Smarter Network Storage - business model …...Smarter Network Storage - business model consultation| 5 technology-type to SNS. This Consultation will complement existing studies on

ukpowernetworks.co.uk

Smarter Network Storage - business model consultation

2 | Smarter Network Storage - business model consultation

4.InvestmentModelTemplates 65

4.1. TemplateIntroduction 66

4.2. ModelSpecification 66

4.2.1. Structure 66

4.2.2. Inputs 68

4.2.3. Calculations 70

4.2.4. Outputs 71

4.2.5. Assumptions 74

4.2.6. Limitations 75

4.3. BusinessModelInsights 75

4.3.1. DNOContractedModel 75

4.3.2. ContractedServices 75

4.4. ConsultationQuestions 76

Appendix1TheValueofStorage 77

LocalSecurityofSupply(SOS) 78

FirmFrequencyResponse(FFR) 78

ShortTermOperatingReserve(STOR) 78

EnergyArbitrage 79

EmbeddedBenefits 79

OtherDNOServices 79

CapacityPayments 79

AncillaryServiceExclusivity 80

ExecutiveSummary 4

1.Introduction 11

1.1. WhatdoesUKPowerNetworksdo? 12

1.2. TheSmarterNetworkStorageProject 12

1.3. AimsandScopeoftheConsultation 14

1.4. TheValueofDistributionConnected

ElectricalEnergyStorage 17

1.5. ThePotentialBenefitsofaWider

RolloutofDistributionConnectedEES 22

1.6. HowtoRespond 23

2.BusinessModelsforStorage 24

2.1. BusinessModels 26

2.1.1. DNOMerchant 28

2.1.2. DistributionSystemOperator 30

2.1.3. DNOContracted 34

2.1.4. ContractedServices 37

2.1.5. ChargingIncentives 39

2.2. QualitativeAssessmentSummary 41

2.3. RegulatoryTreatment 44

2.4. AccountingTreatment 46


3.LeadBusinessModels 48

3.1. DNOContracted 49

3.2. ContractedServices 55

3.3. Variants 59

3.4. SecureCapacityWindows

&TollingContractTerms 61

3.5. Comparison&Conclusions 63


Contents

Smarter Network Storage - business model consultation | 3

Figures

Figure1 BusinessModelKeyFeatures 6

Figure2 LeightonBuzzardReinforcementOptions 13

Figure3 StorageProjectSchematicCostComparison 18

Figure4 ConsultationQuestions 25

Figure5 BusinessModelKeyFeatures 26

Figure6 DNOMerchantModel 28

Figure7 DNOContractedModel 34

Figure8 ContractedServicesModel 37

Figure9 QualitativeAssessmentSummary 43

Figure10 RegulationOptions 44

Figure11 DNOContracted:ContractualStructures 50

Figure12 ContractedServices:ContractualStructures 56

Figure13 DNOandThirdPartyOwnershipVariants 60

Figure14 InvestmentModelTemplateSchematic 67

Figure15 ModelInputCategories 69

Figure16 GenerationCapacity

andSystemInefficiencies 70

Figure17 AssetUtilisationSplitSummary 71

Figure18 TraditionalReinforcement

andStorageOptionCostComparison(£) 72

Figure19 ProjectProfitabilityChecks 72

Figure20 RangeofFeasibleTollingCharges

(Cumulativeoverlife-timeofproject) 73

Tables

Table1 BusinessModelAdvantages

andDisadvantagesSummary 8

Table2 SNSProjectAncillaryBenefits 20

Table3 QualitativeAssessmentCriteria 27

Table4 DNOMerchantModel

QualitativeAssessment 29

Table5 DSOModelQualitativeAssessment 33

Table6 DNOContractedQualitativeAssessment 36

Table7 ContractedServicesModel


Table8 ChargingIncentivesModel


Table9 QualitativeScoringMatrix 42

Table10 ExampleDNOContractTolling

AgreementHeadlineTerms 52

Table11 ExampleContractedServices

TollingAgreementHeadlineTerms 58

Table12 AlternativestoTollingAgreement

HeadlineTerms 62


Executive Summary


technology-type toSNS. This Consultationwill complement

existing studies on the macro-economic benefit of storage,

suchasthosefromImperialCollegeLondon1andtheEnergy

Research Partnership2, and will inform how these might be

achievedinpractice.

ThekeyfeaturesofthetechnologiesthatthisConsultationis

applicabletoare:

• Technology:Electricalenergystorageincludingforexample

lead-acid, lithium-ion, flow and sodium sulphur batteries.

Mechanicalstoragecouldalsobeapplicable.

• Size:Generationcapacityofbetween1MWto10MW,with

storage capacity equivalent to between one and several

hoursofgeneration.

• Location:Distribution-networkconnectedatEHV(ExtraHigh

Voltage)orHV(HighVoltage).Wehaveconsideredbusiness

cases for projects that are located at similar voltages in

distributionnetworks.

• Primary need: The primary need is for mitigation of a

distributionnetworkconstraintthatwouldotherwiserequire

reinforcement. This could be related to network security,

statutory voltage limits or fault-levels. We consider cases

wherethisisamajordriverofthebusinesscase.

• Technology and proposition maturity: Future storage

developments,fromthefirstpost-LCNFundprojectstoafuture

worldinwhichEESformspartofBusinessAsUsualforDNOs.

The consultation closes on30th September 2013. Interested

partiesshouldseeSection2.5forasummaryofquestionsand

instructionsonhowtorespond.

InDecember2012,UKPowerNetworkswasawarded£13.2

millionof funding fromOfgem’s LowCarbonNetworks Fund

(LCNFund)fortheSmarterNetworkStorage(SNS)Tier2project.

TheSmarterNetworkStorageprojectwilldevelopandutilise

6MW/10MWh of advanced electrical storage technology to

support the local distribution network while exploring the

potential commercialopportunitiesassociatedwithproviding

balancing energy through the wholesale electricity markets,

andancillaryservicestotheSystemOperator,andtheextentto

whichmultipleservicescanbeprovidedsimultaneously.

1.ConsultationScope

Aspartofthisproject,UKPowerNetworksislaunchingaConsultation

ontheviablebusinessmodelsfordistributionscaleElectricalEnergy

Storage(EES).Weaimtodemonstrateandtestour thinkingon

the possible business models for energy storage, including the

advantagesanddisadvantagesoftheoptionspresented.

The purpose of this Consultation is to gather views from

all interested parties on the possible business models for

distributionconnectedEESasproposedbyUKPowerNetworks.

TheConsultationalsoseeksassurancefromstakeholdersand

subject matter experts that all significant cost and revenue

streamshavebeenaccountedforinthebusinessmodels.We

areseekingviewsonthemicro-economicbusinessmodel:

the business model for a particular installation of energy

storage, from an investor or ‘controlling entity’ perspective.

We are considering the suitability of the business models

for projects of a similar distribution-scale and of similar

1 Strategic Assessment of the Role and Value of Energy Storage Systems in the UK Low Carbon Energy Future, Energy Futures Lab, Imperial College, EDF UK R&D, 20122 The future role of energy storage in the UK, Energy Research Partnership 2011


2.BusinessModels

We considered five broad categories of business model in

Figure1.

Figure 1 Business Model Key Features

Fullmerchantrisk,exposedtopowerpriceandbalancingservices

DNOexposedtoincentivescheme

DNOexposedtoconstructionandoperationalrisks

LowcommercialriskforDNO

Noguaranteeofassetbeingbuild

• DNObuilds,ownsandoperatestheasset.Fulloperationalcontrol.

• DNOmonetisesadditionalvaluestreamsdirectlyonashorttermbasis(e.g.trading).

• Possiblebarriers:Costsofaccessingthemarket,DNOskillsandcapabilities,regulationandshareholderexpectationsofrisk.

• DNObuilds,ownsandoperatestheasset.DNOhasfulloperationalcontrol.

• DNOhasDSOrole;coordinatingportfoliosofflexibilityforbothdistributionandwidersystembenefitthroughacentralizedcontrolmechanism.

• DNOcommercialriskisdependantondesignofincentivescheme.


• Priortoconstruction,longtermcontracts(e.g10years)forthecommercialcontroloftheassetoutsideofspecifiedwindowsareagreed.

• Dependantonthefeasibilityoflongtermcontracts.

• DNOoffersalongtermcontract(e.g.10years)forservicesataspecificlocationwithcommercialcontrolincertainperiods.

• Thirdpartyresponsibleforbuilding,owningandoperatingtheassetandmonetisingadditionalrevenuestreams.

• DNOsetsDUoStocreatesignalsforpeakshavingthatreflectthevalueofreinforcement.

• Barriers:nooperationalcontrolforDNO,thereforenoguaranteeonsecurity.

Model Keypoints Comments

DNOMerchant

DSO

DNOContracted

ContractedServices

Charging Incentives

Incr

easi

ngc

omm

erci

alri

skfo

rDN

O

Incr

easi

ngc

omm

erci

alri

skfo

rThi

rdP

arty


In the DNO Merchant business model, the DNO takes full

ownershipandoperationof theassetand is responsible for

monetising thevalue fromthewholesaleelectricitymarkets

and ancillary services provided to the System Operator. The

DNOdoesthisdirectlyintherelevantmarketsandthereisno

thirdpartywithadirectrelationshiptotheasset.

Under a DSO model the DNO would own, operate and

maintaintheassetaspartofawiderroleofactivelymanaging

itsnetworkunderaregulatoryincentivescheme,akintothe

role that National Grid plays at the transmission level. The

DNOwouldalso lead thedevelopmentand constructionof

the asset, finance its construction and operation and then

holditsfullcommercialcontrol.TheDNOwouldaccrueallthe

projectcostsandbenefitsandtherewouldbenodirectthird

partyinvolvement.

The DNO Contracted model differs from the DNO Merchant

model due to the involvement of a third party to manage

thecapacityoftheassetwhenit isnotrequiredforsecurity

purposes.TheDNOwouldstillfinance,maintainandoperate

the asset, but would dispatch for ancillary services at the

instructionofathirdparty.

UndertheContractedServicesmodel,theDNOrunsatender

forthirdpartiestobuildandoperatestorageataspecificsite.

The DNO makes a fixed annual payment in return for the

distributionnetworkservicesprovidedbythethirdparty.The

thirdpartymanagesthecapacityoftheassetwhenit isnot

requiredforsecuritypurposes.

TheChargingIncentivesmodelisoneunderwhichtheDNO

ensuresthattheDUoSchargingcreatestherightincentivesin

thelocationrequiringreinforcement.Thirdpartiesmayormay

notrespondtotheincentivesbybuildingstorage.

Asummaryofthepotentialadvantagesanddisadvantagesof

eachmodelisshowninTable1.


Table 1 Business Model Advantages and Disadvantages Summary

Businessmodel

DNOMerchant

DistributionSystem

Operator(DSO)

DNOContracted

ContractedServices

ChargingIncentives

Advantages

• DNOhasfulloperationalcontrol.

• Maybelowercostoffinancingiffinancedasa

regulatedasset(dependingonrisksharingbetween

DNO&Customers).


• SpecificincentivesonDNOtomanagecostsof

balancingthegrid.



DNO&consumers).



DNO&Customers).

• CommercialriskforDNOsignificantlydecreased.

• Thirdpartymaybebetterplacedtomanage

commercialvaluestreams.

• Thirdpartymaybeabletoaggregateacrossmultiple

assetswhichincreasesscalabilityandoverallsystem

efficiency.

• CommercialriskforDNOsignificantlydecreased.

• Thirdpartymaybebetterplacedtomanage

commercialvaluestreams.

• Thirdpartymaybeabletoaggregateacrossmultiple

assetswhichincreasesscalabilityandoverallsystem

efficiency.

• DNO(andCustomers)takesnocommercialrisk.

• Incentivesbasedapproachmaybeeconomically

efficient.

Disadvantages

• DNOrequiresnewskillsand

capabilitiestotradeinthewholesale

energymarketandparticipate

inprocurementmechanismsfor

ancillaryservices.

• MaynotbeconsistentwithDNO

shareholderexpectationsofrisk.

• Regulatoryregimenotyetinplace

• CommercialriskremainswithDNO

andCustomers.

• Complextollingcontractrequired(i.e.

aservicescontractbetweentheDNO

andathirdparty).

• Thirdpartymayheavilydiscountlong

termvalueofadditionalrevenues.

• DNOdoesnothavedirectoperational

control.

• Complextollingcontractrequired.

• Thirdpartymayheavilydiscountlong

termvalueofadditionalrevenues.

• Noguaranteeofstoragebeingbuilt.

• NoDNOcontrolonassetbeing.

availablefornetworksecuritywhen

required.

• Thirdpartyexposedtoannual

changestoincentives.


Impactonassetvalue,costsandrisks

Bothmodelsdependonawell-structuredtollingcontractthat

givesasmuchavailabilitytothethirdpartyaspossiblewithout

compromisingsecurity.Bothmodelsplacetheoptimisationof

thevaluestreamswithathirdpartywhichislikelytohavea

moredevelopedsetofskillsandcapabilitiestogeneratevalue

fromthestoragewithoutimposinghighcostsoftrading,and

dependona thirdparty’swillingness to take long term risk

on the additional value streams. However, under the DNO

ContractedmodelthereismoreflexibilityfortheDNOtoshare

someofthisriskiftheDNOcantakesomemerchantexposure.

A relative advantage of the DNO Contracted model may be

a lowercostofcapital.However thismaynotpersistas the

deploymentofstorageincreasesandtheriskprofileoftheutility

changes.TheDNOContractedoptioncreatesthepossibilityof

sharingtherisk(andadditionalbenefit)withCustomers(ifthis

wasconsidereddesirableforCustomers),whereasthis isnot

easilypossiblewiththeContractedServicesmodel.

Impactonwiderbenefits

Under both models, the tolling contract will need to clearly

specifythetermsonwhichcapacityismadeavailable.Thereis

ariskthatthefullbenefitsfortheGBsystemarenotcaptured

duetoalackofflexibilityintheseterms.

TheDNOcontractedmodelallowsforcompetitionbetween

thirdpartiesintheprovisionoftradingandaggregation.The

third party has the ability to transfer their knowledge and

expertise to develop projects in other DNO licence areas.

UndertheContractedServicesmodel,thethirdpartycanalso

transfertheirexperienceofdesigning,buildingandoperating

other storage assets. Additionally, third parties that have a

portfolio of assets in planning, construction or operation

mayhavealreadyestablishedfairervaluearrangementsand

Afterreviewingthesemodels,weselectedDNOContractedand

ContractedServicesasthetwoleadbusinessmodelsforfurther

consideration.TheDNOMerchantmodelwasexcludedmainly

because of the requirement for the DNO to build a trading

capability and take wholesale market risk. The DSO model,

whileattractiveinprinciplewasexcludedatthistimebecause

the underlying regulation that would define this model has

yettobedevelopedandassuchcannotbecriticallyappraised.

However,werecognisethatanumberofotherLCNFprojects

maybenefitfromthiskindofapproach.Assuch,wedonot

ruleoutthepossibilitythatthisbusinessmodelbecomesmore

relevant in the future. The Charging Incentives model was

excludedbecauseitprovidesnoguaranteeofthestoragebeing

builtor,oncebuilt,beingavailabletoprovidenetworksecurity.

3.LeadBusinessModels

Under the DNO Contracted model, the DNO would own the

asset,whereasunderContractedServicesitwouldbeowned

andoperatedbyathirdparty.

Theleadbusinessmodelsshareacommonconceptofalong

termcontractbetweentheDNOandathirdpartytosharethe

risksandrewardsfromcommercialopportunitiesthatcanbe

capturedbeyondtheasset’sprimaryroleofprovidingnetwork

security.Thetermsforthistollingcontractwouldneedtogive

thethirdpartythegreatestfreedompossibletooptimisethe

value of additional value streams whilst ensuring that the

DNO’ssecurityrequirementismet.

Impactonnetworksecurity

TheDNOContractedmodelgivestheDNOdirectcontroloverthe

operationofthestorage.ThesecurityprovidedbytheContracted

Services model is dependent on the contractual obligations

placedonthethirdparty,andhowitfulfilstheseobligations.

Whilst not as direct as operational control, this model could

providesufficientsecurityifthetermsarewellstructured.


barriers for these models are shared: the complexity of the

tollingcontractandthewillingnessofathirdpartytotakelong

termriskontheadditionalvaluestreams.

4.InvestmentModelTemplates

Wehavealsodevelopeda template investmentmodel for

each of the two lead business models. These templates

demonstrate how the business case for distribution-

connected EES might be assessed. The investment model

template illustrates the possible financing costs, operating

models and revenue streams in each case, and enables

comparisons across the business models for the same

configurationofstorage.

TheConsultationresponseswillallowtheprojecttoupdate

theinvestmentmodeltemplates,andtounderstandthelevel

ofacceptancefrompotentialbuyersofservicesandthelevel

ofappetite frompotentialstorageowners for thedifferent

businessmodels.Theinvestmentmodeltemplatesareakey

tool intheprojectandoneoftheaimsoftheproject isto

populateanddisseminate thesewith real costandbenefit

figuresbasedonexperiencefromtheoperationaltrials.

contractswiththeprovidersofthesedesign,procurementand

buildingservices.

Under both models, the DNO is able to set terms of the

technologyconsidered, toensure thata lowcarbonsolution

isprocured.

Futureproofingofbusinessmodel

Bothmodels allow for aggregationof thedispatchofmultiple

assetsbythethirdparty.TheContractedServicesmodelallowsone

thirdpartytooperatestorageacrossmultipleDNOlicenceareas.

FromaregulatoryperspectivetheDNOContractedmodelcould

face barriers as theDNOapproaches its present deminimis

non-distribution activity limit, which is 2.5 per cent of the

sumofthelicensee’ssharecapital,itssharepremium,andits

consolidatedreserves.

Based on our review of the lead models, both the DNO

Contracted and Contracted Services appear to be feasible

business models for distribution-connected storage. The key


Introduction1


1.1.WhatdoesUKPowerNetworksdo?

UKPowerNetworksowns,operatesandmanagesthreeofthe

fourteenregionalelectricitydistributionnetworksintheUK.Our

licenseddistributionnetworksareintheEastofEngland(Eastern

PowerNetworksplc),London(LondonPowerNetworksplc)and

theSouthEast(SouthEasternPowerNetworksplc).UKPower

NetworksisoneofthelargestDistributionNetworkOperators

(DNOs)intheUK,coveringanareaofapproximately30,000km,

extending from the Wash in the east, through London, to

LittlehamptonontheSussexcoast.Approximatelyeightmillion

connectedcustomersdependonusfortheirpower.

1.2.TheSmarterNetworkStorageProject

TheSNSprojectwilldevelopa6MW/10MWhbattery facility

atLeightonBuzzardprimarysubstation.Thissystemwillhave

thecapabilitytogenerate6MWforapproximately1.5hoursfor

eachfullchargeanddischargecycle,andtochangeoutputlevel

withinafewseconds.Thisisintendedtodelay,orpotentially

avoid the requirement for traditional reinforcement to the

distributionnetworkasdescribedbelow.

LeightonBuzzard substation comprises two33/11kV38MVA

transformers fed by two 33kV overhead Lines (OHLs), each

withawinterratingof35.6MVA.Networksecurityofsupply

standard Engineering Recommendation P2/6 sets out the

minimumdemandthatneedstobemetfollowingthelossofa

circuitdependingonthegroupdemand.Thisrequiresalevelof

redundancysuchthatifonelineweretofail,acertainproportion

ofgroupdemandcanstillbemet.AtLeightonBuzzard,thesite

‘firmcapacity’ –definedas themaximumcapacityavailable

during an N-1 event – is currently restricted by the thermal

ratingofthe33kVOHLandistherefore35.6MVA.

Peak demand at LeightonBuzzard has been above this firm

capacitylimitbetween9and37daysineachofthelastfiveyears

(typicallyduringperiodsofverycoldweather).Theadditional

capacityrequiredhasbeenprovidedbytransfercapacityfrom

InDecember2012,UKPowerNetworkswasawarded£13.2

millionof funding fromOfgem’s LowCarbonNetworks Fund

(LCNFund)fortheSmarterNetworkStorage(SNS)Tier2project.

The Smarter Network Storage project will develop and utilise

6MW/10MWh of advanced electrical storage technology to

support the local distribution network while exploring the

potential commercial arrangements that will support overall

systembalancingandstability,thewholesaleelectricitymarkets

and the viability in providingmultiple services simultaneously.

Theprojectisdifferentiatedfromotherstoragedemonstrationsin

thatitsaimistoundertakearangeofcommercialandtechnical

innovationstoexploreandimprovetheeconomicsofelectrical

energy storage when leveraged for full-system benefit, and

howtheseadditionalvaluestreamsmaysupportviablebusiness

modelsforstorageinthefuture.Theobjectiveistosupportthe

developmentofthestorageindustrytobenefitnetworkoperators

andcustomers,allowingstoragetobenefitanupgradedelectricity

systeminamoresustainableandcostefficientway.

