LoSchiavo ERRA Budapest 08May rev0€¦ · Staff of the Regulatory Authority have the duty to disclaim in public that only personal opinions are presented when speaking in public

05/05/2018

1

1

Budapest, 08.05.18ERRA Course «Principles of Electricity Markets»

Incentive Regulation of Smart Distribution Systems: from Input-based Demonstration to Output-based Deployment

Luca Lo Schiavo ARERA - Regulatory Authority for Energy, Networks and Environment Infrastructure Regulation, deputy directorMember of CEER DS (distribution system) WG Member of ACER INF (elecricity infrastructure) TF

Staff of the Regulatory Authority have the duty to disclaim in public that only personal opinions are presented when speaking in public at conferences, workshops and seminars.

Ethical code of ARERA,10(2)

Content of this presentation

Luca Lo Schiavo, AEEGSI (Italy) 2ERRA 08.05.18

RES evolution in Italy and effects of “system trans formation”The Italian power system and the impact on distribution networks

Overall approach to innovationInput-based vs output-based incentive mechanisms

Input-based regulatory incentives for demonstration phaseLearning from pilot projects of active distribution grids

Output-based regulatory incentives for deployment p haseThe mechanism for promoting roll-out of smart functionalities

Annex 1: Output-based regulatory incentives for Quality of ServiceAnnex 2: Smart metering (from 1st to 2nd generation)Annex 3: Electro-mobility and recharge infrastructure

05/05/2018

2

KEY FIGURES: energy demand in Italy

Luca Lo Schiavo, ARERA (Italy) 3ERRA 08.05.18

Total demand 2016: 320.000 GWh/yr

Self-consumption:30.000 GWh/yr (estimate)

System peak: 55-59 GW~20 GW min.load night, ~30 GW Sundays, daylight

Household consumption~2.200 kWh/user3.3 kW capacity

0,6 Million prosumers

0%

20%

40%

60%

80%

100%

23%

2%

25%

36%

15% HV

MV

LV(small

business)

LV(households )

KEY FIGURES: electricity infrastructures in Italy

Distributionand metering

Transmission

Number of companies 137 DSOs 1 TSO

Voltage levelsMV (10-20-25 kV)

LV (230-400 V)HV (130-150 kV)

EHV (220-380 kV)

Type of network Radial(with back-feeding in MV)

Meshed

Dimension Total: 37 Million cust. Around 60.000 km(of which 23.000 EHV)

Aggregated RAB 22,700 M€ 11,700 M€

UnbundlingFunctional and legal

De-branding (>100k cust)

Full ownershipunbundling


05/05/2018

3

KEY FIGURES: electricity distribution companies in It aly

DSONumber of

LV customer

1. E-distribuzione 32.2 million

2. Areti (Roma) 1.6 million

3. Unareti (Milano/Brescia) 1.1 million

4. Ireti (Torino) 0.5 million

5.SET distribuzione (Trento province)

0.3 million

6. – 10.5 DSO «medium size»250.000 - 100.000 cust.

0.8 million (aggreg.)

11. – 42.32 DSO «small size»100.000 – 5.000 cust.


43. – 137.95 DSOs «micro-size»Less than 5.000 cust.



«SYSTEM TRANSFORMATION»: main driving forces

6

The electricity sector is undergoing a radical transformation in a very narrow timeframe:

• strong growth of renewable sources All RES are 44.6% of total installed in 2016 compared with 24% in 2004; All RES produce 37.3% of total production in 2016 compared to 18.4% in 2004;

• strong spread of unpredictable renewable sourcesWind+PV are 24.5% of the total installed in 2016 compared to 1.3% in 2004; Wind+PV produce 13.7% of the total production in 2016 vs 0.6% in 2004

• strong spread of distributed generation (up to 10 MVA) DG is 21% of the total installed in 2015 compared with 4.6% in 2004; DG produce 18.1% of the total produced in 2015 compared with 4.7% in 2004

• radical changes of electricity flowsThe electricity flows in Italy have changed in recent years , due to the high penetration of new RES especially in the south, where the load is lower; transmission investments are relevant (around 1 billion/year)

ERRA 08.05.18 Luca Lo Schiavo, ARERA (Italy)