As part of this project, UK Power Networks is launching a

consultationontheviablebusinessmodelsfordistributionscale

ElectricalEnergyStorage(EES).Weaimtodemonstrateandtest

ourthinkingonthepossiblebusinessmodelsforenergystorage,

including the advantages/disadvantages of the options. The

Consultationalsoseeksassurancefromstakeholdersandsubject

matter experts that all significant cost and revenue streams

havebeenaccountedforinthebusinessmodels.

Wearealsoseekingfeedbackonatemplateinvestmentmodel

fortwospecificbusinessmodelsthatwebelievearethemost

likelytobesuitablefordistribution-connectedstorage.These

models demonstratehow thebusiness case for distribution-

connectedenergystoragemightbeassessedforeachbusiness

model.Thetemplateinvestmentmodelillustratesthepossible

financingcosts,operatingmodelsandrevenuemodelsforeach

businessmodel.


to LeightonBuzzardprimary substation,anda third38MVA

transformer at Leighton Buzzard. This reinforcement would

provide an additional 36 MVA of firm capacity at Leighton

Buzzard,which is significantlyabovepredicted requirements

for the medium-long term. The traditional reinforcement

option, and alternative approach using storage as described

below,isshowninFigure2.

neighbouring sections of the network. Peak demand at this

locationisforecasttocontinuetogrow,andtransfercapacityis

limitedat2MVA,meaningthatlimitsmaybebreachedinfuture

years.Thisisatriggerforreinforcementofthenetwork.

The conventional reinforcement option for Leighton Buzzard

hasbeenevaluatedasathird33kVcircuitfromSundonGrid

Figure 2 Leighton Buzzard Reinforcement Options

GridSupply

ToCustomers

OverheadLines

ConventionalIntervention:

3rdLine

EnergyStorage

Primary Substation 33kV to 11kV

GridSubstation132kVto33kV

PowerFlow


Theprimarypurposeofthestorageassetistoprovidenetwork

security,andadditionalbenefitsmustbecompatiblewiththis

requirement.

DuringthecourseoftheSNSproject,UKPowerNetworkswill

demonstrate:

1.Deployment and multi-purpose application of large-scale

distribution-connectedEES

2.ImplementationofaSmartOptimisation&Controlsystemin

ordertomanageandoptimisethestorageflexibility

3.Innovativecommercialarrangementstosupporttheshared

useofenergystorage inprovidingwidersystembenefits,

includingstandbyreserveandmanagingfrequency

4.Assessmentandvalidationofthefullvaluethatstoragecan

provide to DNOs and the wider system to support future

businessmodelsforstorage

This Consultation will primarily support the third of these

objectives,asdescribedinthefollowingsection.

1.3.AimsandScopeoftheConsultation

The purpose of this Consultation is to gather views from

interested parties on the possible business models for

distributionconnectedEESasproposedbyUKPowerNetworks.

Therearemultipledifferentownershipandoperatingmodels

thatcouldevolvearoundEES.TheSNSprojectaimstocapture

learning,demonstrateanalysisandprovidethought-leadership

thatwillsupportthedevelopmentofviablebusinessmodels

forfutureelectricitystorageprojectsatthedistributionlevel.

Note that we are not consulting on the business model for

the SNS project itself. The business model and commercial

arrangements for the Leighton Buzzard EES facility reflect a

variationofthe‘DNOContracted’businessmodel,inwhichthe

DNOownsandoperatesthestorage,asdescribedlaterinthis

report. This approach has been designed to ensure learning

Analternativetobuildinganewcircuitandtransformer is to

considerinnovativesolutionsthatwillgiveUKPowerNetworks

theabilitytoreduce(net)peakofftakeatLeightonBuzzardto

maintain demand below the firm capacity rating. This could

taketheformofembeddedgeneration,DemandSideResponse

(DSR),orElectricalEnergyStorage(EES),allofwhichcouldreduce

therequiredofftakefromthenetworkatLeightonBuzzard.The

reductionofpeakdemandcoulddelaytheneedfortraditional

reinforcement foranumberofyears. Thismaybecombined

withincrementalupgradestothenetwork(e.g.reconductoring

of theexistingoverhead lines) to furtherdelayorpotentially

avoidtheneedfortraditionalreinforcement.

Distributionconnectedstorageisstillarelativelynewconceptthat

attractsasetof“FirstofaKind”coststhatwouldnotbeaccrued

tootherestablishedtechnologiesandsolutions.Thesecostsare

expectedtoreduceinfuturewithgreaterdeploymentofstorage.

Ontheotherhand,distributedstoragecanaccessarangeof

additionalbenefitsnotaccessibletotraditionalreinforcement.

Thevalueoftheseadditionalbenefitsrelativetotheadditional

cost of storage governs whether storage is cost effective

comparedtothealternativetraditionalreinforcementoptions.

The aim of the SNS project is to explore how EES can fully

maximise these ancillary benefits. These benefits are above

andbeyondtheresolutionofnetworkconstraintsatLeighton

Buzzard and can both improve the project economics and

advance theuseof storageasa cost-effectivealternative to

networkreinforcement.Theseadditionalbenefitscouldinclude:

• Provisionofpowerqualityservicesonthedistributionnetwork

• Provisionof balancingenergy to themarket andancillary

services to the System Operator to reduce the costs of

managing the GB transmission system with increasing

proportionsofintermittentrenewables

• Reductioninrequirementforpeakgenerationcapacity


thesearedescribedinBox1.UKPowerNetworksanalysis

has indicated thata reasonableestimate for thepotential

nationalcapacityofsitessimilartotheSNSprojectisofthe

orderof2GWby20403.Thereareanumberofkeyfeatures

whichwillbecommontotheseopportunities:

• Technology: Electrical energy storage including for

example lead-acid, lithium-ion, flow and sodium

sulphurbatteries.Mechanical storage, if scalableandat

similar Technology Readiness Levels (TRLs) could also

be applicable, such as some flywheel systems. Storage

requiring specialist geological requirements, such as

undergroundcompressed-air,areoutofscopeduetothe

differingCAPEXandOPEXrequirements.

• Size:Generationcapacityofbetween1MWto10MW,with

storagecapacityequivalenttobetweenoneandseveral

hoursofgeneration.

• Location: Distribution-network connected at EHV or HV.

Wehaveconsideredbusinesscasesforprojectsthatare

locatedatsimilarvoltagesindistributionnetworks.

• Primary need: the primary need is for mitigation of a

distribution network constraint that would otherwise

requirenetwork reinforcement. This couldbe related to

network security, statutory voltage limits or fault-levels.

We consider cases where this is a major driver of the

businesscase.

• Technologyandpropositionmaturity:weareconsulting

on the business models for future storage developments,

from the first post-LCN Fund projects to a future world in

whichEESformspartofbusinessasusualforDNOs.

canbegained inrelationtoawiderangeof futurepossible

businessmodelsthatmayinvolvethestoragebeingentirely

managed,operatedorevenownedbydifferenttypesofthird-

partyorganisation.Furthermore,thisapproachwillensurefull

transparencyof the valuegenerated fromeach service area

andmakesmosteffectiveuseofexistingbusinesscapability

andoperationalexperienceacrossourprojectpartners.Thiswill

notnecessarilybe typicalof futurestorageand thescope is

rathertoconsiderthefuturebusinessmodelsforstoragethat

aretobedevelopedwithoutLCNFundsupport.

Businessmodels

Thescopeofthebusinessmodelsconsultationisdefinedas:

• Micro-economicbusinessmodel:thebusinessmodelfora

particularinstallationofenergystorage,fromaninvestoror

‘controllingentity’perspective.Thiswillincludealllifecycle

andinvestmentcostsindetailandallowinstallationstobe

compared in terms of economic viability, and sensitivity

analysis.Outof scopeare ‘non-bankable’ societalbenefits

accruingto‘UKPlc’,suchascarbonemissionssavings.These

are relevant to the overall benefits case for storage and

mayinfluencethedesignofpolicyorregulationbutdonot

directlycontributetorevenues.

• Suitable for energy storage projects of a similar

distribution-scale and of similar technology-type to

SNS: in order to restrict the scope of discussions we will

evaluate business models for projects similar to the SNS

project in terms of scale and storage type. We note that

thereareawiderangeofpotentialtechnologies,sizesand

usesofstorageonthedistributionnetworks,someofwhich

arebeingexploredunderotherLCNFundprojects.Someof

3 LCNF Full Submission Pro-Forma. Appendix G, Section 2 http://www.ofgem.gov.uk/Networks/ElecDist/lcnf/stlcnp/year3/smarter-network-storage/Documents1/SNS%20RE-Submission%20+%20Appendices%20Redacted.pdf


Therangeofscalesandusesofstorageondistributionnetworksrangesfromdomesticorstreetlevelbatterytechnologies,upto

megawattscalestoragethatmaybeusedtoabsorbexcessembeddedrenewablegeneration(exportconstraintmanagement)

ormeetpeakdemandinimportconstrainedareas.Storagemayalsobeinstalledforpowerqualitymanagementpurposes.

Box 1 Storage on distribution networks

ProjectName

Demonstratingthebenefitsofshort-termdischargeenergystorageonan11kVdistributionnetwork

ShetlandandtheNINESproject

CHALVEY

DNO

UKPN

SSEPD

SSEPD

Location

Hemsby,Norfolk

LerwickPowerStation,Shetland

Chalvey,Slough,Berkshire

Description

UKPowerNetworkshasinstalledadynamicenergystoragesystem(ESS)atHemsbyinNorfolk,incollaborationwithABB.ThesystemisbasedonABB’sSVCLightproduct,combinedwithaLithium-ionbatterystoragedeviceandislocatedonan11kVdistributionnetworkwithsomepenetrationofwindpower(http://www.ukpowernetworks.co.uk/internet/en/innovation/documents/Hemsby_Progress_report%20_Oct_2012_FINAL.pdf)

ThisProjectinvolvesinstallinga1MWeconnectedbatteryattheLerwickPowerStationonShetland.ThisProjectwillprovidelearningregardingtheoperationofthebatteryanditsintegrationwithlocalDemandSideResponsetoremovestationpeaksprovidingadditionalDemandcapacity(inasimilarwaytomanaginganetworkloadconstraint).(http://www.ofgem.gov.uk/Networks/ElecDist/lcnf/ftp/sse/Pages/index.aspx)

Thisprojectseekstounderstandthepotentialbenefits,practicalitiesandcostsofinstallingelectricalenergystorage(EES)connectedviafourquadrantpowerconversionsystems(PCS)ontheLVnetwork.Themainobjectiveistoinformandde-riskthelargerscaledeploymentofstreetbatteriesasdetailedintheNTVVTier2project.(http://www.ofgem.gov.uk/Networks/ElecDist/lcnf/ftp/sse/Pages/index.aspx)

These threeexamples are shown to illustrate the rangeof possible storageapplications and that thebusinessmodels

presentedinthisconsultationwhilegenericasfarasispossible,maynotbeapplicabletootherstorageprojects.


We are requesting responses from a broad range of interested

parties with an interest in distribution-scale electricity storage.

This may include, but is not limited to:

• Supply chain (e.g. manufacturers)

• Generation (renewable and conventional generators)

• Transmission (Transmission Owners and the GB System

Operator)

• Distribution (Distribution Network Operators)

• Suppliers and aggregators

• Customers (e.g. Industrial, Community)

• Regulation (Ofgem)

• Government

We will review the responses and publish a summary along with

our findings. The output of the Consultation will be used to inform

the on-going SNS project. In particular we expect this to guide

the further assessment of regulatory and commercial barriers and

the assessment of the business case for distribution connected

storage. Feedback on the template investment model will be

incorporated into a version which will be populated throughout

the project lifecycle as learning relating to the operational costs

and value streams is validated during the trial phases.

1.4. The Value of Distribution Connected Electrical Energy

Storage

An understanding of the business models for storage depends

on an understanding of the value streams available to EES. For

We have reviewed a range of potential business models and

made a qualitative assessment of these in order to select two

models for further consideration. We have considered the

details of how these two models would operate, and what the

advantages, disadvantages and risks are of these models.

Investment model template

We have also developed a template investment model for each

of the two lead business models. These templates demonstrate

how the business case for distribution-connected EES might

be assessed. The template investment model illustrates the

possible financing costs, operating models and revenue models

in each case, and enables comparisons across the business

models for the same configuration of storage.

The investment templates are provided for illustration of the

business models only. Although we have populated them with

representative values as an aid to users to understand the

workings of each model, we do not present these as a business

case assessment for any particular EES project.

Timelines

Responses should be sent to:

[email protected]

by 30th September 2013. The Consultation process is shown

schematically in the below diagram.

Incorporating learning into SNS project

Review responses and publish summary

Stakeholders respond to consultation

Issue consultationDevelop business models and investment model templates


traditionalreinforcementoptions.Thebusinesscaseisdriven

primarilybythisconstraintwhichcouldberelatedtonetwork

security, or, statutory voltage limits for example, whilst

additional value streams increase the competitiveness of

storagewhenconsideredalongsidetraditionalreinforcement.

This is shown schematically in Figure 3 where the current

storage project cost is adjusted for future technology and

system cost reductions, along with the expected ancillary

revenuestreams.ThisreducedcostisshownastheLongRun

Cost.Thisisthenetcostwhichcanbecomparedtothecost

oftraditionalreinforcementtoappraisethefutureviabilityof

storageprojects.

this analysis we focus on the monetisation of benefits to a

particularproject, ratherthanthewidereconomicandsocial

benefits provided to theGB systemas awhole (it isworth

notinghoweverintheparticularcaseoftheDSOtheincentive

islikelytohavebeensetbasedon,anddesignedtoachieve

thesewidereconomicandsocialbenefits).

Forthescopeofdistribution-scaleEESunderconsideration,the

“NeedsCase”forstorageisinitiatedbyadistributionnetwork

requirement.Whereflexibilityindemandand/orgeneration

would help to mitigate network constraints and potentially

deferinvestment,thenstorageshouldbeconsideredalongside

Figure 3 Storage Project Schematic Cost Comparison

Conventional Once proven successful

£m

ConventionalReinforcement

SNSProjectCost

1stofakindcosts

TechCostReduction

FutureIncomeStreams

SystemCostSavings

LongRunCost


InthecaseoftheSNSproject,thestorageisrequiredtocontribute

tonetworksecurity.TheimplicationoftheNeedsCaseisthat

storagemustbeavailableforsecuritypurposeswhenrequired,

takingpriorityoverallotheruses.Thisisbecausethesecurity

providedbythestorageassetcannotbeprovidedbyanother

storageassetorgeneratorelsewhereonthenetwork–itmust

beconnectedtothespecificsubstationthatrequiressupport.

There are alternative applications for distribution-connected

EES other than to meet security obligations, for example

the management of the output of embedded intermittent

generation.Insuchcasestheassetcouldbeoperatedwithout

restrictioninapurelyprofitoptimisingmanner.

Anyadditionalvaluestreamsmustbecompatiblewiththe

security requirement. Given the nature of timing of the

demand,itishighlyprobablethatthestoragewillberequired

toprovidesecurityforonlyasmallwindowwhenparticularly

highpeakdemandmayoccur.Currently this is likely tobe

timedcloseto theGBsystempeakdemand,althoughthis

couldvaryinfutureforareaswithsignificantpenetrationof

embeddedgenerationbehindthesameconstraint.For the

LeightonBuzzardexample,theassetislikelytoberequired

for securityonly inwintermonthsandonlyatpeak times

(e.g. 4-7pm).Wedefine this periodof timeas the Secure

Capacity Window. Outside of this window the storage can

provideotherservices,subjecttobeinginastatetoprovide

security to the system at the start of the Secure Capacity

Window.Itmayalsobepossibleforsomecompatibleservices

tobeprovidedduringtheSecureCapacityWindowitself.This

isanareathattheSNSprojectwillinvestigateastohowthis

mightworkinpractice.

Table2 showsa listof the future incomestreamsshown in

theStorageProjectCostBreakupinFigure3above.Onlysome

of these income streams have been considered in the SNS

projectandinvestmenttemplates,alistofwhichisgivenina

followingsection(furtherdetaileddescriptionsanddiscussions

areincludedinAppendix1).


Table 2 SNS Project Ancillary Benefits

Valuestream

Local SecurityProvisionofcapacityattimesofpeaklocaldemand

‘Embedded Benefits’ e.g. avoided demand TNUoS, avoided BSUoS, avoided losses

Reactive PowerProvisionofreactivepowertoreducelosses

Voltage SupportProvisiontolocalnetwork

Firm Frequency Response (FFR)ManagingtheGBsystemfrequency

Short Term Operating Reserve (STOR)ReservefortheGBsystem

Energy ArbitrageTradingofwholesalepower

Capacity Payments

Finalcustomer

DNO

Supplier

DNO.Mayalsobeopportunitiesontransmissionnetwork

DNO.Mayalsobeopportunitiesontransmissionnetwork

TSO

TSO

Wholesalemarket/Supplier

GeneratorsorEnergyproviders

Driversofvalue

Avoided/delayedreinforcementcost

DemandTNUoS,BSUoS,transmissionlosses

Reducedlossesandpossibleavoidedreinforcementduetoreactivepowerallowingmoreefficientuseofnetwork

Possiblevalueinspecificlocationse.g.withlocalrenewablegeneration

GrowthinFFRrequirementduetosizeoflargestlossontransmissionsystem

GrowthinSTORrequirementduetointermittency

Powerpriceshape&volatility

Designofcapacitymechanism

Valuerisks

Technicalperformanceofstorage(availability)

Onlyrelevantwhenstorageisexporting

Opportunitiesmaybelimited/specific

Opportunitiesmaybelimited/specific

CompetingsourcesofFFR

CompetingsourcesofSTOR

Lackofmarketliquidity

Penaltiesfornon-delivery

Likelysuitability

Primaryneed

Secondaryvalue

stream

Opportunityspecific

Opportunityspecific

Secondaryvalue

stream

Secondaryvalue

stream

Secondary

valuestream

Potentialfuture

benefit


Table 2 SNS Project Ancillary Benefits

Valuestream

Managing Intermittent Renewable Generation (Balancing Market Services)Absorbingexcessrenewablegeneration

Fast Reserve

Transmission Constraints

Inertia ServicePossibleTSOservice

Responsive Flexibility ServicePossiblefutureTSOservice

Distribution LossesGenerationatpeakreduceslossesbymorethatincreasewhencharging

Finalcustomer

RenewablegeneratororDNO

TSO

TSO

TSO

TSO

DNO

Driversofvalue

GrowthinIntermittentgeneration

Sizeofmarketforfastreserve

Transmissionconstraintinlocationofstorage

Maybeincreasinglyrequiredasamountofasynchronousgenerationincreases

Possiblefutureproducttospecificallyreflectbenefitsprovidedbyresponsivestorageorothersourcesoflow-carbonflexibility

Dependsonincentivescheme.BenefitmayaccruetoDNObutnottostoragedirectly

Valuerisks

Requirementisunpredictable

Competingsources

Unlikelytobeapplicableunlessaggregated

Competingsources

Productcurrentlyundefined

Likelysuitability

Valuableifsignificant

localembedded

generation

Limitedtocapacity

>50MW

Notconsidered

Potentialfuture

benefit

Potentialfuture

benefit

Anancillarybenefit

ofoperationofthe

asset–unlikelyto

drivedispatch.Not

considered


networks once EES is proven as an economic alternative

solutiontoreinforcement.Theanalysisconservativelysuggests

around2GWofdistribution-connectedstoragecapacity could

beintegratedintothesystemacrossGBby20404.

UKPowerNetworkshasalsocalculatedthepresentvalueof

net benefits of this additional flexible capacity at anational

leveltobearound£0.7bn,resultingfromsavingsindistribution

and transmission investment, value from supporting system

balancing, displacement of peaking generation capacity and

reducedcostsofcurtailmentoflow-carbongeneration.These

benefits assume that the storage is leveraged across only

a limited number of applications simultaneously for short

periods;althoughinpracticeitisexpectedthatstoragecapacity

couldbemuchmoreflexible5.

AstheintegrationofEESintotheGBsystemincreasessodo

the possible portfolio and scalability benefit to the project

parties.Forexample,athirdparty(orindeedaDNOwithan

establishedenergytradingcapability)couldactivelymanage

anddispatchassetsoveralargespatialarea,tofirstlyincrease

economiesofscaleandreduceoperatingcosts,andsecondly

toincreasetheirownsystemredundanciesandimprovetheir

system efficiency (where increasing system redundancies

refers to theirability tomitigateunavailabilityof individual

assets by picking up lost generation across their asset

portfolio).Similarly,anEESassetcouldbe incorporated into

a portfolio of intermittent generation assets, allowing the

intermittentgeneratorstodischargetheassettoavoidcostly

energyimbalancecharges.

TheSNSprojectwillfocusonthevalueavailablefromasubset

ofthesevaluestreams,whichhavebeenassessedasbeing

themostsignificantinthecurrentmarketandarecompatible

with the characteristics and likely operation of the storage

assetenvisaged.