05/05/2018

4

«SYSTEM TRANSFORMATION»: gross installed capacity

7ERRA 08.05.18 Luca Lo Schiavo, ARERA (Italy)

0

5.000

10.000

15.000

20.000

25.000

30.000

35.000

40.000

45.000

50.000

55.000

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016

MW

Biomass and waste Photovoltaics Wind Geothermal Hydro

22,5% of the total

installed

44,6% of the total

installed

PV: 3,5 GW in 2010 18,4 GW in 2013 19,3 GW in 2016

«SYSTEM TRANSFORMATION»: gross energy production


05.000

10.00015.00020.00025.00030.00035.00040.00045.00050.00055.00060.00065.00070.00075.00080.00085.00090.00095.000

100.000105.000110.000115.000120.000125.000

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016

GWh Biomass and waste Photovoltaics Wind Geothermal Hydro

19,0% of the total

37,3% of the total

Unpredictable renewablesources (PV and wind): 13,7% of the total

05/05/2018

5

«SYSTEM TRANSFORMATION»: peak load and reserve margin


19,0% of the total

Unpredictable renewablesources (PV and wind): 13,7% of the total

«SYSTEM TRANSFORMATION»: residual demand (net DG)


…from 2010…

… to 2015

-11,0 GW

-3,0 GW

Average Sunday in April , residual demand

The phenomenon is really crucial on Sundays and Holidays in Spring and Summer(low load, high PV production if the sun is shining, combined in spring

with still low temperatures and high performance of PV panels)

05/05/2018

6


«SYSTEM TRANSFORMATION»: changes in hourly profiles

2010 2016

Higher ramps

Curtailment risk

Loadcovered

by wind and PV

Week-days

Week-ends


«SYSTEM TRANSFORMATION»: changes in energy flows

2004 2015

05/05/2018

7


«SYSTEM TRANSFORMATION»: cost of State incentives

Mill

ions

Euro

per

year

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

2008 2009 2010 2011 2012 2013 2014 2015 2016 2017

[TW

h]

Electrical energy produced from RES and encouraged

New incentives for renewables

Photovoltaic incentives

Feed in tariff l. n. 244/07

Incentives substituting Green certificates

Green certificates

Cip 6 (only renewables)Energ

y pro

duce

d[T

Wh]

State incentives have beenvery generous especially for

PV

From 2013, new PV units are no

longer incentivized

0500

1.0001.5002.0002.5003.0003.5004.0004.5005.0005.5006.0006.5007.0007.5008.0008.5009.0009.500

10.00010.50011.00011.50012.00012.50013.00013.50014.000

2008 2009 2010 2011 2012 2013 2014 2015 2016 2017

[Mili

oni d

i eur

o]

Costs of incentives for electrical energy produced by RES

New incentives for renewables

Photovoltaic incentives

Feed in tariff l. n. 244/07

Incentives substituting Green certificates

Green certificates

Cip 6 (only renewables)

14 di 65

RES Connection: Critical areas

• in white, all the non critical areas;

• in yellow, areas served through primary stations, in standard operating mode, for which Pimm > 0,5 * Pcmin;

• in orange, aeras served through primary stations, in standard operating mode, for which Pimm> Pcmin;

• in red, aeras served through primary stations, in standard operating mode, for which

Pimm – Pcmin > 0,9 * Pn. The red areas are considered critical areas.

For low and medium voltage networks, the grid operator identifies:

where:

• area is a province or group of municipalities or a municipal territory or part of it;

• Pcmin is the min load, defined as the load of the area in the quarter of hour of the minimum regional ;

• Pn is the sum of the rated power of all HV/MV transformers installed in the primary stations of the area;

• Pimm is the sum of the Required Injected Power, as requested by appliers for connection.