• Localsecurity

• ShortTermOperatingReserve(STOR)

• FirmFrequencyResponse

• EnergyArbitrage

ForthepurposesofthisConsultationandinvestmenttemplate

we include two additional sets of value streams as being

significantinthecurrentmarkettocomplementthoseoriginally

identifiedascorevaluestreams;

• EmbeddedBenefits

• CapacityPayments.

We use these value streams as examples throughout this

Consultation document and in the investment templates.

However we recognise that other value streams mentioned

abovemaybecomeequallyormoreimportantinfuture,and

sotheinvestmenttemplateincludestheabilitytoincludeother

user-definedbenefits.

1.5.ThePotentialBenefitsofaWiderRolloutofDistribution

ConnectedEES

UK Power Networks has modelled the potential number of

typicalstoragedeploymentsthatcouldbeappliedtodistribution

4,5 LCNF Full Submission Pro-Forma. Appendix G, Section 2 http://www.ofgem.gov.uk/Networks/ElecDist/lcnf/stlcnp/year3/smarter-network-storage/Documents1/SNS%20RE-Submission%20+%20Appendices%20Redacted.pdf


a.Arethereotherscenariosthatshouldbeincluded?

b.Arethereinherentregulatorycreatedlimitationsthathave

notbeendiscussed?

c.Arethereadditionaladvantagesordisadvantagesforeach

ofthesescenariosthatshouldbeconsidered?

SECTION3LEADBUSINESSMODELS

5.Do you agree with the respective advantages and

disadvantages of the two lead business models as

describedintheSection3?

a.Are there other limitations, barriers or features of these

businessmodels,orEESprojects ingeneral thathavenot

beenconsidered?

b.Doeitheroftheseleadbusinessmodelsdisproportionally

favouronepartyovertheother?

6.Fromyourperspective,whichofthetwoleadbusiness

modelsismostlikelytobefavoured?

SECTION4INVESTMENTMODELTEMPLATE

7.Arethereothertechnologyparameters,costsorrevenue

streams that should have been considered in the

investmentmodeltemplate?

a.Ifsopleasegivedetails

8.Doyouagreewiththeinterrelationsoftheseancillary

servicesandtheirassociatedrevenuestreams?

a.Arethereadditionalcomplexitiesinthedispatchoftheassetto

utilisetheserevenuestreamsthathavenotbeenconsidered?

Are they all mutually exclusive or potentially dispatchable

inunison,arethereadditionalcomplexitiesintheknock-on

effecttobatteryperformancethathavenotbeenconsidered?

9.Doyouagreewiththestatedassumptionsandmodel

limitations?

a.Should any of these stated assumptions or limitations

havebeendealtwithdifferently?

1.6.HowtoRespond

Responsesshouldbesentto

[email protected]

by30thSeptember2013.

Thequestionsthatweinviteinterestedpartiestosubmitwritten

evidenceandanalysisonareasfollows:

SECTION2BUSINESSMODELSFORSTORAGE

1.Do you agree with the range of business models

presentedinSection2.1?

a.Arethesebusinessmodelsandtheirvariantsrepresentative

oftherangeofplausiblebusinessmodels?

b.Doyouagreewiththecharacterisationsofeachofthese

businessmodelsintheirrespectiveSections2.1.1through

toSection2.1.5?

2.Doyouagreewiththechoiceofassessmentcriteriaas

describedinSection2.1?

a.Are these thekeyassessment criteria that thebusiness

modelsshouldbecomparedagainst?

b.Are there any additional advantages, disadvantages or

barrierstoanyoftheindividualbusinessmodelsthathave

notbeenincluded?

3.Doyouagreewith thechoiceof the two leadmodels

andthereasonsforselectingtheseasdescribedinthe

QualitativeAssessmentSummary,Section2.2?

a.Are these options the most likely to be suitable for

distribution-connected storage, or should one or more

oftheotherbusinessmodels(orvariantsofthese)have

beenconsideredovertheDNOContractedandContracted

Servicesmodels?

b.Ifso,whichmodelsshouldhavebeenconsideredandwhy?

4.Doyouagreewiththerangeoffourregulatorytreatment

scenariospresentedinSection2.3?


Business Models for Storage2


inherentroleasthepartyresponsibleforthesecurityneedcase.

Therearealsoawiderangeof thirdparties fromthroughout

theelectricityvaluechainwhocouldplayaroleinthebusiness

model. In describing the business models, we consider the

relationshipbetweentheDNOandasinglethirdpartyonly.In

reality, the thirdpartycould take the formofa jointventure

orconsortium,oraspectsofthebusinessmodelcouldbesplit

betweenmultiplethirdparties,aswillbethecaseforthetrials

withintheSNSprojectitself.

Inthissectionwedescribearangeofpossiblebusinessmodelsfor

distributionconnectedEES.Subsequently,wedescribehowwehave

selectedandfurtherdevelopedtwoleadmodelsforconsultation.

Thedefinitionofabusinessmodelforourpurposescoversthe

development, ownership, operation and maintenance, and

marketingofservices.TheConsultationaimstodevelopmodels

inlinewiththekeyquestionsinFigure4.

Thereareabroadrangeofpartiesthatcouldplayaroleina

distribution connected EES business model. The DNO has an

Figure 4 Consultation Questions

Ownership CommercialOperation Market

Whoshouldtakeriskonconstructionandoperation

oflargescalestorage(e.g.availabilityrisk)?

Whoshouldtakethebenefitandmanagetherisk

associatedwithcapturingthevalueofstorage

(i.e.market/operationrisk)?

Whichservicesandmarketscouldtheoperators

participateinandhowdotheycomplement/cannibaliseeachother?

KeyQuestions

Options

DNO

Energytraders/Aggregators

InstitutionalInvestors

IndustrialCustomers

RenewableProjectDevelopers

PrivateEquity

Suppliers

Community

SecurityofSupply

EmbeddedBenefits

FirmFrequencyResponse

STOR

EnergyArbitrage

CapacityPayments

PrimaryServicese.g.

SecondaryServicese.g.


modelwhereathirdpartybuildstheassetandtheDNOhas

no commercial relationship with the storage other than a

connectionagreement.Betweentheseextremesaremodels

whichinvolveacontractualrelationshipbetweentheDNOand

athirdparty.

2.1.BusinessModels

Wehaveselectedarangeofbusinessmodelstoillustratethe

landscapeofpotentialoptions.Thesearesummarisedatahigh

levelinFigure5.Thefivechosenbusinessmodelsrangefrom

aDNOmerchantcasewheretheDNOtakestotaloperational

controlandnothirdpartyisinvolved,toaChargingIncentives

Figure 5 Business Model Key Features

Fullmerchantrisk,exposedtopowerpriceandbalancingservices

DNOexposedtoincentivescheme

DNOexposedtoconstructionandoperationalrisks

LowcommercialriskforDNO

Noguaranteeofassetbeingbuild

• DNObuilds,ownsandoperatestheasset.Fulloperationalcontrol.

• DNOmonetisesadditionalvaluestreamsdirectlyonashorttermbasis(e.g.trading).

• Possiblebarriers:Costsofaccessingthemarket,DNOskillsandcapabilities,regulationandshareholderexpectationsofrisk.


• DNOhasDSOrole;coordinatingportfoliosofflexibilityforbothdistributionandwidersystembenefitthroughacentralizedcontrolmechanism.

• DNOcommercialriskisdependantondesignofincentivescheme.


• Priortoconstruction,longtermcontracts(e.g10years)forthecommercialcontroloftheassetoutsideofspecifiedwindowsareagreed.

• Dependantonthefeasibilityoflongtermcontracts.

• DNOoffersalongtermcontract(e.g.10years)forservicesataspecificlocationwithcommercialcontrolincertainperiods.

• Thirdpartyresponsibleforbuilding,owningandoperatingtheassetandmonetisingadditionalrevenuestreams.

• DNOsetsDUoStocreatesignalsforpeakshavingthatreflectthevalueofreinforcement.

• Barriers:nooperationalcontrolforDNO,thereforenoguaranteeonsecurity.

Model Keypoints Comments

DNOMerchant

DSO

DNOContracted

ContractedServices

Charging Incentives

Incr

easi

ngc

omm

erci

alri

skfo

rDN

O

Incr

easi

ngc

omm

erci

alri

skfo

rThi

rdP

arty


listofcriteriarepresentsahighlevelviewofthekeybusiness

model considerations. The relative importance of the criteria

maydifferbetweeninterestedparties.Wehavenotattempted

torankthesecriteria.HoweverwenotethattheDNOControl

requirementisconsideredaprimaryfactorbecausetheprovision

ofsecuritytothelocalnetworkisthedriveroftheNeedsCase.

Thecommonsetofqualitativecriteriathataretobeusedto

assess the business models, along with their descriptions is

showninTable3.

Inthefollowingsections,eachbusinessmodelisfirstcharacterised

according to its key exhibited features, namely storage

developmentandconstruction,financing,ownership,operations

andmaintenanceandcommercialcontrol.Second,eachbusiness

modelisassessedagainstacommonsetofqualitativecriteria.

ThecriteriaaimtocoverabroadspectrumoftheaspectsofanEES

projectonwhichthebusinessmodelmayhaveanimpact.

Thecriteriaareclusteredwithinfourbroadthemes–security,

assetvaluecostsandrisks,widerbenefitsandfutureproof.The

Table 3 Qualitative Assessment Criteria

Grouping

Security

AssetValue,

CostandRisks

WiderBenefit

FutureProof

Criteria

DNOControl

OptimisingValue

ofAsset

RiskAllocation

Financing

OptimisingValuefor

System

DynamicEfficiency

CarbonEfficiency

Scalability

Regulatory

Compatibility

Flexibility/

Optionality

CriteriaDescription&Examples

• EnsurethattheassetisavailableforDNOsecurity/constraintmanagementpurposes

whenrequired.

• DNOconfidenceinassetavailability.

• Incentiviseefficientusageoftheasset.

• Optimiseviablevaluestreams.

• Easeofadministrationandavoidanceofprohibitivecostsofdoingbusiness.

• Marketrisksallocatedtopartyabletotradeinthemarkets.

• Constructionriskallocatedtopartieswiththeappetiteandabilitytomanagethis.

• Operationsandmaintenancebypartywithabilitytomanagethis.

• Enableaccesstolowcostfinancing.

• Optimiseoverallmacro-economicandwelfarebenefits.

• Abilityforinnovationtooccurontechnologydevelopmentandtechnologychoice.

• Abilityforbusinessmodeltoevolveovertime.

• Incentivisesandenableslowcarbonpowersystem.

• Ensurethatbusinessmodelisscalableandcanevolveinresponsetochangingvalues

orvaluestreams.

• Compatiblewithcurrentmarketdesignsandcurrentregulation.

• Futureproofagainstexpectationsofchangestomarketdesignsorregulation.

• Consistentwithexpectationsofprojectlife:interactionwithpotentialfuturereinforcement.

• Allowsforthepossibilityofrelocationofthestorageassettoanotherpartofthenetwork.


TheDNOwouldleadthedevelopmentandconstructionofthe

storage facility, including all planning and consents. The asset

wouldthenbefinancedbytheDNOonbalancesheet,potentially

asaregulatedasset.Thiscouldbeagainstabaselineoftraditional

reinforcementwithDNOandCustomerssharingthebenefitsand

risksofanysavingsorcostoverrunsthroughasharingfactor.

2.1.1. DNOMerchant

In the DNO Merchant business model, the DNO takes full

ownershipandoperationof theassetand is responsible for

monetising theadditionalvalue streams. TheDNOdoes this

directlyintherelevantmarketsandthereisnothirdpartywith

adirectrelationshiptotheasset.Figure6belowillustratesthe

keyaspectsofthebusinessmodel.

Figure 6 DNO Merchant Model

Ownership

Development & Construction

• DNOmanagesdevelopmentasforanyothernetworkasset

Operations & Dispatch

• DNOhasfulloperationalcontrol

Finance

• FinancedbyDNO• Mayformpartof

theDNO’sRegulatedAssetBase

Maintenance

• DNOmaintainsasset

DN

O tr

adin

g op

erat

ion

Security of Supply

Embedded Benefits

Reactive Power & Voltage Support

Frequency Response

STOR

Energy Arbitrage (buy and sell)

Capacity Payments

Operation Services/ValueStreams

EnergySupply(internaltransfer)

DNO

DNO 3rdparty Contract Market

MonthlyFFRtenders

AnnualSTORtenders

EnergyTrading

CapacityAuction

RelationshipwithSupplier


Asummaryoftheprincipaladvantagesanddisadvantagesofthe

modelaccordingtothecommonsetofqualitativecriteriaisgiven

inTable4(seeTable3fordescriptionsoftheassessmentcriteria).

Table 4 DNO Merchant Model Qualitative Assessment

Grouping

Security

AssetValue,

CostandRisks

WiderBenefit

FutureProof

Criteria

DNOControl

OptimisingValue

ofAsset

RiskAllocation

Financing

OptimisingValuefor

System

DynamicEfficiency

CarbonEfficiency

Scalability

Regulatory

Compatibility

Flexibility/Optionality

Description


• DNOscurrentlylacktheexperience,skillsandcapabilitiestotradeandriskmanagetheasset.

• ThefirstofakindcostofdevelopingthesecapabilitieswillbehighunlesstheDNOhas

alargeportfolioofassetstospreadthecostsacross.

• UntilDNOshavedevelopedtheinternalexpertisethereisariskthattheassetis

utilisedsub-optimally.

• DNOisrequiredtotakeriskthatvalueofresidualrevenuestreamswillnotdrive

asavingwhencomparedagainsttraditionalreinforcement,whichisasignificant

departurefromtheirbusinessasusual.

• RiskallocationwithCustomersmustbealignedwiththeproportionofregulatoryfinancing

• Customerscouldshareinadditionalcapturedvaluenotpreviouslyavailable,depending

ontheproportionofregulatoryfinancing.

• Financingasaregulatedassetcouldreducethecostofcapital.

• Forasingleassetthereisnoaggregationthereforecostsoftradingwillbehighasa

proportionofrevenues.

• MaybemoresuitableifDNOcandevelopmultipleassetstospreadcostsoftradingand

acceleratedevelopmentofinternalexpertise.

• AssetmaybeaggregatedwithassetsofotherDNOsinajointtradingorganisation.

• DNOhasfullcontroloverthestoragetechnologyandcantargetthedevelopmentof

specifictechnologiesifrequired.

• Thebusinessmodelwouldbecomemoreefficientasprojectsarerolledoutandthe

DNObuildsitsinternaltradingandriskmanagementexpertise.


• Canbescaledbutwouldbelimitedbydevelopmentofinternaltradingandrisk

managementexpertise.

• TheremayberegulatorybarrierstotradingofpowerbyDNOandthefinancingofthe

projectasaregulatedasset.

• DNOhasfullcontroltorelocatetheassettonewconstrainedlocationifrequired.


potentially aggregate the asset with other DNOs in a joint

tradingorganization.

Anadditionalcomplicationwouldbehowtheassetisfinanced

and operated within the DNO business. If the asset was a

regulated asset to be fully financed by the Customers, but

operated by the DNO there could be a misplaced incentive

on the DNO to operate the asset’s ancillary services in a

profit optimising manner as they would not accrue any of

this additional financial benefit. (The effect of the different

financingoptionsonpricecontrols,alongwiththecost,riskand

benefitsharingbetweentheDNO,thirdpartyandCustomersis

discussedindetailinSection2.3).

2.1.2.DistributionSystemOperator

Our definition of a Distribution System Operator (DSO) is

discussedinBox2.TheDSOmodelforstorageassumesafuture

world inwhichthenecessary regulatorychangeshavebeen

made.UnderaDSOmodeltheDNOmightown,operateand

maintainelectricalenergystorageaspartofawider roleof

activelymanagingitsnetwork,akintotherolethatNationalGrid

playsatthetransmissionlevel.UnderthismodeltheDSOmight

leadthedevelopmentandconstructionof theasset,finance

itsconstructionandoperationandthentake fullcommercial

control.TheDNOwouldaccruealltheprojectcostsandbenefits

andtherewouldbenodirectthirdpartyinvolvement.

AlternativelytheDSOmightsimplycontractforservicesfrom

electricalenergystorageonanexclusiveorprioritybasis.Thisis

subtlydifferentfromtheGBTSOmodelwheretheGBTSOdoes

notownanygenerationorstorage.

TheDSOmodelcanbeconsideredasavariantoftheDNOMerchant

model,butwiththeadditionofanewregulatoryincentiveregime

tomanagetherisksassociatedwiththevalueoftheasset.

Under the DNO Merchant model the DNO would have full

operational control of the asset and therefore the ability to

meet any security or constraint management obligations.

TheassetwouldbefinancedontheDNO’sbalancesheetasa

regulatedasset.Thereareanumberofvariationsontheextent

whichCustomerswouldfundtheassetandtakesomeofthe

riskaroundthereturns,discussedinSection2.3.ADNOislikely

tohavealowerWeightedAverageCostofCapital(WACC)than

themajorityofpotentialthirdparties.Thismaymeanthatthe

asset can be financed at lower cost (albeit with Customers

taking some or all of the risk). However, the risk profile of

thestorageishigherthantherestoftheDNO’sbusinessand

creates exposure to new risks, and a significant portfolio of

storagemayincreasetheDNO’scostoffunding.Thereforethe

DNO may assess the development using a marginal cost of

capitalimplyingahigherrateofreturn.Thismaydependon

howmuchriskissharedwithCustomers.

TheDNOMerchantmodelcarriesthegreatestcommercialrisk

fortheDNO.TheDNOwouldneedtotradeinthewholesale

energy market and participate in procurement mechanisms

for ancillary services (see Appendix 1 for a description of

the possible ancillary revenue streams). The value that can

bedrawn from theseancillary serviceswouldbe limitedby

theDNO’sde-minimisthreshold,whichis2.5percentofthe

sumofthelicensee’ssharecapital,itssharepremium,andits

consolidatedreserves.

A DNO trading and risk management operation would also

be required. This would be a significant deviation from the

core business of the DNO and would require additional set

upandoperationalcostsforthefirstofakindproject.These

costswouldbeexpected todecreasewith increasing rollout

ofsimilarprojectsandtheestablishmentoftheDNOsinternal

trading and risk management expertise. DNOs could also


ADistributionSystemOperator(DSO)hasaccesstoaportfolioofresponsivedemand,storageandcontrollablegeneration

assetsthatcanbeusedtoactivelycontributetobothdistributionnetworkandwidersystemoperation.ADSObuildsand

operatesaflexiblenetworkwiththeabilitytocontrolloadflows.Thecombinationofahighlyflexiblenetworkandaccess

todemandandgenerationresponseallowstheDSOtocontributetotheincreasingchallengeofencouragingdemandto

followgeneration.Thiswillbecomeincreasinglyimportantashighervolumesoflowcarbon,zeromarginalcostintermittent

generation(suchaswindandsolar)becomeavailabletosupplyGBdemand.However,changeddemandprofilesmightalso

giverisetohigherpeakdemandsoccurringwhenavailabilityoflowcarbongenerationishigh.Moreover,inordertoprovide

fastlow-costconnectionstorenewabledistributedgenerators,theuseofinterruptibleconnectionarrangements,suchasis

beingtrialledthroughUKPowerNetworks’FlexiblePlug&PlayNetworksproject,mightbecomemorecommon,particularly

foronshorewindgenerationinrespectofwhichatypicalloadfactormightbearound25%.

Itfollowsthataswellashelpingtosupportthemarket,aDSOwillalsohavetoconsiderdistributionnetworkconstraintsand

theopportunitiesforusingcommercialinnovationssuchasdemandsideresponse(DSR)inordertoreducetherequirement

fornetworkreinforcement.TheDSOrolecouldalsoentailcloserinteractionswiththeNationalElectricityTransmissionSystem

Operator(NETSO),suchasaresponsibilitytoassistwithbalancingatanationallevelandprovidingancillaryservicessuchas

reserve,frequencyresponse,andvoltageandreactivepowermanagement.

ItfollowsthattheDSOroleisconceptuallysimilartotheNETSOroleunderwhichNationalGridmanagesbalancing,reserveand

constraintsontheGBinterconnectedtransmissionsystemovertimeperiodsrangingfromoutagemanagementplanningtimescales

tosecondbysecondsystemfrequencycontrol.TheNETSObalancesthesystemusingarangeofmechanisms.Theseinclude:

• Ancillaryandcommercialservicessuchasreactivepower,frequencyresponseandreserveserviceswhichtheNETSOwill

contractfordirectlywiththeserviceproviders;

• Contractnotifications–wherebytheNETSOcanbuyandsellelectricityaheadofGateClosuredependingonwhetherthere

isexpectedtobeasurplusorshortfallofgeneration;and

• TheBalancingMechanismwherebytheSOinstructsBalancingMechanismUnitstoincreasegeneration/reducedemand

(acceptanceofanOffer)orincreasedemand/reducegeneration(acceptanceofaBid).Bid/Offeracceptancesaremade

onlyfollowingGateClosure6.

NationalGridisalsoresponsibleforensuringthattheGBinterconnectedsystemremainswithinsafeoperatinglimitsandthatthe

patternofgenerationanddemandisconsistentwithanysystemtransmissionrelatedconstraints(forexampleduetoaplannedoutage

ofacircuit).WhilstNationalGridwillendeavourtocoordinatenetworkoutagescoincidentwithrelevantgenerationoutagesinorderto

minimiseconstraintcosts,itmaybenecessarytotakeactions(byenteringintoaTransmissionConstraintAgreement,tradingortaking

actionsintheBalancingMechanismwithgenerators,suppliersandlargecustomers)toresolveconstraintsonthetransmissionsystem.

Box 2 Definition of a DSO

6 Note: whilst a Bid/Offer acceptance might move a generator or Supplier from their contracted position such that they are technically imbalanced, this does not affect their settlements (cash-out) position.


NationalGridisincentivisedontheprocurementandutilisationofservicestomaintaintheenergyandsystembalanceand

othercostsassociatedwithoperatingthesystem.Userspayforthecostoftheseservicesandanyincentivisedpayment/

receiptsthroughtheBalancingServicesUseofSystem(BSUoS)charge.TheSOforecaststhecostsatthestartofapricecontrol

period,andOfgemplacesincentivesontheSOtokeepcostswithintheseforecasts.Thesechargesarereconciledagainst

actualcostssothattheSOisnotexposedtoexcessiveoverorunderspend.Forexample,iftheSOoverorunderspends

againstitsforecastitisexposedtosomeoftheextracostorsavingrespectively,subjecttoadead-band.TheamounttheSO

isexposedto-the‘sharingfactors’-were25%foreitheroverspendorunderspendfortheperiod2011-20137.

ADSOmighttakeanactiveroleinmanagingthedistributionnetworkthroughdispatchingorcurtailingelectricalenergyat

differentlocations,forexamplethrough:

• curtailingorconstraining-ongeneration(dependingonwhetherthenetworkisconstrainedforgeneratorexportorrequires

generatorsupportduetolossofsecurecapacity);

• dispatchofelectricalenergystorage;or

• curtailmentofdemandthroughDSRcontracts,curtailmentofgenerationoruseofstorage)tominimisethecostofresolving

constraintsandensuringnetworksecurity.

TheDNOregulatoryframeworkdoesnotexplicitlyprovideforsuchmarket-based/relativelyhighrisksolutions.Itisdesigned

forcapitalandoperatingexpenditurethatcanbeaccuratelyforecastandthensetasabaselineex-anteallowancebyOfgem

aheadofanapplicablepricecontrolperiod.Moreover,therearecurrentlylimitationsembeddedintheDNOlicencethatlimits

theamountofgenerationorstorage8whichtheymayown.TheDNOdoeshoweverhaveincentivestomakeefficiency

savingsthroughacost-sharingmechanismwherebysavedcostsaresharedwithconsumersbutanyoverspendisonlypart

fundedbyconsumers.Someflexibilityisprovidedthroughtheuseofuncertaintymechanisms9,whichallowtheDNOtoalter

itsallowedexpenditureduringthepricecontrolperiod.

TheDSOmodelmaybemostrelevantinfutureasDNOstakeamoreactiveroleinmanagingthedistributionnetwork,e.g.

curtailmentofembeddedgenerationanddispatchofDSR.HoweveritcouldalsoofferDNOsamechanismthroughwhichto

manageadditionalvaluestreamsofstorageoutsideofthepricecontrolstructure.ItcouldalsogiveOfgemanumberoflevers

toplaceincentivesontheDNOtomaximisethevaluestreamstoprovidethemostbenefitforCustomers.

Box 2 Definition of a DSO

7 Ofgem, 2011. National Grid Electricity Transmission System Operator Incentives from 1 April 20118 Existing storage is operated under Generation Licences, There is some current debate as to whether storage should be treated as generation for the purposes of the regulatory

limitation on DNOs.9 Uncertainty mechanisms can include indexing, volume drivers, triggers, logging-up provisions and re-openers.


therefore forward compatibilitywithDSO-typearrangements

shouldbeaconsideration.

TheSNSprojectisintendingtoexplorefurthertheseregulatory

barriersfollowingthefirst-of-a-kindtrialsofthestoragefacility

forarangeofvaluestreams,whichwillbesharedwithDNOs

and industry. A summary of the principal advantages and

disadvantagesofthemodelaccordingtothecommonsetof

qualitativecriteriaisgiveninTable5.

It should be noted that while the DSO model is contingent

on significant regulatory development, these developments

arenotnecessarilyrequiredfordistribution-scaleEESprojects.

UndertheSmartGridsForumWorkstream610,DECC,Ofgemand

theindustryareexploringthepotentialscopeoftheDistribution

SystemOperator(DSO)roleforDNOs.Theinitialviewfromthe

SmartGridsForumisthataDSOmodelisnotrequiredforRIIO-

ED1.Howeverdistribution-connectedEESbuiltunderRIIO-ED1

may find itself operating under a DSO framework in future;

Table 5 DSO Model Qualitative Assessment

Grouping

Security

Assetvalue,

costandrisks

Widerbenefit

Futureproof

Criteria

DNOControl

OptimisingValue

ofAsset

RiskAllocation

Financing

OptimisingValuefor

System

DynamicEfficiency

CarbonEfficiency

Scalability

Regulatory

Compatibility


Description


• InaDSOworldtheDNOwouldactivelymanagethebalancingofthenetworkandwould

thereforebebetterplacedtooptimisethevalueoftheassetthanapresent-dayDNO

• Anexperiencedthirdpartyaggregatormightstillbebetterplacedtooptimisetheoperation

oftheasset.

• Ifaregulatedasset,DNOcansharerisksandbenefitswithCustomersdependingon

thesplitofregulatoryfunding–agreaterproportionofsystemvaluecapturedfor

customers.

• DSOrolestillatconceptualstage(WS6ofSGF).

• Financingasregulatedassetcouldreducethecostofcapital.

• Customerscouldshareinadditionalcapturedvaluenotpreviouslyavailable,depending

ontheproportionofregulatoryfinancing.

• ForDSO,maybeadditionalincentivesonpowerquality,lossesetc.whichcangenerate

additionalvalue.




• RelevantinfutureasactivenetworkmanagementbyDNOdevelops,activeassets

increase,andregulationsaredeveloped.

• Maybemoreappropriatewhenmultipleactivelymanagedassetsareinplace.

• Themodelisreliantonfutureregulatorydevelopments.

• DNOcontroltorelocateassettonewconstrainedlocationifrequired.

10 http://www.ofgem.gov.uk/Networks/SGF/Pages/SGF.aspx


asset’soperationand theassociatedcommercial riskwere to

becontractedtoathirdparty(whowouldbeexpectedtobe

abletobettermanagethisrisk).Also,whilethebalancingof

thenetworkwouldbeacorepartoftheDNO’sbusinessundera

DSOscenario,aspecialistthirdpartymaystillbebetterplacedto

operateandrisk-managetheassettoensuregeneratedpower

andancillaryservicesachievethemaximumpossiblereturns.

2.1.3. DNOContracted

TheDNOContractedmodeldiffersfromtheDNOMerchantmodel

duetotheinvolvementofathirdpartytomanagethecapacityof

theassetwhenitisnotrequiredforsecuritypurposes.TheDNO

wouldstillfinance,maintainandoperatetheasset,butwould

dispatchitforancillaryservicesattheinstructionofathirdparty.

Under a futureDSO incentive scheme theDNOwould still be

incentivised to actively manage the distribution network to

reducecosts.Assuch,theoperationoftheassetwouldalready

beacorefunctionoftheDNO’sbusiness,decreasingoperational

costs in setting up a trading and risk management business

andincreasingrevenueefficienciesthroughalreadyestablished

expertise.Storagecouldbeoneofmanytechnologiesforenabling

this, forming one part of a portfolio of active management

technologiesandsystemswithagreaterwidersystemvalue.

However,theobviousdrawbackofthismodelisthatitisstill

verymuchahypotheticalcase.Also,whiletheextenttowhich

theDNOwouldneedtocarrythecommercialriskoftheasset’s

operation is unknown, it would still be greater than if the

Figure 7 DNO Contracted Model

Ownership


• DNOmanagesdevelopmentasforanyothernetworkasset


• DNOhasfulloperationalcontrol

• DispatchedforancillaryservicesatinstructionofThirdparty

Finance

• FinancedbyDNO• Mayformpartof

theDNO’sRegulatedAssetBase

Maintenance

• DNOmaintainsasset

Security of Supply

Embedded Benefits


Frequency Response

STOR


Capacity Payments

Operation Services/Valuestreams

AnnualSTORtenders

Energytrading

Capacityauction


DNO


Trad

ing/

aggr

egat

ion

MonthlyFFRtenders

EnergysupplyTollingcharge


asset,andisresponsibleformeetingthethirdparty’sdispatch

instructionsintheperiodswhenthethirdpartyhascommercial

controloftheasset,improvingtheoverallDNOconfidencein

theavailabilityofthecapacitywhenrequired.

The disadvantage with the DNO Contracted model is that

thetermsofthethirdparty’saccesstothecapacitycouldbe

complexandneedtobeclearlydefinedpriortodevelopment

of the asset. The ability to define the times at which the

third party will have access to the asset is dependent on

forecastingof the specificnetwork constraint requirement,

which ismainlydependenton the localnetworkareaand

local demandgrowthwhich is largely uncertain (offset by

any growth in embedded generation or DSR). Also, these

contracted security requirement windows would aim to

allow as much flexibility as possible for the monetization

of additional value streams. As such, they would aim to

usethesmallestwindowthatguaranteestheDNOsecurity

requirementsaremet.However,therewouldbeaninherent

and largely unavoidable conservatism in these contracted

security requirementwindows,whichwould likelybe long

termcoarseagreementswithpoorgranularityforrealtime

deviationsofrequirement.Thiswouldcreateapotentialvalue

losstotherevenuestreamsoftheassetancillaryservices.

Underthismodel,thethirdpartywouldenterintoalongterm

capacitycontract(ideallytotheendoftheoperationallifeof

theasset)whereadditionalvaluewouldaccrueto theDNO

throughfixedannualavailabilitypaymentstotheDNO,orsome

form of percentage pass-through of value. The DNO would

thereforebeinapositiontomakeitsinvestmentdecisionby

comparingthenetpresentvalueoftheavailabilitypayments

underthelongtermcapacitycontractsagainsttheincremental

costofstorageoverandabovetraditionalreinforcement.

Asummaryoftheprincipaladvantagesanddisadvantagesof

themodelaccordingtothecommonsetofqualitativecriteria

isgiveninTable6.

TheprincipaladvantageoftheDNOcontractedmodelisthat

some, or all, of the risk around monetising the additional

valuestreams is transferredaway fromtheDNO. Inviewof

theinherentuncertaintyastothelongtermvalueofadditional

revenuestreams,thethirdpartywouldbeexpectedtobidat

adiscounttothe longtermexpectedvaluefor thiscapacity.

Theextentoftheriskswillobviouslydeterminetheextentof

thediscountthatitapplies,whichwillinturnaffectthelikely

viabilityoftheinvestmentfromtheDNO’sperspective.Afurther

advantageisthattheDNOretainsfulloperationalcontrolofthe


Grouping

Security

AssetValue,

CostandRisks

WiderBenefit

FutureProof

Criteria

DNOControl

OptimisingValue

ofAsset

RiskAllocation

Financing

OptimisingValue

forSystem

DynamicEfficiency

CarbonEfficiency

Scalability

Regulatory

Compatibility

Flexibility/

Optionality

Description

• DNOhasfulloperationalcontrol,andcommercialcontrolindefinedsecurityperiodsonly.

• Possibilityofcomplexcontractualtermstoensurethemostvaluableservicesare

availabletothirdpartywhilstensuringsystemsecurity.

• LikelytobeconservativeSecureCapacityWindowswithsignificantvaluelosses(i.e.lost

ancillaryservicebenefits).

• AssumesthatSecureCapacityWindowscanbedefinedwithenoughcertainty.

• Thirdpartymayheavilydiscountlongtermvalueofadditionalrevenues,dependingon

theriskinessoftheserevenues.

• Thirdpartyexpertisehoweverwould,intheory,ensuretheassetisoptimallyutilised.

• TheDNOwouldnotberequiredtoestablishanenergytradingandriskmanagement

functionwithintheirbusiness.

• TheDNOwouldpasssomeofthecommercialrisktothethirdparty,theextentof

whichwoulddependontheSecureCapacityWindowsandtollingagreementbetween

thetwoparties.

• Financingasaregulatedassetcouldreducethecostofcapital.

• Customersshareinadditionalcapturedvaluenotpreviouslyavailable,dependingon

theproportionofregulatoryfinancing.

• Thetermsofthetollingagreementcouldallowfortheassettobeaggregatedintoa

largerthirdpartyportfolio.



• Thebusinessmodelwouldbecomemoreefficientasprojectsarerolledoutandsubject

totheDNOestablishinganinternaltradingandriskmanagementexpertise,the

organicgrowthofthesecapabilities.


• Thirdpartymaybeabletoaggregateacrossmultipleassetswhichincreasesscalability.

• Theremayberegulatorybarrierstothefinancingoftheprojectasaregulatedasset.

• IftheDNOwastocarrysomeofthemerchantexposure(amodelvariation),theremay

beregulatorybarrierstotradingofpowerbytheDNO.

• Optionalitytore-locateassetmayberestrictedifnotexplicitlycoveredbytermsof

agreementwiththirdparty.

Table 6 DNO Contracted Qualitative Assessment


Figure 8 Contracted Services Model

Ownership


• Thirdpartymanagesdevelopment.LikelytorequireDNOinputatthisstagetodefinerequirements


• Thirdpartyhasfulloperationalcontrol,subjecttoobligationstoDNO

Finance

• Financedbythirdparty

Maintenance

• Thirdpartymaintainsasset

Trad

ing/

aggr

egat

ion

Security of Supply

Embedded Benefits


Frequency Response

STOR


Capacity Payments

Operation Services/Valuestreams

MonthlyFFRtenders

AnnualSTORtenders

Energytrading

Capacityauction


Thirdparty


DNOservicescontract(TollingCharge)

2.1.4.ContractedServices

Underthismodel,theDNOidentifiesthesecurityrequirement,

orothernetworkconstraintformitigation,selectsthesiteand

thenrunsanopentenderforthirdpartiestobuildandoperate

storage.TheDNOsetstechnicalrequirementsthatmustbe

met, such as, the Secure Capacity Windows, the minimum

exportingcapacityandstoragecapacityoftheasset.TheDNO

offersafixedannualpaymentinreturnforthesecurityservices

providedbythethirdparty.


Grouping

Security

AssetValue,

CostandRisks

WiderBenefit

FutureProof

Criteria

DNOControl

OptimisingValue

ofAsset

RiskAllocation

Financing

OptimisingValue

forSystem

DynamicEfficiency

CarbonEfficiency

Scalability

Regulatory

Compatibility


Description

• Thirdpartyhasfullcontroloftheasset.

• DNOdoesnothaveoperationalcontrol,ortangibleassuranceofsafetyortechnical

aspects.

• Possibilityofcontractualchallengesinensuringsystemsecurityatbestvalue.

• Thirdpartymayheavilydiscountlongtermvalueofadditionalrevenues.

• Littleexistingappetitefromthird-partieswhileunproven.

• Fullcommercialriskwiththethirdparty.

• LimitedinitialinvestmentbyDNOs/Customers.

• Additionalsystemvalueisnotdeliveredtocustomers.

• CostofcapitalmaybehigherthaniffinancedbyDNO.

• AssetcanbeaggregatedintoalargerThirdPartyportfolio.

• Dependingonthetenderrequirements,notnecessarilypromotingthedevelopmentof

energystorageaboveothersolutions.

• DNOcouldstilldefinethestoragetechnology,oropenuptoalltechnologies.


• Thirdpartymaybeabletoaggregateacrossmultipleassets,whichincreasesscalability.

• Likelytofacefewregulatorybarriers.

• DNOoptionalityonrelocationofassetisremoved.

fullcommercialcontroloftheassetandwouldreceivetolling

paymentsfromtheDNOcombinedwithuncertainreturnsfrom

additionalvaluestreams.


themodelaccordingtothecommonsetofqualitativecriteriais

giveninTable7below.

UndertheContractedServicesmodelthesuccessfulthirdparty

would leadthedevelopmentandconstructionof thefacility,

including all consents and planning. The third party would

thenfinance,own,operateandcontroltheasset.Theoriginal

tender agreement would guarantee them revenue streams

across the expected life of the asset in return for meeting

theDNO’ssecurityrequirements.Thethirdpartywouldhave

Table 7 Contracted Services Model Qualitative Assessment


2.1.5.ChargingIncentives

The Charging Incentives model is one under which the

DNO ensures that the DUoS charging (for both generation

and demand) creates the right incentives for decreases

in (net) demand at peak times in the location requiring

reinforcement. Third parties may or may not respond to

the incentives by building storage (or equally by building

embeddedgenerationorwithDSM,which are outside the

scope of this consultation). The DNO has no control over

howmuchcapacityisbuiltorwhatthetechnicalcapabilities

are.TheDNOalsohasnowaytoensurethatthecapacityis

availableforsecuritypurposeswhenrequired.TheDNOmay

thereforehavetoover-incentivisethethirdpartyintermsof

thepaymentstructure.

Thethirdpartywouldleadthedevelopmentandconstruction

of thestorage facilitywitha technologyoptionand location

of their choice. The thirdpartywouldown,finance, operate

andmaintaintheasset,holdingfullcommercialcontrolofits

operation.TheDNOwouldholdnooperationalcontrolofthe

assetandhavenoguaranteeonsecurity,butwouldsetcharging

signals(suchasnegativeDUoSorcreditsforgenerationduring

the‘SuperRed’timeperiod11)forpeakshavingthatreflectthe

valueofreinforcement.Thesesignalswouldpotentiallyneedto

reactdynamicallytosystemconditions,whichisnotcurrently

thecase.


themodelaccordingtothecommonsetofqualitativecriteria

isgiveninTable8.

FromtheperspectiveoftheDNO,theprincipaladvantageofthe

ContractedServicesmodelisthattheconstruction,operational

andcommercialriskisremovedorsignificantlydecreasedfrom

theDNOMerchantandDNOContractedmodels.Theproject’s

construction, operational and commercial risk is instead

passed toa thirdparty,who couldhavealreadyestablished

anexpertisethatwouldallowthemtoeffectivelymanagethis

risk.TheDNOalsoavoidstheupfrontcapitalcostoffinancing

theasset,insteadspreadingthiscostoutoverthelifetimeofthe

assetasanannualtollingcharge.TheDNOalsoavoidshaving

toestablishaninternalenergytradingandriskmanagement

servicewhichinsteadcouldbetakenbyanestablishedfunction

withinthethirdparty’sbusiness.

However,astheDNOpassescontroloftheassettoathirdparty

theyalsolosedirectcontrolofitsoperation.TheDNOinstead

relies on the Third Party to provide the storage for security

whenrequired.Thisinitselfaddsaconsiderableprojectriskto

theDNOwiththechancethatthethirdpartywouldoverutilise

theassettomaximizetheircommercialbenefits,thusfailing

tomeettheirsecurityandconstraintrequirementobligations.

Forthefirstprojects,therisksassociatedmaylimitthenumber

of potential providers and lead to a lack of competition in

the near term. This may be compounded by the lack of

understandingfromthemarketastothenatureoftheservices

required.Thereisariskthattheconsiderableupfrontcostand

commercialriskthatwouldneedtobecarriedbythethirdparty

wouldagainlimitthemarketsizewiththelackofsuitableand

interestedthirdparties.

11 Super Red time band is a seasonal time of day period determined by each DNO to reflect the time of system peak


Grouping

Security

AssetValue,

CostandRisks

WiderBenefit

FutureProof

Criteria

DNOControl

OptimisingValue

ofAsset

RiskAllocation

Financing

OptimisingValue

forSystem

DynamicEfficiency

CarbonEfficiency

Scalability

Regulatory

Compatibility

Flexibility/

Optionality

Description

• Thirdpartyhasfullcontroloftheassetanditsoperation.

• NooperationalcontrolforDNO–noguaranteeonsecurityorthatassetisbuiltinoptimal

locationorisbuiltatall.

• Chargingincentivesmayneedtobeverylarge.

• ThirdpartymayheavilydiscountlongtermvalueofadditionalrevenuesandofDUoS

incentives.

• Amarketledsolutioncould,intheory,producethemosteconomicalsolution.However

thisdoesnotrecognisethebarriersandrisks.

• Allcommercialriskispassedtothethirdparty.

• NocommercialrisktoDNOorCustomers.

• CostofcapitalmaybehigherthanforDNO.

• Thirdpartiescouldaggregatetheassetwithothersinawiderportfolio.

• Allowsmosteconomictechnologychoice(notnecessarilystorage).

• Doesnotnecessarilydriveinnovationinstorage.

• Thereisnoguaranteethatacarbonefficientsolutionwouldbeproposedbythethirdparty.

• Appearstobeeasilyscalable.

• Likelytofacefewregulatorybarriers.

• DNOoptionalityonlocationofassetisremoved.

• TheThirdPartyhasfullcontrolovertheasset’slocation.

Table 8 Charging Incentives Model Qualitative Assessment


Under thismodel theDNOallows themarket to define the

mosteconomictechnologychoiceandlocationsolutionforthe

project.TheDNOwouldcarrynoneofthecommercialrisk,with

alltheprojectandcommercialriskpassedtoaspecialistthird

party (who would be expected to have the experience and

capabilitiestomanagethisrisk).

However,withthismarketbasedsolutionthereisariskthat

thirdpartieswouldnotreacttothepricesignalsandtheasset

wouldnotbebuiltatall.Ifthestorageassetweretobebuiltit

wouldbebyathirdpartyandtheDNOwouldhavenocontrol

ofitslocationortechnologychoice,bothofwhichmightnot

besufficienttomeettheDNOssecurityrequirements.Also,the

DNOwouldhavenocontroloftheasset’soperationandassuch

noguaranteethattheassetwouldbeavailabletoprovidethe

securityrequired.Thismaybeacceptableiftherewerealarge

numberofsmalldistributedassetssupportingasingleregion,

butisabarrierifsecurityisbasedonasingleasset.

Anotherconsiderationisthatthechangestochargingrequired

tocreatetheeconomicsignalscouldaddsignificantcomplexity

toDUoStariffs.

2.2.QualitativeAssessmentSummary

We have performed an assessment of the business models

againstthesetofqualitativecriteriapresentedinTable3,and

usedthistoaidtheselectionoftwobusinessmodelsforfurther

development.Thelistofcriteriarepresentsahighlevelviewof

thekeybusinessmodelconsiderations.Wehavenotassigned

weightingstothecriteriaasthesewouldbehighlysubjective

depending on stakeholder viewpoint. We invite interested

partiestopresenttheirownviewsinthisarea.

Wehavescoredeachmodelagainst thecriteriaabove.Each

model is scored from1 to5where5 is thebest score. The

scoringisnecessarilysubjectivebutprovidesacommonbasis

fordiscussion.


Table 9 Qualitative Scoring Matrix

DNOControl

Optimising

ValueofAsset

RiskAllocation

Financing

OptimisingValue

forSystem

Dynamic

Efficiency

Carbon

Efficiency

Scalability

Regulatory

Compatibility

Flexibility/

Optionality

DNOMerchant

5

3

1

4

3

2

4

2

2

5

DistributionSystemOperator

5

3

2

4

3

2

4

3

1

5

DNOContracted

4

4

4

4

4

3

4

4

4

3

ContractedServices

3

4

3

2

4

5

4

5

5

1

Comments

Highestscoreformodels

whereDNOhasfull

operationalcontrol.

Assumethirdpartybetter

placedtooptimisevalue

thanDNO.

Allocationofmarketrisksis

akeydriver.

AssumeDNOhaslower

costofcapitalforstorage

investments.

Assumethatthirdparty

aggregationcreatesmore

valueforsystem.

Modelswhichenable

competitionandtechnology

neutralityscorehighly.

Allscorehighlyexpectfor

ChargingIncentives,where

thereisnoguaranteethat

storagewillbebuilt.

AssumeDNOdominated

modelsarelessscalable.

AssumethatDNOlicence

conditionsmaylimitDNO

tradingofpower.

UnderDNOownedmodels,

DNOkeepslongterm

optionalitytore-locateasset.

ChargingIncentives

1

2

2

2

2

5

2

2

5

1


Figure 9 Qualitative Assessment Summary

1

2

3

4

5

DNO Control

Dynamic Efficiency

Optimising Value for System

Carbon Efficiency

ScalabilityFinancing

Regulatory Compatibility Risk

Allocation


Optimising Value of Asset

Key

DNO Merchant

DNO Contracted

Charging Incentives

Distribution System Operator

Contracted Services


While the DNO would have full control over the asset, they

wouldberequiredtobuildanin-houseenergytradingandrisk

management capability tomonetize theseadditionalbenefits.

ThisinitselfisasignificantdeviationfromtheDNOscorebusiness

andisaprohibitiveup-frontcost.Thereisalsoanincreasedrisk

thattheprojectwouldlosethevalueoftheseancillarybenefits

whicharecriticaltothecommercialviabilityoftheproject.

TheDSOmodelscorespoorlybecausetheregulatoryregime

is yet to get beyond the earliest stages of discussion. This

modelwillbeofmoreinterestinthefuturebutforthecurrent

discussion it is relevant toexplore thecompatibilityofother

modelswithafutureDSOworld.

Based on this assessment the lead models selected for

furtherconsiderationaretheDNOContractedandContracted

Servicesmodels.

2.3.RegulatoryTreatment

WithregardstotheDNO’spricecontrol,therearefourdifferent

regulatory treatment options of interest across the five

proposedbusinessmodels.Apartfromthemodelswherethe

asset is financedbya thirdparty, theDNOparent company

has the option to finance the asset as a non-regulated or

The DNO contracted model scores consistently well across

mostmetrics,duetotheallocationofrisksandresponsibilities

betweentheDNOandthirdparty.Itscoresneutrallyondynamic

efficiencybecauseitdoesnotencouragetheopencompetition

in technologiesormodels that thethirdpartymodelsallow.

The flexibility/optionality for the DNO is restricted by the

contractualrelationshipwiththethirdparty.

The Contracted Services model scores well on dynamic

efficiency, scalabilityand regulatory compatibility,asa result

of being third party owned. It scores poorly on flexibility/

optionalityfortheDNObecausethethirdpartyownstheasset

andthereforeredeploymentoftheassetmaynotbepossible.

Theremainingthreemodelsscorepoorlyonspecificareas.The

ChargingIncentivesmodelofferstheDNOnocontroloverthe

assetstechnologychoice,itslocation,oritsoperationtoprovide

securityorancillaryservices.TheChargingIncentivesmodelas

suchdoesnotguaranteetheassetscorefunctionofmeeting

theDNOssecurityobligationwouldbemet,andistherefore

notconsideredfurther.

TheDNOMerchantmodelscorespoorlyonriskallocationasthe

DNOretainstheriskofmonetizingtheadditionalvaluestreams.

Figure 10 Regulation Options

• ContractedServices

• ChargingIncentives

RegulatoryHandling

RelevantModels

Non-regulated Regulated

ProjectCostAccruedto ThirdParty DNOParentCompany

DNOParentCompanyandConsumerSplit

Consumer

• DNOMerchant

• DSO

• DNOContracted

• DNOMerchant

• DSO

• DNOContracted

• DNOMerchant

• DSO

• DNOContracted


assetfinancedon theDNOparentcompany’sbalancesheet

theDNOparentcompanywouldcarry theprojectcostsand

all the commercial risk. The regulated value of the asset

underthepricecontrolwouldbethecostofoffsettraditional

reinforcement.Customerswouldtakenoriskonthefinancial

benefits from the assets additional services, which would

accruedirectlytotheDNOparentcompany.Inthisregulated

DNOparent companyfinanced scenario the cost of security

ispassedtotheCustomersviaappropriatepricecontrolsina

similarmannertotypicalnetworkreinforcementcosts.

CustomerRegulatedFinancedScenario

The third option is the opposite extreme of the regulated

scenario,where the regulated value of the asset is the full

capital cost of the asset. In this scenario Customers would

seethesecuritybenefitandwouldalsoaccruetheadditional

financial benefit from the ancillary revenue streams. This

scenariocreatesamisalignmentofincentives,withtheDNO

facing no direct incentive to operate the asset in a profit

optimizingmanner. If theassetweretobeoperatedbythe

DNO,asisthecaseintheDNOMerchantmodel,therewould

beamisalignmentoftheproject’scostandrevenuestructures

wherethebenefitofoperatingtheassetwouldnotaccrueto

the party that would carry its operational cost, namely the

DNO.(Thisexcludesthepossibleresetofallowedrevenuesthat

mightoccuratpricecontrolreviewiftherevenuegeneratedby

storageissignificantlyoutoflinewithotherDNOs)

Thiscasealsoposes thequestionofhowtheseadditional

benefitswouldbepassedtotheCustomersunderexisting

pricecontrolarrangements.Thecostoftheassetalongwith

the benefit of the offset traditional reinforcement could

bedealtwithandpassedtoCustomersviatreatmentasa

regulatedasset. Theuncertainancillary revenuesover the

lifetimeoftheassetwouldthenoffsetthepaymentsmade

byCustomers.

regulatedasset.Furthermore,iftheprojectwastobefinanced

asaregulatedassettheproportionthatwouldbefinancedby

theDNOparentcompanyandCustomerscouldalsovary.This

wouldinprincipledependontheappropriatecostandbenefit

splittobecarriedbytheDNOparentcompanyandCustomers.

This prompts two immediate questions, firstly what is the

appropriateriskandcorrespondingcostandbenefitthatshould

bepassedtoConsumersforboththesecuritybenefitandthe

ancillary services. And secondly, how this cost (and annual

revenues in particular) would be transferred to Customers

throughpricecontrolsorotherwise.Wediscusstheprinciplesof

theregulatorytreatment,ratherthanseekingtoconsiderhow

thismayalignwiththeemergingRIIO-ED1arrangements12.

Non-regulated&ThirdPartyFinancedScenario

In the third party financedmodels there are few regulatory

compatibilityconcernsasallcommercialriskispassedtothe

third party. The DNO parent company would see the tolling

chargeasanoperationalcostwhichwouldbetreatedlikeany

otherunderthepricecontrol.

Wenotethatthereisavariantunderwhichthethirdpartyisinfact

thenon-regulatedbusinessarmoftheDNOparentcompany.This

wouldnotprecludetheDNOparentcompanyfrommakinguseof

anaggregatortomanagetheoptimisationoftheadditionalvalue

streams. Under this variant the DNO’s non-regulated business

buildsandoperatestheasset.CustomersandtheDNO’sregulated

businesstakenoriskontheasset,assumingthatafixedtransfer

priceisagreedforthevalueoftheavoidedreinforcement.The

DNO’s non-regulated business takes the risk and value of the

additionalvaluestreams(variantofDNOMerchant),orcontracts

thecapacityouttoathirdparty(VariantofDNOContracted).

DNOParentCompanyRegulatedFinancedScenario

Inthesecondpossiblescenariowheretheassetisaregulated

12 http://www.ofgem.gov.uk/Networks/ElecDist/PriceCntrls/riio-ed1/consultations/Documents1/RIIOED1DecOverview.pdf


longtermagreementsfortheeconomicoutputofspecificassets

(suchasthetollingcontractbetweenthethirdpartyandthe

DNO)mayincertaincircumstancesbecategorisedasaleasing

agreement.Ifso,theliabilitiesoftheassetmayconsolidateon

thebalancesheetofthepartywhoissuesdispatchinstructions

for the asset and carries the commercial risk, irrespective of

whohasownershiporoperationalcontroloftheasset.Similarly,

there may be an adverse effect on the credit rating of the

projectpartywhotakestheassetandsubsequentcommercial

risk,againirrespectiveofwhocontrolsorfinancestheasset.

The accounting treatment of the project’s liabilities on the

balancesheetsoftherespectiveprojectpartieswoulddepend

onboththemagnitudeof,alongwithtowhomtheproject’s

assetsandliabilitiesaccruedto.Thisassetliabilitysplitwould

beexaminedinthediscountedcashflowofthespecificproject

and would be tested according to the Financial Reporting

Standard (FRS) 5. Careful consideration of the commercial

structure would be required to ensure that the accounting

treatmentalignswiththeintentionsofthebusinessmodel.

2.5.ConsultationQuestions

The questions that we invite interested parties to submit

writtenevidenceandanalysisonareasfollows:

1.Do you agree with the range of business models

presentedinSection2.1?

a.Arethesebusinessmodelsandtheirvariantsrepresentative

oftherangeofplausiblebusinessmodels?

b.Doyouagreewiththecharacterisationsofeachofthese

businessmodelsintheirrespectiveSections2.1.1through

toSection2.1.5?

2.Doyouagreewiththechoiceofassessmentcriteriaas

describedinSection2.1?

a.Are these thekeyassessment criteria that thebusiness

modelsshouldbecomparedagainst?

Regulated & Split Customers DNO Parent Company

FinancedScenario

The fourth financing option is for the asset to again be a

regulatedasset,butitisnowsplitfinancedbetweentheDNO

parentcompanyandCustomers. In thiscase theDNOparent

company,ifithasoperationalcontroloftheasset,isincentivised

tooperate theasset inaprofitoptimizingmanner (theDNO

parentcompanywouldaccruesomeof thefinancialbenefits

fromtheancillaryservices,thusavoidingmisplacedincentives).

Customerswouldagainseethesecuritybenefit,butthefinancial

benefitsoftheancillaryrevenuestreamswouldnowbesplit

betweentheDNOparentcompanyandCustomers.

Acrossalltheregulatedcasesthekeyistoalignthecostsand

risksthatCustomersaretakingwiththebenefits.Itisclearthat

theDNO should take someexposure to theadditional costs

andrevenuesassociatedwiththeproject,buttheremaybean

argumentforCustomerstotakingmoreorlessofthebenefits

andcorrespondingrisk.

We also note that storage assets do not have the typical long

economiclifeofdistributionassets.Moststorageassets,suchas

batteries,wouldhavealifetimeof10-15yearsfortheenergystorage

mediumand30yearsforthebalanceofplant.Incomparison,most

conventionaldistributionequipmenthasalifetimeof30-50years.

Storageassetsalsodifferfromtypicaldistributionassetsasthey

providepositiverevenuesthroughoutthelifeoftheasset.Therefore

treatmentasatypicalregulatedasset(withreturnsmadeovera45

yearassumedassetlife)maynotbewellalignedwiththetrue

costsandrevenuestotheDNOofowningandoperatingstorage.

2.4.AccountingTreatment

Anadditionalprojectdynamicthatvariesbetweenthedifferent

businessmodelsistheaccountingtreatmentoftheassetand

itsliabilitiesonthebalancesheetsoftheprojectparties.Most

notably,underaccountingrulesasoutlinedintheInternational

FinancialStandardsasadoptedintheEuropeanUnion(EU-IFRS),


b.Are there any additional advantages, disadvantages or

barrierstoanyoftheindividualbusinessmodelsthathave

notbeenincluded?

3.Doyouagreewith thechoiceof the two leadmodels

andthereasonsforselectingtheseasdescribedinthe

QualitativeAssessmentSummary,Section2.2?

a.Are these options the most likely to be suitable for

distribution-connected storage, or should one or more

oftheotherbusinessmodels(orvariantsofthese)have

beenconsideredovertheDNOContractedandContracted

Servicesmodels?

b.Ifso,whichmodelsshouldhavebeenconsideredandwhy?

4.Doyouagreewiththerangeoffourregulatorytreatment

scenariospresentedinSection2.3?

a.Arethereotherscenariosthatshouldbeincluded?

b.Arethereinherentregulatorycreatedlimitationsthathave

notbeendiscussed?

c.Arethereadditionaladvantagesordisadvantagesforeach

ofthesescenariosthatshouldbeconsidered?


Lead Business Models3


Projectdevelopment

The trigger for pre-development of the project is the

identificationofaneedforreinforcementinaspecificlocation

tosupportthelocalnetwork.TheDNOwouldinvestigatethe

rangeofoptionsavailabletoit,whichmightincludeinstallation

of new transformers and or lines, uprating of capacity of

current infrastructure through incremental investments, and

of non-transmission options such as EES and/or demand

side response. If the DNO identifies EES as being the most

technicallyandeconomicallyfeasibleoptionthatisdeliverable

inthetimescalesrequired,itwillinitiatepre-developmentof

thestoragefacility.

At an early stage in the development process, the DNO

would initiate discussionswith potential third parties to test

themarketforthestoragetollingagreement.Assumingthat

the discussions indicate sufficient value in potential tolling

payments,theDNOwouldselectathirdpartyandagreeterms.

Thismay involvebilateralnegotiation,or amore formalised

auctionprocess.

Theassetwouldbebuiltasaregulatedassetundertherules

ofthepricecontrolinforceatthatpointintime.Theoptions

for regulatory treatment are discussed in general terms in

section2.3.

Inthissectionweoutlinetwoproposedleadbusinessmodelsto

beassessedwithintheSNSprojectinmoredetail.Weconsider

the project lifecycle, contractual arrangements, and possible

barriers.Finally,weconsidersomepossiblevariantsofthemodels.

TheDNOContracted and Contracted Servicesmodels share a

numberofcommonfeatures.Theybothrequirea thirdparty

to takesomeorallof thecommercial riskon the long term

valueoftheadditionalvaluestreams,andrequireacontractthat

ensuresthattheDNOhasprimaryaccesstotheassetwhenitis

requiredforsecuritypurposes,orotherconstraintmanagement.

Thekeydifferencebetweenthemodelsistheownershipand

operationof theasset,which lieswith theDNOunderDNO

ContractedandthethirdpartyunderContractedServices.

3.1.DNOContracted

TheDNOContractedmodeltreatsthestorageasadistribution

system asset and the development approach may in many

ways be analogous to a traditional DNO investment, albeit

withthefurthercomplexityofadditionalvaluestreamsanda

contractualrelationshipwithathirdparty.

Thismodelisclosesttothatwhichwillbedemonstratedwithin

theSNSprojectfollowingcommissioningofthestoragefacility.


Contractingstructure

Figure 11 shows a possible ownership and contracting

structurefortheDNOContractedmodel.TheDNOwouldlead

construction of the asset, contracting out for construction as

required.TheassetwouldbeownedbytheDNOasaregulated

distributionasset.Onceoperational,themaintenancestrategy

ofthefacilitywouldfalltotheDNO’sNetworkOperations.The

actual maintenance may be carried out by the DNO’s field

engineersdirectly,orcontractedouttothemanufacturerora

maintenanceprovider.

Figure 11 DNO Contracted: Contractual Structures

Manufacturer ManufacturerO&M Provider/DNO Function

Insurer

Asset

DNO

Ofgem

Third Party

AdditionalInsurance

O&MAgreement

AssetOwnership

LongTermServicesAgreement

EPCorTurnkeyContract

CapacityOfftakeAgreement

DNOLicence AdditionalServices


thatensurethestorageisatorabovesomepre-agreedcharging

levelatthestartofeachsecuritywindow(i.e.thethirdparty

mustensurethatthestoragedeviceisattheagreedcharging

levelatthestartofeachsecuritywindow).

DependingonwhethertheDNOhassecuredafixedcapacity

tollingcontractfortheentireeconomic lifetimeoftheasset,

orwhetherarevenuesharearrangement isestablished, the

DNOmaynotholdanypriceormarketriskassociatedwiththe

valueoftheadditionalservices.Ineithercase,theremaybe

somesmallexposuretothenetcostofpowerboughtandsold

intheDNO’suse(e.g.ifafaultwastooccurandtheassetwas

requiredtoexport).

TheDNOretainscounterpartyriskonthepossibilityofthethird

partydefaulting.TheDNOmayrequiresomeformofcollateral,

securityorguaranteeasacontingency.Therisk for theDNO

isthatitisleftwithnoroutetomarket,andwithnorevenue

security.Weassumethat thecounterparty risk that thethird

party takes isnotamajorconsiderationdue to the typically

highcreditworthinessofDNOs.

Note that it is assumed that no Connecting Construction

Agreement (section 16 agreement, controlling the terms

underwhichgeneratorordemandconnectstothenetwork)is

required,becausetheconnectionoftheassettothenetwork

isunder the full controlof theDNO,andoperational control

remainswiththeDNOsothereisnotaneedtospecifythisina

connectionagreement.

ThecapacityofftakeagreementbetweentheDNOandthethird

partydefinesthetermsunderwhichthecapacityisreleasedto

thethirdparty.Theagreementgivesthethirdpartycommercial

control(tolling)oftheassetoutsideofaspecifiedsetofSecure

Capacity Windows. Under the simplest arrangement, these

windowswouldbefixed in advance for thedurationof the

contract(e.g.4-7pmonNovember–Marchweekdays).Asthe

requirementsmightchangeacrossthelifeoftheprojectthe

DNOwouldneedtobeconservativeindefiningthesewindows,

whichmayreducethevaluetothethirdpartyandtherefore

thepaymenttotheDNO.Insection3.3weexploretheoptions

tobuildflexibilityintothiscontract.Aconditionofthecontract

wouldbethethirdpartyisobligedtomaketollinginstructions


TypicalContractTerms

Whileany tolling contractbetween theDNOand thirdparty

would be project specific, Table 10 gives a summary and

examplesoftheexpectedheadlinecontracttermsthatcould

Terms

SecureCapacityWindow

ContractTenure

GenerationCapacity

OperationalConstraints

DispatchNotice

TollingCharge

Non-performancePenalties

Table 10 Example DNO Contract Tolling Agreement Headline Terms

Description

• FixedtimesthattheassetisrequiredfortheDNO’ssecurity/constraintmanagement

purposes(e.g.4-7pmdailyfromOctobertoMarch)

• Contracttenure(years)

• Thestoredenergy(MWh)thatisrequiredatthestartofaSecureCapacityWindow

• The operational constraints outside of Secure Capacity Windows (e.g. depth of

discharge(%ofbeginningoflifecapacity,responsetime(s))

• Thirdparty’sMWhavailability/holdingrequirementforagivenperiodforancillaryservices

• Thirdparty’sMWhdischargeforagivenperiodifDNOisinstructedbythethirdparty

todispatchtheassetforancillaryservices

• An annual fixed payment from the DNO to the third party (in £ or in £/MWh

ofavailability)

• A£/hrpaymentfromtheDNOtothethirdpartyifthethirdparty’sdispatchnotices

foravailabilityand/ordischargearenotmet

• A£/hrpaymentfromthethirdpartytotheDNOifthethirdpartyissuesdispatch

noticeswhichareinconsistentwiththeSecureCapacityWindowrequirements(e.g.

notenoughenergystoredatstartofaSecureCapacityWindow)

define the commercial elements of any tolling contract. A

summaryofhow these termscould changeas thebusiness

models evolve and the technology becomes established is

giveninSection3.4.


on the operation of generation by DNOs is a potential

barrier as generation and distribution licenses are taken to

beincompatible.

As network companies, DNOs are prevented from owning

generationlicencesundertheEuropeanThirdEnergyPackage.

However, under theGB regime therearegeneration licence

exemptions for small generators in the Electricity Act.

EffectivelytheseallowaDNOtoowngenerationupto10MWin

capacity.Storageprovidingdistributionnetworkreinforcement

cantypicallybesizedbelowthislimit,andthereforethisneed

notbeabarrier,assumingthatthelimitismaintainedatthe

10MWlevel.

Howevertheneedforstoragetobeoperatedwithageneration

licencehasnot beenvalidated. If storagewasexempt from

beingconsideredgenerationandholdingagenerationlicence

thiswouldsimplifyarrangements.

De Minimis Business restrictions

DNOs are restricted from conducting activity outside of

distributionsubjecttoademinimisthreshold14ofboth;

• 2.5percentofthetotalturnoverofdistributionbusiness:and

• 2.5percentof thesumof the licensee’ssharecapital, its

sharepremium,anditsconsolidatedreserves.

IntheDNOContractedmodelweassumethatstoragedoesfall

undertheregulateddistributionbusiness.However,ifthiswere

notthecasethentheamountofancillaryrevenuegenerated

fromstorageownedbytheDNOscouldbesubjecttoacap.This

wouldnotbeanissueuntillargenumbersofstorageprojects

hadbeendeveloped.

Restrictions on DNOs buying and selling electricity

UndertheDNOContractedmodel,theDNOuseofthestorage

Decommissioningandterminalvalue

Throughout the project life, the DNO would regularly assess

whether storage remains capable of fulfilling the security

requirement, taking account of the useable capacity of the

storage(afteranydegradation)andforecastdemandgrowth.

AtthepointatwhichtheDNOforecaststhatthestoragewillno

longerbesufficienttofulfilthesecurityrequirement,theDNO

willconsiderwhethertoreneworaddtothestoragecapacity

or to make alternative interventions including traditional

reinforcement.IftheDNOfollowsthetraditionalreinforcement

option and the storage still has some usable economic life,

itmaybeeconomicfortheDNOtore-deploythestoragein

another location. This could only occur within the contract

durationiftheoptiontorelocate,alongwithanycompensation

forthetemporarylossofavailability,waswrittenintotheterms

ofthetollingcontract.

Outsideofthecontractterm,theDNOcouldrelocateandaimto

extendthecontractwiththethirdparty.

PotentialbarrierstotheDNOContractedmodel

There are potential regulatory barriers to the ownership or

operation of storage by DNOs, which we explore below.

Workstream 6 (WS6) of the Smart Grids Forum (SGF)13 is

consideringtheseandUKPowerNetworkswillconductfurther

workinthisareainthecourseoftheSNSproject.Herewegive

anoverviewofourcurrentunderstanding.

Generation licence

Storage is distinct fromgeneration in that it both consumes

and releases electricity (with the consumed amount being

slightlyhigherduetolossesinthechargingcycle).However,

storagesuchastheexistinglargescaletransmissionconnected

pumpedstorageatDinorwigandFfestinioghasbeentreated

asgenerationforregulatorypurposes.Assuch,theoperators

currently hold generation licenses. Therefore, the restriction

13 DNO ownership and operation of storage facilities.pdf (unpublished)14 Ref to Standard Licence Condition 29


technologywouldexpect todegradepast its functional limit

in10-12years,butthebalanceofplantcouldoperatefor30

yearsbeforereplacement.Accordingly,theDNOmaywishto

replacethestoragetechnologytofurthertheoperationallifeof

theasset.Alternatively,theyDNOmaylooktoselltheassetto

anotherpartywhowouldsimilarlyusetheintrinsicassetvalue

toextenditsoperationallifetime.

If the asset and balance of plant have no further intrinsic

operational value, thematerials (and in the caseof EES the

storage medium in particular) and land would still have an

intrinsicvaluethattheDNOcouldmonetizethroughthesaleof

theassetandproperty.

Other barriers

As we have already noted, the model may require complex

contractualtermsinordertoensuremostvaluableservicesare

dispatchedwhilstensuringsystemsecurity.Thiscomplexitymay

alsolimitthepotentialforaggregationofthestorageintoathird

party’sportfolioofflexibleassets,whichmaylimitthevalueof

theassettothethirdpartyorincreasethecostofmanagingit.

Thefuturevalueoftheadditionalrevenuestreamsisinherently

uncertain.Underafixedannualpayment,allmarketrisksits

withthethirdparty,whichmayleadathirdpartytoheavily

discountthefuturevalueoftherevenueswhenevaluatinga

potentialopportunity.OthervariantswheretheDNOsharesthe

riskmayleadtolowerdiscounts.

mayconsumeandgeneratesmallamountsofelectricity.Whilst

theremaybesomerestrictiononthedirecttradingofelectricity

byDNO’s,itseemsthatasupplyagreementwiththethirdparty

wouldcircumventthispotentialissue.

Disposal of assets

When the asset (storage technology and balance of

plant) reaches itsendof life theDNOwouldbe required to

decommissionanddisposeoftheassetaccordingtocondition

26 “Disposal of Relevant Assets” of the Distribution License

Conditions. This condition states that, in general, consent

is given for assetswithoutgivingpriornotice if theasset is

obsolete, redundant or the disposal will not constitute the

disposalofalegal(ratherthananequitable)interestuntilthe

assetisobsoleteorredundant.Howeverthisdoesnotapplyif;

• Thevalueexceeds£200kinanyregulatoryyear

• It does not apply in respect of a relevant asset that is

obsolete,unlessanappropriatereplacementoralternative

arrangementhasbeeninstalled

• It does not apply if the disposal of the relevant asset

constitutesasaleandleasebackarrangement

• Itdoesnotapplyiftherelevantassetisintendedtoremainin

operationalusebutnotundertheoperationalcontrolofthe

licenseeanditsvalueexceeds£20,000

Attheendoflifeitisexpectedthattheassetwouldstillhave

some intrinsic operational value. For example, the storage


Box 3 SSE Orkney project

TheprimarypurposeoftheSSEOrkneyProjectistomanagetheintermittentgenerationofrenewablesonOrkney.Thiswillbe

runaspartoftheOrkneyActiveNetworkManagement,underwhichSSEPDhastheabilitytocurtailrenewablegenerationif

required.Inthiscase,theuseofstorageisanalternativetocurtailmentofrenewablesandthereforenotspecificallyrequired

fornetworksecurity.

SSEhasprocuredstoragethroughatenderprocess,underwhichSSEwillbuystorageservicesfromthirdpartyowner(s)/

operator(s)underpre-agreedterms(analogoustotheContractedServicesmodel).Whilethereisafundamentalconditionto

makestorageavailabletoreceivesurpluspower,afundamentaldifferencetotheUKPowerNetworksSNSprojectisthatthe

Orkneystoragedeviceisnotrequiredfordistributionnetworksecuritypurposes.

15 http://www.ssepd.co.uk/HaveYourSay/Innovation/Portfolio/OrkneyPhase1/

3.2.ContractedServices

TheContractedServicesmodeltreatsthestorageasaservice

providertotheDNO.Itsharessomesimilaritywiththetreatment

foranembeddedgenerator,withtheadditionoftheprovisionof

capacitytotheDNOwithinspecificwindows.Anotherexample

ofaDNO-connectedstoragewithadifferentprimarypurpose

toUKPowerNetworks’SNSprojectisSSEPowerDistribution’s

OrkneyEnergyStorageproject15describedinBox3.


CapacityOfftakeAgreement

Alternatively,theDNOcouldissueatechnologyneutraltender,

open to DSR or embedded generation, however it may be

difficulttocompareavailabilityonanequalfooting.Thismight

reducethecostof theservicetotheDNO,however itcould

lead to less innovative,highercarbonemitting technologies

beingselected.

Contractingstructure

Figure12showsapossibleownershipandcontractingstructure

fortheContractedServicesmodel.Thestoragewouldbeowned

bythethirdparty,eitherdirectlyorthroughaSpecialPurpose

Vehicle(SPV).

Projectdevelopment

AsfortheDNOContractedmodel,thetriggerforpre-development

oftheprojectistheidentificationofaneedforreinforcementina

specificlocationtosupportthelocalnetwork.IftheDNOidentifies

EES as being the most technically and economically feasible

optionitwillinitiateatenderforthestoragefacility.

Thistenderwilldefinetheparametersoftheservicerequired

bytheDNO,includingtheexportingcapacity(MW)andstorage

capacity(MWh),aswellasthesecuritywindowsinwhichthe

assetisrequiredbytheDNO.Itwouldalsoidentifythesitefor

theassetandthestartandenddateofthesecurityrequirement.

Figure 12 Contracted Services: Contractual Structures

Manufacturer Manufacturer O&M Provider

Insurer

Project SPV:Storage Co DNO

Third Party

Insurance

O&MAgreement

Ownership

LongTermServicesAgreement

EPCorTurnkeyContract

DNOConnectionAgreement

AdditionalServices


butcounteredbytheincentivesofmorethanonebidderwho

maybeparticipatinginacontractorauction.

TheDNOisexposedtocounterpartyrisk–inthiscase,therisk

that the third party fails to provide the contracted services

(eitherforisolatedinstances,orinthelongterm).Inthiscase

itisunlikelythattheDNOwillbeabletoreplacethesecurity

providedbytheassetatshortnoticeandmightincursignificant

costsindoingso.Thisisnotaseasilymitigatedwithfinancial

guaranteesasfortheDNOContractedmodelwheretheDNO

isonlyexposedtoafinancialloss.Thismaybeabarriertothe

ContractedServicesmodel,asdiscussedlaterinthissection.

Thethirdpartyislessexposedtocounterpartyriskbecauseof

thetypicalDNOcreditworthinessandthelimitedalternatives

availabletotheDNOintermsofsecurity.

TypicalContractTerms

AsintheDNOContractedmodel,anytollingcontractbetween

theDNOandthirdpartyinaContractedServicesmodelwould

beprojectspecific.However,Table11givesasummaryand

examplesof theexpectedheadline contract terms.Again, a

summaryofhow these termscould changeas thebusiness

models evolve and the technology becomes established is

giveninSection3.4.

To ensure that asset is built in a location that both meets

theDNO’s reinforcement requirementsandcanbeoptimally

incorporated into the existing network, the DNO would

identifythesiteonwhichthestorageistobebuilt.However,

whilsttheDNOwouldidentifythesitethethirdpartywould

beresponsible forsecuringthenecessaryplanningconsents.

Weassumethatthethirdpartywouldberequiredtosigna

connection agreement with DNO. This would most likely be

based on a generation connection agreement, albeit with

someadjustments to reflect theparticular technical features

ofstorage.

ThecapacityofftakeagreementbetweentheDNOandthethird

partywouldhaveatermequaltotheanticipatedeconomiclife

oftheasset.Itdefinesthewindowsinwhichthestoragewill

beheldavailableandatsomeminimumchargelevel.Asthe

requirementsmightchangeacrossthelifeoftheprojectthe

DNOwouldneedtobeconservativeindefiningthesewindows,

whichmay increase the cost to theDNO. In section3.3we

exploretheoptionstobuildflexibilityintothiscontract.

In this model the DNO does not take any market risk. This

remainswiththethirdparty.Thethirdpartyalsotakesriskon

the availability and operational performance of the storage.

ThisislikelytobereflectedinlowertermsofferedtotheDNO,


Table 11 Example Contracted Services Tolling Agreement Headline Terms

Terms


ContractTenure

GenerationCapacity


ThirdPartyDispatchNotice

TollingCharge

Non-performancepenalties

Description

• FixedtimesthattheassetisrequiredfortheDNO’ssecurity/constraintmanagement

purposes(e.g.4-7pmdailyfromOctobertoMarch).

• Contracttenure(years).

• Thestoredenergy(MWh)thatisrequiredatthestartofaSecureCapacityWindow.

• The operational constraints during these Secure Capacity Windows (e.g. depth of

discharge(%ofbeginningoflifecapacity,responsetime(s)).

• MWhdischargeforagivenperiodifassetisrequiredtodischargeforSoSobligations.

• Anannualfixedpayment from theDNO to the thirdparty (in£or in£/MWhof

availability).

• A£/hrpaymentfromthethirdpartytotheDNOifthecontractedSecureCapacity

Windowrequirementsarenotmet.

• A£/MWhpaymentfromthethirdpartytotheDNOiftheinstructeddischargecapacity

isnotmet.


which can all provide the same service. This is not true in

thecaseofdistribution-connectedEESemployeddirectly for

networksecurity.Apotentialoutcomeofaprobabilitybased

approachmightbetoresultintheDNOcontractingforagreater

capacityofstorage than is required,withoutaddressing the

underlyingissue.

Complexity

As we have already noted, the model may require complex

contractual terms inorder toensuremostvaluable servicesare

dispatchedwhilstensuringsystemsecurity.Thelevelofcomplexity

islikelytobesimilartotheDNOContractedmodel.Thiscomplexity

mayalsolimitthepotentialforaggregationofthestorageintoa

thirdparty’sportfolioofflexibleassets,whichmaylimitthevalue

oftheassettothethirdpartyorincreasethecostofmanagingit.

ItmaybemoredifficultfortheDNOtomakeuseofadditional

benefits, such as power quality control and power factor

correctionimprovementsfromthepowerelectronicsofsuch

storagetechnologiesthathavethecapability.

3.3.Variants

Therearemanypotentialvariationsonthetwoleadmodels

described above.Herewe consider variantswhich separate

ownershipfromoperation,variations incontract length,and

variantsontheregulatorytreatmentofuncertainty.

Ownershipvariants

Inthemodelsdescribedaboveweassumethattheownerof

theassetisalsotheoperator.Byseparatingthesetwofunctions

wegeneratetwonewvariants,showninFigure13.

• Third party operator leased from DNO. Under this

model, the DNO builds the storage facility and then

leases ittothethirdpartytooperate.Comparedtothe

DNOContractedmodelthisremovescontroloftheasset

fromtheDNO.ComparedtoaContractedServicesmodel

Decommissioningandterminalvalue

Under the Contracted Services model, the DNO may have

littleornooptionalityonrelocatingtheasset if thesecurity

requirement canno longerbemetby theasset.Given the

capitalcommittedbythethirdparty, it is likely torequirea

longtermcontractwithafixedpriceforthesecurityprovided.

Thethirdpartyisunlikelytorelocatetheassetunlessthisis

aspecificterminthecontractwhichcompensatedthethird

partyfortheadditionalcostsandrisksincurredinthisprocess

aswellasthelossofrevenue.

PotentialbarrierstotheContractedServicesmodel

ApossiblebarriertotheContractServicesmodelisthelackof

directoperationalcontrolofthestoragebytheDNO.Thisisa

specificissueforEESrequiredfornetworksecuritypurposes,

andmaynotbeanissueforotherusesofEESoutsideofthe

scopeofthisConsultation.

Oneoptionistoputfinancialpenaltiesonthethirdpartyifthe

storageisnotmadeavailableintherequiredwindowswith

thestoragecharged toapre-specified level. The issuewith

thisisthattheimpactontheDNOofnotmeetingasecurity

requirementisnoteasilyquantifiable,andanycompensation

value couldbe sohigh that itmaterially reduced thevalue

thatathirdpartyassignstothecontract.Anotheroptionmay

be for theDNOtohavedirectoverridecontrolof theasset;

howeverthisimpliesalevelofoversightandinterventionthat

is unusual in tolling contracts. Individual DNOs will need to

evaluatetheirownattitudetothisrisk.

TheprobabilitythatDNOsshouldassigntotheavailabilityof

embeddedgenerationiscurrentlydefinedinERP2/6.Areview

ofthesestandardshasbeencitedasawayofaddressingthe

issue for storage, and SNS is expecting to deliver learning

specificallyrelatingtothecontributionofstoragetosecurity

ofsupply.Aprobabilitybasedapproachtotheavailabilityof

capacityissuitablewheretherearemultiplesmallresources


ContractedServices

itmaintains theDNO’spotential costof capitalbenefit.

However disposal of an operational asset by a DNO to

another operator requires regulatory approval (Section

3.1).

Third PartyLeased from

DNO

DNO Leasedfrom Third

Party

DNOContracted

Third Party Owner

Third Party OperatorThird Party Owner

Third Party Operator

Figure 13 DNO and Third Party Ownership Variants

• DNOoperatorleasedfromthirdparty.Underthismodel,

thethirdpartybuildsthestoragefacilityandthenleases

it to theDNO tooperate. Thismodelmight beofmerit

inacasewheretheDNOrequiredoperationalcontrolof

theassetforsecuritypurposes,butisrestrictedfromasset

ownershipe.g.byregulatoryrestrictions.


agreeinadvancethetermsofthecontractedSecureCapacity

Windowsandtollingcontractterms.Asummaryoftheexpected

headlinecontracttermsforthetwoleadbusinessmodelswas

giveninSections3.1&3.2.

TheSecureCapacityWindowswoulddefinethecapacityandtime

thatwouldberequiredbytheDNO,withthebalanceofcapacityand

timebeingavailabletothethirdpartytousetheasset.Anincrease

intimeavailabletothethirdpartyislikelytoincreasethevalueof

additionalvaluestreams,althoughthisincreasemaybesmall if

overridingrestrictionsremain.Toensureminimalvaluelossestothe

ancillaryservicesthecontractedSecureCapacityWindowsshould

be carefully considered, balancing security of supply obligations

whilemaximisingtheasset’sutilisation.Forexample,onemodel

isthattheDNOhasarightofoverrideatanypoint,toensurethat

securityofsupplyrequirementsaremet.However,thisavailability

uncertaintyforthethirdpartyisanadditionalriskthatcouldresult

inincreasedcommercialriskpremiums.

Similarly,ifthethirdpartyhadoperationalcontroloftheasset

theDNOcouldincludeinthecontractsetdiurnalSecureCapacity

Windowsthataretobeavailableforthedurationofthecontract.

Thisoptionwould,however,likelyresultinsignificantancillary

servicevaluelossesastheseSecureCapacityWindowswould

beunavoidablyconservative,aswasshownfortheexample

termsoftheleadbusinessmodelsinSections3.1&3.2.

Anoptimal scenario that couldevolveasallpartiesbecome

morefamiliarwiththeassetsoperationwouldbeadynamic

reporting of the asset’s Secure Capacity Windows at a day

and month ahead frequency coupled with aggregation of

thestorageassetsintoawiderportfolio.Thiswouldlimitthe

conservatism in the Secure Capacity Windows and ensure

minimal ancillary revenue stream value losses. A summary

ofthealternativestothetollingcontracttermsaspenetration

of EES assets increase, technology learning increaseand the

businessmodelsevolveisgiveninTable12.

Lengthofcontract

Fortheleadmodelswehaveassumedthatthetollingcontractis

fortheexpectedeconomiclifeofthestoragefacility.FortheDNO

Contractedmodel,theimpactofshorteningthecontractistogive

theDNOmoreflexibilitye.g.ifdemandgrowthisdifferentfrom

forecast.However,thisleavestheDNOwiththeneedtorenegotiate

thecontractatafuturepointintimewhentheexpectedvalueof

the additional revenue streams may be significantly higher or

lower.Ashortercontractmayalsoberequiredifthirdpartiesare

unwillingtotakeonalongtermposition.

UndertheContractedServicesmodel,ashortercontractwould

beharder toenactbecause thiswould leave the thirdparty

withtheriskofastrandedassetinfuture.Alsooncetheasset

isinplacetheDNOistheonlyCustomers,thereforethethird

partywillwishtolockinthefullvalueofthesecuritypayments

upfrontbeforetakingafinalinvestmentdecisionontheasset.

Shortercontractperiodsmayentailhigherpaymentsfromthe

DNOacrosstheshortercontract.

SharingofbenefitsandriskswithCustomers

Ifstorageistreatedasanyotherdistributionasset,thesharing

oftherisksandrewardsoftheassetwouldbesharedaccording

to the price control. However if the risks of storage were

consideredtobemateriallydifferentfromthosefortraditional

distributionassets(e.g.duetotechnologyrisk,ormarketrisk

if theDNO isunable to securea long termcontract) thena

separateuncertaintymechanismcouldbeproposed.Underthis

approach,theDNOandOfgemwouldagreeaspecificstorage

uncertaintymechanism in theDNO’spricecontrol.Thiscould

bespecificallyfocusedonthedriversofstoragevaluethatare

beyondtheDNO’scontrol.

3.4.SecureCapacityWindows&TollingContractTerms

If theasset is tobecommerciallycontrolledbyathirdparty

(withorwithoutoperationalcontrol),asisthecaseinthetwo

leadbusinessmodels, theDNOand thirdpartywillneed to


Table 12 Alternatives to Tolling Agreement Headline Terms

Terms


ContractTenure

GenerationCapacity


ThirdPartyDispatchNotice

TollingCharges

Non-performancePenalties

TollingContractTermAlternatives

TheSecureCapacityWindowsareexpectedtobeconservativetobeginwith,butwithincreased

projectlearningwouldexpecttoberationalisedovertime,forexample;

• Aminimumavailabilitycouldbecontractedinadvanceasabaseline.Thiscouldbereviewedat

setperiodsovertheprojectcontractandrationalisedwherepossibleasallpartiesbecomemore

familiarwiththeasset’soperation.

• Forecasting and contracting of security requirements could increase in frequency to month,

or week ahead reporting, allowing more active real-time control of the asset and avoiding

unnecessaryvaluelosses.

• TheSecureCapacityWindowcouldbeaggregatedforanumberofassetsacrossawiderstorage

portfolio.

• Contracttenuresarepresentlylimitedbytechnologylife-times,butasEEStechnologiesimprove

sowouldthecontracttenuresexpecttoincrease.

• Shorttermrollingcontractsmaybefavourabletosomethirdparties.

• Longtermcontractsacrossportfolioswithassetreplacementsattheendoftheirtechnological

life-timewouldalsobepossibleandmayalsobefavourabletootherthirdparties.

• AswiththeSecureCapacityWindows,thereportingofthegenerationcapacitycouldbecome

more dynamic and reported closer to real time as project party’s familiarity of the asset’s

operationimproves.

• Portfolioeffectswouldincreasetheassetsredundancywhendischargingforancillaryservices

(i.e.risksofnotmeetingtheirobligationwouldbemitigatedbycapacityfromotheravailableassets).

• Theavailablegenerationcapacitywouldincrease,reducingancillaryrevenuestreamvaluelosses.

• Astechnologyperformanceimproves,sowouldoperationalconstraintsbeexpectedtobelessonerous.

• Againincreasingtheavailablegenerationcapacityandreducingancillaryrevenuestreamvaluelosses.

• Withincreasedprojectlearningandeconomiesofscale(asotherassetsarebuilt)dispatchwould

beagreedclosertorealtimetoavoidancillaryservicevaluelosses.

• Withportfoliosdispatchnoticescouldbecomeareaspecific,allowingmultiplestorageassetsto

dischargeinpart(orinfull)inunison.

Alternativetollingchargearrangementscouldbeused,suchas;

• Apaymentindexedtotheasset’sancillaryrevenues.

• Discountsontherevenuesofrespectiveancillaryrevenuestreams(withorwithoutafloorprice).

• Thesenon-performancepenaltieswouldnotbeexpectedtochangeovertheprojectcontract.

• However, security of supply forecasting and technology performance would be expected to

improve,alongwiththethirdparty’sabilitytoeffectivelymanagetheancillaryservices,thus

decreasingthelikelihoodofthethirdpartyaccruingthesepunitivecharges.


ArelativeadvantageoftheDNOContractedmodelmaybealower

costofcapital.Howeverthismaynotpersistasthedeploymentof

storageincreases.TheDNOContractedoptioncreatesthepossibility

ofsharingtheriskandadditionalbenefitswithCustomers(ifthis

wasconsidereddesirableforCustomers),whereasthisisnoteasily

possiblewiththeContractedServicesmodel.

Widerbenefits

Underbothmodels,thetollingcontractwillclearlyspecifythe

termsonwhichcapacityismadeavailable.Thereisariskthat

thefullbenefitsfortheGBsystemarenotcapturedduetoa

lackofflexibilityintheseterms.

TheDNO contractedmodel allows for competitionbetween

thirdpartiesintheprovisionoftradingandaggregation.The

third party has the ability to transfer the knowledge and

expertise to develop projects in other DNO licence areas.

UndertheContractedServicesmodel,thethirdpartycanalso

transfer experience of building and operating storage, and

thereisthepotentialforfurthercompetitionintheprovision

oftheseservices.

Under both models, the DNO is able to set terms of the

technologyconsidered,toensurethatalowcarbonsolutionis

procured.ItislikelythataContractedServicesmodelwouldbe

lessprescriptiveintermsoftechnologychoice.

Futureproof

Both models allow for aggregation across multiple assets

bythethirdparty;howeverthisislimitedbytherestrictions

imposed by the Secure Capacity Windows. A large number

ofstorageassetswithsimilarSecureCapacityWindowscould

beeasilyaggregated,butifthesearealldistinctthismaybe

less effective. The Contracted Services model may be more

scalable,withonethirdpartyabletooperateandaggregate

storageacrossmultipleDNOlicenceareas.

3.5.Comparison&Conclusions

Our review of the two lead business models presented for

consultation suggests that both the DNO Contracted and

ContractedServicesmodelsareviablemodelsfordistribution-

connectedstorage. Inthehighlevelqualitativescoring,both

modelsperformrelativelywellagainstmostcriteria.

Thetwoleadmodelsarecomparedbelowunderthegroupings

fromthequalitativeassessmentcriteria.

Security

TheDNOContractedmodelgivestheDNOrobustconfidence

inavailabilitythroughdirectcontrolovertheoperationofthe

storage,andcanensurethatthethirdpartyinstructionsdo

notcompromisetheuseofthestoragetomanagenetwork

constraints when required (assuming that the contractual

obligationsdonotpreventthis).Thesecurityprovidedbythe

ContractedServicesmodel isdependentonthecontractual

obligationsplacedonthethirdparty,andhowitmeetsthose

obligations.Whilstnotasdirectasoperationalcontrol, this

modelcouldprovidesufficientsecurityifthetermsarewell

structured.

Assetvalue,costandrisks

Both models place the optimisation of the value streams

withathirdpartywhichislikelytohaveamoredeveloped

set of skills and capabilities to generate value from the

storagewithoutimposinghighcostsoftrading.Bothmodels

alsodependonawell-structuredtollingcontractthatgives

asmuch availability to the third party as possiblewithout

compromisingsecurity.

Bothmodelsdependonathirdparty’swillingnesstotakelong

termriskontheadditionalvaluestreams.However,underthe

DNOContractedmodelthereismoreflexibilityfortheDNOto

share someof this risk if required,and if theDNOcan take

somemerchantexposure.


FromaregulatoryperspectivetheDNOContractedmodelcould

face barriers as theDNOapproaches its present deminimis

thresholdfornon-distributionactivity.

The DNO Contracted model may allow the DNO flexibility

in the long term location of the storage, with the ability

to redeploy the asset to a new location (e.g. after full

reinforcementatacurrentlocation).However,theDNOmay

requireregulatoryapprovaltodisposeoftheassettoathird

party,whichitmightwishtodoifthestoragenolongerhad

significantvalueintermsofnetworksupportandcouldnot

be economically redeployed. Under a Contracted Services

modeltherewouldbenosuchrestrictionsbutrelocationof

theassettosupportthenetworkinadifferentlocationwould

bemorechallenging.

Conclusions

Based on our review of the lead models, both the DNO

Contracted and Contracted Services appear to be feasible

business models for distribution-connected storage. The key

barriers for these models are shared: the complexity of the

tollingcontract,andthewillingnessofathirdpartytotakelong

termriskontheadditionalvaluestreams.




5. Do you agree with the respective advantages and

disadvantages of the two lead business models as

described in Section 3?

a.Are thereother limitations,barriersor featuresof these

businessmodels,orEESprojectsingeneralthathavenot

beenconsidered?

b.Doeitheroftheseleadbusinessmodelsdisproportionally

favoronepartyovertheother?

6. From your experience, which of the two lead business

models is most likely to be favoured?


Investment Model Templates4


price projections and storage performance scenarios can be

easilypopulatedtoshowthesensitivityofthekeyperformance

indicators (for each project party) to these respective

parameters.Itshouldbenotedthatthemodelhasbeenissued

with default input parameters, but these are notional and

should be updated with project specific values before users

appraiseandcomparetheirownbusinesscases.

Finally,themodelcalculatesthefeasiblerangeofcumulative

tollingchargesbetweentheDNOandThirdPartythatwould

give eachof theproject parties positive net present values

(NPVs),effectivelydefiningamaxandmintollingchargethat

would stillmake theprojectprofitable forbothparties. The

actualuserdefined tolling charge is thencomparedagainst

thisfeasiblerangetoassessiftheprofit/costsplitisweighted

towardsonepartyoranother(Thetollingchargeisthetotal

charge made over the life time of the project by the third

partytotheDNOinthecaseoftheDNOContractedmodel,and

fromtheDNOtothethirdpartyinthecaseoftheContracted

Services mode (See Section 3.4 for a definition of tolling

chargeterms)).

4.2.ModelSpecification

A summary of the model structure, the inputs parameters,

calculationsandoutputsisgivenbelow.Themainassumptions

that were made when preparing the model along with its

inherentlimitationsarealsostated.

4.2.1.Structure

Theinvestmentmodeltemplateissplitintofourmainsections

with supplementary user aids where applicable. A high-

level summary of these four main sections along with their

interrelationsisgiveninthemodelschematicgiveninFigure14.

ToaidtheConsultationprocessaninvestmentmodeltemplate

describing possible business models for distribution-connected

energy storage applications was developed. This investment

modeltemplateisavailablefromUKPowerNetworksandisbeing

issuedinconjunctionwiththisConsultationdocumenttoallow

furtherdiscussionandconstructivefeedbackontheConsultation.

4.1.TemplateIntroduction

Theinvestmentmodeltemplateexaminestheinvestmentcases

forthetwoleadbusinessmodels,namelytheDNOContracted

and Contracted Services models. A third “Project Model” has

beendevelopedinthetemplateasapartyagnostic,hypothetical

referencecasemodelagainstwhichtheinvestmentmodelsfor

thetwoleadbusinesscasescanbebenchmarked.

The investmentmodel templateshouldbeusedby interested

parties to further investigate the two lead business models

described and characterised above. The investment model

templatefirstlyallowsuserstoinvestigateindetailthemagnitude

andinterrelationsofcost,profitandrisksastheyaccruetothe

differentprojectpartiesinthedifferentbusinessmodels.

Secondly, interested parties can input their own project

parameterstoexploretheseinterrelationsfortheirownproject

specific EES cases (the investment models are generic as

faras ispossiblebutaredesigned forprojects fallingwithin

the scope of this Consultation as defined in Section 1.3).As

directcomparisonscanbemadebetweenthethreebusiness

models users can then test the sensitivity of the indicative

financial performance of their project. The project’s financial

performancecanbeassessedfurtherfortherespectiveproject

partiesfordifferentinputparametersacrossthethreebusiness

models.Forexample,differentlife-timeassetutilisationsplits,


Investment Model Particulars

Usersinputthetechnicalparameters,theproject’sprimaryandancillaryrevenuestreamvolumesandrevenuesandtheproject’sfinancials

Figure 14 Investment Model Template Schematic

Project and Investment Model Introduction

Cover Sheet

Business Model Definition

Adefinitionoftheproject,theinvestmentmodeltemplateandthebusinesscasesconsidered

Discounted Cash Flows

Project Cash Flow Model

DNO Contracted Cash Flow Model

Contracted ServicesCash Flow Model

Traditional reinforcement Cash Flow Model

DCFsarepopulatedfromtheinputstothe“InvestmentModelParticulars”

Investment Model Summary

Asset Utilisation Summary

Business Model Comprision

Asummaryoftheassetutilisationsplit,thecostandrevenuestructuresandtheprojectprofitabilityfromaproject,DNOandthirdpartyperspective

User Aids

Assumptions

Glossary of Terms

Checklist

Technical Parameters

Revenue Streams

Financials


for the three business models is given in the “Business

ModelComparison”sheet.Thissheetalsosummarisesthe

total costsand revenuesas theyaccrue to the respective

parties,alongwiththeprofitabilityofthedifferentbusiness

models,onceagainsplitforthedifferentprojectparties.The

BusinessModelComparisonsheetalsocalculatestherange

of cumulative tolling charges for which all parties would

haveapositiveNPV(i.e.themaximumandminimumtotal

tollingchargethatcanflowbetweentheDNOandthethird

partyoverthelifetimeoftheprojectfortheprojecttostill

beprofitableforbothparties).

4.2.2. Inputs

Alegendofthetemplate’scelltypesincludingthecellswhich

usersarerequiredtoinputvaluesforisgiveninthetemplate

coversheet.Forindicativepurposesonlythesecellshavebeen

populated with notional values throughout the model as it

is issued.Thesedefaultvaluesarenot representativeofany

empiricallybasedrealcasescenarioandassuchusersshould

inputtheirownprojectspecificvalues(SeeSection4.2.5).

Achecklistoftheparametersthatusersareaskedto input

values for, alongwith the level of detail (or granularity) to

whichtheusersareaskedtoinputforeachoftheseparameters

isgiveninthe“Checklist”worksheet.Thedefinitionsofthese

parameters are included in the Glossary of Terms sheet for

reference.Usersareaskedtocompletethischecklisttosatisfy

themselvesthattheyhaveinputtedtheirownparametersfor

eachoftheseinputsbeforeusingthemodeltoappraiseand

comparebusinesscases.Ahighlevelsummaryofthemodel’s

inputcategoriesaregiveninFigure15.

The “Cover Sheet” and “Business Model Definition” sheets

introduce the Consultation process and the characterisations

of the threemodelsexplored in the template. Similarly, the

“Assumptions”, “Glossary of Terms” and “Checklist” sheets

define themainassumptions, thenomenclatureusedanda

qualityassurance(QA)checkoftheusers’inputstothemodel.

Usersarefirstlyaskedtoinputthemodelparametersinthe

mainbodyofthetemplate.Themodelparameterscomprise

thetechnicalparticularsandthecapitalandoperationalcosts

inthe“TechParameters”worksheet,therevenuestreamsin

the“RevenueStreams”worksheet,andthefinancingoptions,

gearinganddebttermsinthe“Financing”worksheet.

Theseinputsarethenallocatedasrevenuesandcostsasthey

accrue to the respective project parties in three discounted

cashflow(DCF)sheets.Todisaggregatethecostandrevenue

splitsfortheprojectpartiestheDNOcontractedandContracted

ServicesDCFsarefurthersplitintotwoseparateDCFsforthe

DNOandthirdpartyrespectively,giving5DCFsintotal(1for

the Project Model and 2 each for the DNO Contracted and

ContractedServicesmodels).Thereisalsoaseparatesimple

DCFmodelforthetraditionalreinforcementoption,whichis

usedasacounterfactualcomparison.

The results of the model are then summarised in two

output sheets. The technical performance of the asset

and its utilisation split for the users’ input parameters

are summarised in the “Asset Utilisation Smry” sheet.

Then, a comparisonof thefinancial performanceand the

netpresentvalues(NPVs)forthedifferentprojectparties


Figure 15 Model Input Categories

Technical Parameters

Project Particulars

Technology Particulars

Revenue Streams

Capital Expenditure

Operational Expenditure

Cost Particulars

Offset Reinforcement Cost

Security of Supply Volumes

Ancillary Services Volumes and Prices

Electricity Arbitrage

Short Term Operating Reserve Volume and Prices

Service Operator Charges

Frequency Response Volumes

and Prices

Embedded Benefits

Financing

Cash-Flow Particulars

Terms of Loans

Financing

Taxation


effort required or, in the case of the Contracted Services

modeloramaximumagreed level in thecontractbefore

penalties are activated. Balance of plant losses refer to

the energy lost in the charge and discharge cycles and

which ismeasuredat thefinalpointofconnectiontothe

DNOnetwork, including losses in step-up and step-down

transformers for example. Notwithstanding the fact that

losses may be reduced elsewhere on the network as a

result, the losses are included here in order to reduce

the“nameplate”capacityofthedevicetoa levelthat is

consistentwiththerevenuestreams.

4.2.3. Calculations

Firstly, the inputs are used to calculate the net available

capacityofthebatterywithrespecttoitschargingvolumeand

durationofdischarge. The technology’sdischargecoefficient,

operational depths of discharge, system inefficiencies, and

storage degradation are then used to calculate the overall

systemefficiencyand the correspondingdischargingvolume

andduration.Asummaryoftheinefficienciesandtheireffect

on the overall system efficiency is graphically shown in the

TechParametersworksheet,anexampleofwhichisgivenin

Figure16.Technicalunavailabilityisduetomaintenanceoutages

andunscheduledfaultrepairs.Thismayrepresentanestimateof

Figure 16 Generation Capacity and System Inefficiencies

10.40

10.20

10.00

9.80

9.60

9.40

9.20

9.00

8.80

8.60

2013 2015 2017 2019 2021 2023

Max

Ava

ilabl

eSy

stem

Dis

char

ge(

MW

h/Cy

cle)

%D

isch

arge

Effi

cien

cy(

%)

Key

Storage Capacity (MWh/cycle)

Gross Storage Energy (MWh/cycle)

Battery Discharge Efficiency (%)

Effective Capacity Drawn from Grid (MWh/cycle)

100.00%

99.00%

98.00%

97.00%

96.00%

95.00%

94.00%

93.00%

92.00%

91.00%

90.00%


as a counterfactual DCF. The adjusted NPV is calculated by

subtractingthiscounterfactualNPV. It is theseadjustedNPVs

thatarepresentedintheoutputsummary.

Theinvestmenttemplatedoesnotattempttocalculatetheimpact

on the DNO’s Regulatory Asset Value (RAV). We assume any

incentivetofavourstorageovertraditionalreinforcementorvice

versawouldbeneutralisedunderawell-designedpricecontrol.

4.2.4.Outputs

Atechnicalsummaryofthesystem’sutilisationsplitispresented

intheAssetUtilisationSummaryworksheet.Theasset’sutilisation

is split according to the system’s availability, unavailability and

chargingtimeperyear,alongwiththetimethatthesystemisboth

availablebutcontractedandwhenit isexporting(discharging)

foreachoftherevenuestreams.Theassetutilisationsplit isa

function of the technical performance of the asset and the

numberofhoursofutilisationthattheuserhasinputtedforeach

oftheprimaryandancillaryservices.Anexampleofthegraphical

outputofthemodelisshowninFigure17.

The charge and discharge capacities and durations are

then coupled with the utilization (and availability where

applicable)pricesanddurationsfortheprimaryandancillary

revenuestreamstocalculatethesystem’scostofchargingand

correspondingrevenuefromdischarging.

Theserevenuestreamsandtheirassociatedcostsfeedintothe

DCFsalongwiththeproject’sCAPEXandOPEX,debtrepayment

costs and tax payments. The DCFs allow users to front load

paymentsandindexpriceinflationifnecessary.Asmentioned

previously,DCFsarepopulatedforthedifferentprojectparties

inthetwoleadbusinessmodelstoallowcomparisonsofthe

cost,profitandriskprofilesacrosstheprojectpartiesforthe

differentbusinessmodeloptions.

FromtheDCFtheNPVsfortheprojectpartiesarecalculated.

Firstly, the unadjusted NPV values are calculated, these

unadjusted metrics do not consider the additional cost

saving from offsetting the traditional reinforcement costs.

Theoffset traditional reinforcement cost is rather considered

Figure 17 Asset Utilisation Split Summary

2013 2015 2017 2019 2021 2023

120.00%

100.00%

80.00%

60.00%

40.00%

20.00%

0.00%

Key

Generating (Discharging)

Charging

Available (Unused)

Available (Contracted)

Unavailable

Gross Discharging Revenue (£/annum)

Charging Cost @ Full Utilisation to DOD (£/annum)

Ass

etU

tilis

atio

nSp

lit(

/yea

r)

Year


forthecommercialviabilityfromtheperspectiveofthediffering

projectpartiesinthecaseoftheDNOContractedandContracted

Services models. For the DNO, the discounted cumulative cash

flowsareplottedagainsttheoffsetreinforcementcosttocheckif

thecumulativecashflowatprojectend(theNPV)islessthanthe

offsetreinforcementcost(similartothefirstoutputabove).Forthe

thirdparty,thecumulativediscountedcashflowsareplottedover

thelifetimeoftheprojecttoassessifthecumulativecashflowat

theprojectend(theNPV)ispositiveasshowninFigure19.

The second output sheet, which is the “Business Model

Comparison”sheet,firstlyconsidersiftheSNSoptionismorecost

effectivethantraditionalreinforcement.Themodelcompares

theunadjustednetpresentvalueof thestorageoption (the

costofstorage)againstthetraditionalreinforcementcost.An

exampleoutputisshowninFigure18below;

The cumulative cash flows are then assessed for the projects’

overallcommercialviabilityinthecaseoftheProjectmodel,and

Figure 18 Traditional Reinforcement and Storage Option Cost Comparison (£)

Figure 19 Project Profitability Checks

4,040,016

3,907,208.49

Key

Storage Cost (For the DNO)

Offset Traditional Reinforcement Cost

1 2 3 4 5 6 7 8 9 10 11

Year-50,000

50,000

100,000

150,000

200,000

250,000

300,000

-

Cum

ulat

ive

Dis

coun

ted

Cash

flow

(£) Key

Project Profitable

Discounted Cumulative Cashflow

1 2 3 4 5 6 7 8 9 10 11

-4,100,000

-3,900,000

-3,800,000

-3,700,000

-3,600,000

-3,400,000

-3,300,000

-4,000,000

Off

setR

einf

Cos

t&C

umul

ativ

eCa

shflo

w(

£)

-3,500,000

Key

Offset Reinforcement

Cost (The Cost of Reinforcement)

Project Profitable

Discounted Cumulative Cashflow (The Cost of Storage)

Year


TheBusinessModelComparisonsheetalsoallowsusers to

runamacrothatcalculatestherangeoftollingchargethat

would give positive NPVs to all of the respective project

parties(i.e.goalseekingtheyearlytollingchargetofindan

NPVofzero).Thetollingchargeisthetotalchargemadeover

thelifetimeoftheprojectbythethirdpartytotheDNOinthe

caseoftheDNOContractedmodel,andfromtheDNOtothe

Figure 20 Range of Feasible Tolling Charges (Cumulative over life-time of project)

thirdpartyinthecaseofthecontractedservicesmodel.The

actualtollingchargeisthenplottedonthisrangeoffeasible

tolling charge to assess if the profit/cost split isweighted

towardsonepartyoranother,oriftheactualtollingcharge

isoutside the feasible range (i.e.oneof thepartieshasa

negativeNPV).AsamplecasefortheDNOcontractedmodel

isgiveninFigure20below;

10,000,000

9,000,000

8,000,000

7,000,000

6,000,000

5,000,000

4,000,000

3,000,000

2,000,000

1,000,000

-

Tota

lPro

ject

Tol

ling

Char

ge(

£)

6,000,000

8,925,088

4,145,893

MaximumTollingChargethattheThirdPartycanaffordtopaytheDNO*

(Increase => Better for DNO)

MinimumTollingChargethattheDNOcanaffordtoreceivetheThirdParty*

(Decrease => Better for Third Party)

AnnualTollingCharge(£ Real)

892,509

600,000

414,589

TotalCharge(Cumulative)

(£ Real)

8,925,088

6,000,000

4,145,893

Max

Actual

Min

RangeoffeasibleTollingCharges

Key

Range of Tolling Charge for which Project has a +ve NPV for all parties

Actual Tolling Charge Payment

*�To�give�that�party�a�positive�NPV��If�line�outside�range�then�-ve�NPV�for�a�party�If�no�range�shown�the�project�cannot�be�profitable�for�both�parties


4. It is assumed that the DNO can transfer the incremental tax

loss benefits internally, as such these are considered as a post

EBITDA benefit in the DCF. This option can be switched off, in

which case these benefits are accrued within the project.

5. The NPV of traditional reinforcement cost is separately

calculated and then used as a counterfactual . Users are asked

to input the CAPEX of the traditional reinforcement, and the

terminal value at the end of the economic assessment period.

6. As the STOR service is typically contracted in two split

diurnal periods it is possible that the asset is called during

both of these periods. If this were the case the asset would

need to charge during the day between these two cycles.

Accordingly the electricity price for charging for use as STOR

is the average day time price, as opposed to the off-peak

night time low price for the rest of the ancillary services.

7. This model has been constructed for GB specific cases.

8. The reporting currency is assumed to be GBP Pound Sterling (£).

9. UK Corporate tax rates can be defined by the user, but have

been assumed in the model to be as per the HMRC published

figures (http://www.hmrc.gov.uk/rates/corp.htm).

10. The model assumes all senior debt drawn down at once

in development start year.

11. The depreciation rate is as per; http://www.hmrc.gov.uk/

capital_allowances/investmentschemes.htm General rates

of capital allowances section (can be changed by user).

12. The model assumes a single charge/discharge rate for all

services per day, therefore the model cannot differentiate

for different charge/discharge rates for different services.

13. The model cannot be used retrospectively on projects

commencing before 2013.

14. The max project lifetime allowed for in the model

mechanics is 20yrs (input as 12yrs as a default).

15. Gearing ratios outside a 20-50% range give a user warning.

16. Years refer to financial years.

17. Self-discharge loss calculation assumes half of battery

capacity stored per cycle.

4.2.5.Assumptions

All technical, utilisationandpriceassumptions thatareused

inthemodel’scalculationsareinputtedasmodelparameters

byusers.However,thefollowingadditionalassumptionshave

beenmadeintheinvestmentmodeltemplate;

1. In the DNO Contracted Model (in addition to the model

characterisation given in Sections 2 & 3);

a.CAPEXisaccruedtotheDNO

b.OPEXisaccruedtotheDNO(withtheexceptionofthe

“Control systems, trading & risk management” OPEX

whichaccruestothethirdparty)

c.Embedded benefits (i.e. avoided TNUos and BSUoS

charges, and savings in transmission and distribution

losses)areaccruedtothethirdparty

d.Capacitypaymentsareaccruedtothethirdparty

e.Thetollingchargeispaidannuallybythethirdpartyto

theDNOand is a fixed sum that canbe indexed and

front-loadedifnecessary

f. The DNO sees the benefit of the offset traditional

reinforcementcostasacounterfactual

g.TheimpactontheDNO’sRegulatoryAssetValue(RAV)is

notcalculated.

2. In the Contracted Services Model (in addition to the model

characterisation given in Sections 2 & 3);

a.CAPEXaccruestothethirdparty

b.OPEXaccruestothethirdparty

c.Embedded benefits (i.e. avoided TNUos and BSUoS

charges, and savings in transmission and distribution

losses)areaccruedtothethirdparty

d.Capacitypaymentsareaccruedtothethirdparty

e.The tollingcharge ispaidannuallyby theDNOto the

thirdpartyisafixedsumthatcanbeindexedandfront-

loaded if necessary The DNO sees the benefit of the

offsettraditionalreinforcementcostasacounterfactual.

3. All values are real unless otherwise stated to be nominal

(i.e. CAPEX, Tax and Debt in DCFs are nominal).


18. A single asset is considered in the model. Models may exhibit

different characteristics if a portfolio were to be considered.

19. Constant and symmetrical charging and discharging

across all services is assumed. This means simultaneous

services cannot be modelled.

4.2.6.Limitations

The following limitations apply to the investment model

templateandshouldbeconsideredbyusersbeforeusingit:

1. The investment model template is a learning aid issued for

illustration and comparison purposes only. The template

should not be used under any circumstances as the primary

investment model for any EES project or otherwise.

2. While the investment model template has been prepared for a

specific application type it is still inherently a generic investment

model, and as such cannot account for every eventuality and

permutation that may arise in specific EES projects.

3. For illustrative purposes the model as it is issued has

been populated with default values for the model’s input

parameters. These default values are not based on empirical

real case values and as such cannot be used to appraise project

specific cases (The model has an in-built functionality to clear

all these user defined inputs and to then check that the user

has inputted values for all the required input parameters).

4. The default input financing terms (debt terms, discount

rates, gearing and WACC) are indicative only and would

need to be inputted by users.

5. The model uses a peak/offpeak spread for the value of

energy arbitrage, and does not capture the full volatility

of wholesale prices.

4.3.BusinessModelInsights

4.3.1.DNOContractedModel

IntheDNOContractedmodeltheDNOfinancestheconstructionof

theassetanditsannualoperation,accruingtheprojectCAPEXand

OPEX.ThemodelallowstheflexibilityfortheDNOtofinancethis

throughamixtureofdebtandequity.Thethirdpartyontheother

handwouldonlyaccruetheannualcostsrequiredtooperatethe

asset’sancillaryservices.Thesearedescribedas“Controlsystems,

trading&riskmanagement”costsinthemodel.Thethirdparty

wouldtheninstructtheDNOontheoperationoftheassettoand

whentocontract(ordispatch)theassetforancillaryservices.The

thirdpartywouldaccrueanyrevenuefromtheutilisationofthe

assetfortheseancillaryservices.Thethirdpartywouldthenreturn

anannualtollingpaymenttotheDNOasapaymentforutilising

the DNO’s asset. The DNO also considers the offset traditional

reinforcementcostwhencalculatingtheiradjustedNPV.

Populating the investment model template with a range of

notionalvaluesandscenariosallowsuserssomeusefulinsights

into thediffering interrelationsofcostand revenuesas they

accruetotheprojectpartiesintheDNOContractedmodel.Most

notably,asthetollingchargeintheDNOContractedmodeldoes

notcontributeaslargeaproportionoftheprojectcashflowas

intheContractedServicesmodel,theNPVoftheprojectisnot

assensitivetothetermsofthetollingchargeagreement(i.e.

thereisagreatercalculatedrangeofpossibletollingcharges).

Also,theDNOContractedmodeltendstogivethethirdparty

ashortpaybackbutalowerNPVcomparedtotheContracted

Servicesmodel(i.e. theydonotaccruetheCAPEXandOPEX

costsandhaveminimalcommercialrisk).

Inbothmodels,thereisaninherentandunavoidableconstraint

on revenue generation during the periods where the DNO

requirestheassettobeavailableforsecurityofsupplyservices,

theSecureCapacityWindows.

4.3.2.ContractedServices

IntheContractedServicesmodelthethirdpartynowfinances

theconstructionoftheassetanditsannualoperation,accruing


The Contracted Services Model gives a longer payback for

the thirdParty thantheDNOContractedModel,buta larger

NPV(i.e.theyneedtopaybackthelargeprojectCAPEXand

wouldexpectlargerfinancialbenefitsforcarryingtheproject’s

commercialrisk).

As in theDNO Contractedmodel there is again an inherent

andunavoidableconstraintonrevenuegenerationduringthe

periodswheretheDNOrequirestheassettobeavailablefor

securityofsupplyservices.




7. Are there other technology parameters, costs or revenue

streams that should have been considered in the

investment model template?

a.Ifsopleasegivedetails

8. Do you agree with the interrelations of these ancillary

services and their associated revenue streams?

b.Arethereadditionalcomplexities inthedispatchofthe

assettoutilisetheserevenuestreamsthathaven’tbeen

considered?Aretheyallmutuallyexclusiveorpotentially

dispatchableinunison,arethereadditionalcomplexities

intheknock-oneffecttobatteryperformancethathave

notbeenconsidered?

9. Do you agree with the stated assumptions and model

limitations?

a.Should any of these stated assumptions or limitations

havebeendealtwithdifferently?

theprojectCAPEXandOPEX.Thethirdpartywouldbeexpected

torequiresomedebtfinancingandwouldaccruethesedebt

and interestrepayments.TheDNOwouldaccruenoneofthe

project’sCAPEXorOPEXbutinsteadwouldpassanannualtolling

paymenttothethirdparty,nowintheoppositedirectiontothe

DNOContractedmodel.Thethirdparty,inadditiontoreceiving

thistollingchargepaymentfromtheDNOwouldalsoaccruethe

revenuesfromtheancillarybenefitswhichitwoulddispatchthe

assetfor.Thethirdpartyalsoaccruesanyembeddedbenefits

and capacity payments as they have full ownership and

operationalcontroloftheasset.TheDNO’scashflowissimpler

thanintheDNOcontractedmodelwiththetollingchargebeing

theironlycost,andtheoffsettraditionalreinforcementcosttheir

onlybenefit.

Again, populating the template with notional values and

scenariosallowsuserssomeusefulinsights.Astheeconomic

benefitsoftheoffsettraditionalreinforcementcostdoesnot

accruetothepartywhoismakingtheinvestment,thethird

party,thereisaneedforacorrespondingtollingpaymentto

thethirdparty.This,inthecaseoftheDNOContractedModel

wouldbeapaymentfromtheDNOtothethirdparty.While

thisdoesnotdifferinprincipletotheofftakeagreementand

paymentintheDNOContractedmodel,itisthevalueofthis

additionalrevenuestreamtotheoverall investmentcaseof

theDNOandthirdpartythatismateriallydifferent(effectively,

thetollingchargeneedstocovertheCAPEXwhenitaccruesto

thethirdparty).

ThecorollarytothisisthattheDNOavoidstheupfrontexpense

offinancingtheasset,andspreadsitoverthelifetimeofthe

projectastollingchargepaymentstothethirdparty.


The Value of StorageAppendix 1


frequency,whichisachieved10secondsfromthetimeof

theFrequencychangeandissustainedthereafter.

TherequirementforFrequencyResponseiscreatedbydemand

forecastinguncertainty,thelossofgenerationfromthesystem

andincreasingwindgenerationforecastuncertainty.

ThetypeofFrequencyResponsewhichwouldbeprovidedby

thestorageisFirmFrequencyResponse.Mandatoryfrequency

responseisanalternativemarketthatisrequiredtobeprovided

under the Grid Code by all sufficiently large generators. An

energy storage device is also eligible to provide Frequency

ControlbyDemandManagement(FCDM),whichisalignedto

providersoffrequencyresponsefromdemandCustomers,but

isalsosuitableforstorage,andthebilateralcontractscanbe

appliedtobothpositiveandnegativeresponse.

Short Term Operating Reserve (STOR)

STOR is one source of reserve for the GB System Operator,

NationalGrid Electricity Transmission.Reserve is required for

thefollowingreasons:

• Demand forecast errors: Most end users of electricity do

notneedtoprovideanystatementoftheirintendedusage

andsoelectricitydemandisuncertainandactualdemandis

oftenquitedifferenttoforecastevenquiteclosetorealtime.

• Unexpected loss of thermal generation:Thelargestinfeed

loss is currently the nuclear unit at Sizewell B. The larger

capacityofthenewnuclearstationsoncetheycomeonline

willleadtoanincreaseinSTORcapacity.

• Variable wind generation: Output from wind capacity is

inherently variable and unpredictable even close to real

time. Therefore reserve is required todealwith situations

wherewindgenerationislowerthanexpected.

STORiscapacitythatNationalGridretainsonstand-bythatcan

becalledontoexportwithin fourhoursof instruction(witha

Local Security of Supply (SOS)

WhenthestorageassetisrequiredtomeetlocalSoSobligations

thestoragemustbeavailablewithasetchargewhencalled.

Itshouldbenotedthatthestoragedoesnotneedtoexport

in order to fulfill the security requirement, but the capacity

mustbeavailableandthestoragemusthaveenoughstored

energytocoverthepeakperiodifrequired.Thestorageasset

willprovidesecurityifitisactivelyexporting.Thiscouldoccur

iftherewasadditionalvaluetoexportingpower(e.g.energy

arbitrage).Inthisinstanceitwillbeimportanttoensurethat

the security requirement was met throughout the security

requirementperiod.Similarly,dependingonthestorageasset

typeanditsoperationtheconstraintcouldalsoberelievedif

theassetwheretostopchargingifitwerebeingchargedatthe

timeoftheconstraint.

Ifafaultdoesoccuronthelocalnetworkatthetimewhenpeak

demandexceedsfirmcapacity,thestoragewillberequiredto

dischargeuntilthedemanddropstoasecurelevel,thenormal

capacity is restored, or the storage is fully discharged. This

wouldtakeprecedenceoverallotheruses.

Firm Frequency Response (FFR)

Frequency Response is the automatic provision of increased

generation or demand reduction in response to a drop in

system frequency. This can be further subdivided into three

typesofresponse:

• Primary response is defined as an initial increase of

generation, with sustained output from 10 seconds to 30

secondsfollowingalossof0.8Hz.

• Secondary responseisdefinedasanincreaseingeneration,

inresponsetosystemfrequencystillbeinglowerthantarget

frequency, with sustained output from 30 seconds to 30

minutesforalossof0.5Hz.

• High response is defined as a decrease in generation, in

response to system frequency being higher than target


storageissufficienttocoverexpectedandunforeseenchanges

indemandoverthepeakperiod.

Embedded Benefits

“Embedded generation” is electricity generation that is

connectedtothelocaldistributionnetworkratherthandirectly

tothetransmissionnetwork.Beingembeddedinthedistribution

networkmayallowparties to avoid various costs,whichare

termed“embeddedbenefits”.Themajorcategoriesofavoided

costsassociatedwithembeddedgenerationareasfollows:

• transmissionlosses

• distributionlosses

• TransmissionNetworkUseofSystem(TNUoS)charges,and

• BalancingServicesUseofSystem(BSUoS)charges.

These embedded benefits may apply to some extent

to distribution-connected storage. As the avoided costs

associatedwithembeddedgenerationoftenaccruedirectly

tothesupplierratherthanthegenerator,embeddedstorage

would need to negotiate with suppliers in order to realise

actualembeddedbenefits.

Other DNO Services

ThepowercontrolsystemassociatedwithsomeEEStechnologies

canprovideusefulsecondarybenefitsofimprovedpowerfactor

andvoltagesupport,reducedsystemlosses,andpowerquality

enhancementbymeansoftheassociatedharmonicfilters.

Capacity Payments

UndertheElectricityMarketReform(EMR)programme,theUK

Governmentis introducinglegislationtoprovideforafuture

capacitymechanism.Tomitigatetheriskstosecurityofsupply,

theBillprovidesfornewpowersfortheSecretaryofStateto

introduceaCMtoensurethereisenoughcapacityavailableto

meetexpecteddemand.

focus on <20min). The STOR service retains spare generation

capacityonstand-byduringcertainhoursof theday(typically

periods when demand is changing rapidly). There are two

categoriesofSTOR:

• Committedprovidersmustbeavailableinalloftherequired

availabilitywindowsineachseasontheyarecontracted.

• Flexible providers are not obliged to offer services in all

availability windows and National Grid is not obliged to

accepttheservicewhenoffered.

STOR is open to both Balancing Mechanism (BM) participants

(generally transmission connected generation from large power

stationsites)andnon-BMparticipants(generallysmalltransmission

or distribution connected generation and demand). However BM

participants must be ‘committed’ providers of STOR generation.

Distributionscalestoragewouldbemostlikelytobeanon-BMflexible

provider(althoughitmaybepossibletocommitforcertainSTOR).

Energy Arbitrage

Energyarbitrageisthetradingofwholesaleelectricitytobenefit

fromthespreadsbetweenpricesatdifferenttimes.Thistypically

involvesbuyingpowercheaplyovernightandsellingitatpeak

whenpricesarehigher.Asthevolumeofintermittentgeneration

increasesinfuture,thevolatilityinpricesislikelytoincreaseand

timingofhighestandlowestpricesmaybemorevariable.

Energyarbitragerequiresthespreadinpricestobegreatenough

tooffsetthecostoflostpowerduetotheefficiencyofthecycle

(charginganddischargingtheasset).Anotherconsiderationisthat

multiplechargingcycleshaveanimpactonthelifetimeofthe

battery,whichshouldbeaccountedforinthedispatchdecision.

Intheorydistribution-connectedEEScoulddischargeoverthe

peakperiodandstillprovidesecurity,but thiswouldrequire

careful control to ensure that the remaining energy in the


TheCapacityMechanismwouldpayprovidersofcapacitysuch

asgenerators,storageandDemandSideManagementbased

ontheresultsofacapacityauction.Thefirstauctionisplanned

for2014withdeliveryofcapacityintheyears2018/19.Itis

proposed to hold early trial auctions in 2014 for delivery in

2016whicharetargetedatDSMandstorage.

Ancillary Service Exclusivity

Inmostcasestheasset’sancillaryservicesandsecurityofsupply

obligationswouldbemutuallyexclusive(i.e.Ancillaryservices

couldnotbeusedinsidetheSecureCapacityWindows).For

example,STORandenergyarbitrageifusedinsidethecapacity

windowswouldadverselyaffecttheassetscapacitytomeet

itsSoSobligationasbothoftheseancillaryservicesnecessitate

thedischargeoftheasset’sstoredcapacity.Asdiscussedin

theSecurityofSupplySection,theassetwouldberequiredto

dischargeifafaultweretooccuronthelocalnetworkatthe

timewhenpeakdemandexceedsfirmcapacity.

However, while the terms of the capacity windows would

definehowtheassetwouldbeutiliseditcouldbethecase

thatforsomeoftheancillaryserviceswhenspecificstorage

technology types are used that the asset would not be

requiredexclusivelyforprovisionofSoS.Forexample,inthe

caseoffirmfrequencyresponsetheassetcouldprovideadual

service,providingfirm frequency responseavailabilitywhile

not adversely affecting its capacity to meet its security of

supplyobligation.Thisispossibleastheassetwhenusedfor

firmfrequencyresponseisunlikelytoberequiredtodischarge

significantly,duetotherelativelyshortdurationofoperation.