05/05/2018

8

15 di 65

Critical areas

��

��

EHV/HVMV

EHV/HVMV

EHV/HVMV

EHV/HVMV

EHV/HVMV

EHV/HVMV

T1 T2

T3

T4 T5

T6

Pcmin

area

Required injected power = 250 kWRequired injected

power = 250 kWRequired injected power = 250 kWRequired Injected

Power (RIP) = 250 kW

��

�

��

M1

M2

M3

M4

IMPACT ON DISTRIBUTION NETWORK: DSOs’ VIEWPOINT

Luca Lo Schiavo, ARERA (Italy) 16

Reverse Power Flow Time (RPT) : % of time (hours) during which energy flows from Medium Voltage (distribution) to High Voltage (Transmission)� RPT is an extremely relevant

indicator of distribution criticalities, used by the Italian Energy Regulator

Due to «connect and forget»approach, for the time being we do not have congestions on MV lines: RES have been connected only after proper network reinforcement. This approach may lead in the long run to overcost in network development (in comparison with smart solutions)

ERRA 08.05.18

05/05/2018

9







Annex 1: Output-based regulatory incentives for Quality of ServiceAnnex 2: Smart metering (from 1st to 2nd generation)

European Regulators’ recommendations for innovation (2 010)

Perform societal cost-benefit assessment

Introduce output regulation: value for money of users

Distinguish grid-related versus market-related activities

Improve consumer awareness for energy use and market opportunities

Learn from best regulatory practices

Ensure stable regulatory framework and long-term return on investments

Decouple profits and volume for grid operators

Incentivise innovative solutions (demonstration pilots)

Adopt open protocols and standards for interoperability

Disseminate the results and lessons learned from the demonstration projects

Luca Lo Schiavo, AEEGSI (Italy) 18

First, keep the system secure

ERRA 08.05.18

Source: CEER Smart Grid Position paper. June 2010 (E10-EQS-38-05)

ROLL OUT

PILOT

05/05/2018

10

Potential impact on system security (voltage dips on T- grid)

Due to reduction of rotating machinesconnected to Transmission grid, there isless Shortcircuit-Power available and therefore voltage dips generated atT-level have larger impact (in this simulation the spatial distributionof DG has been assumed homogeneous)

80% Vnante PV

80% Vnpost PV

Distance

Residual

Voltage

0

1 p.u.


DG frequency tolerance: outcome of retrofit (mandator y by ARERA)

Source: ENTSO-e «Dispersed Generation Impact on Continental Europe Region» position paper, 15 Nov.2014


A first EU-wide incident alreadyhappened on 4th Nov. 2006:split of Europeansynchronous area due to a RES-relatedproblem in NorthernGermany: ~2 GW of DG disconnected in Italy due to previousnarrow bands; nowadays, the sameincident would havea far larger effect, without the new tolerance windows

ERRA 08.05.18

05/05/2018

11

Reduced risk in case of deep voltage dip in the grid

With new FRT requirementsWithout new FRT requirements


SMART SYSTEMS =

SMART GRID + SMART USERS

DISTR. NETWORKAUTOMATION

CONNECTION RULES FOR DG

VOLTAGEREGULATION

NETWORKRESILIENCE

smart grids

ELECTRICVEHICLES

RECHARGINGINFRASTRUCTURE E-mobility

ELECTRONICMETERS

smart metering

COMMUNICATIONINFRASTRUCT.

FLEXIBILITYSERVICES

«HAN»: HOMEAREA NETWORK

Demandresponse

IN-HOMEDEVICES

DEMANDAGGREGATION

LARGE SCALEINTERMITT.GEN.

windintegration

STORAGESYSTEMS

storageV-2-G

SERVICES

ARERADECISIONS

292/06393/13

87/16222/17(SM 2g)

ARERADECISION 96/11POSITION 5/15

ARERA DECISIONS 39/10 > 12/11 > 646/15

ARERADECISION

5/10

ARERADECISION

43/1366/13

ARERACONSULTATIONS

354/13 AND 298/16DECISION 300/17


Focus of pilotdemonstration projects

What «innovation» includes?

05/05/2018

12

Output-based vs Input-based incentives: a frame

OUTPUT-BASED

• e.g. Quality of Supply (from 2000)

• Key outcome indicators (e.g.SAIDI, SAIFI+MAIFI, …)

• Metrics ought to be: clear and fair, trackable and auditable

• Output valuation is needed (based on estimates of avoided externalities – e.g. VOLL)

• May be used as stable regulation, however needs periodic tuning (typically every regulatory period)

INPUT-BASED

• esp. suited: Innovation (up to 2015)

• Benefit/Cost ratio used to select pilot projects

• Demonstration projects are evaluated ex-ante

• Input-based ncentive is not based on value for the customer/system

• Intended to be a transitional regulation: when metrics are developed, evolution towards output-based incentives is preferable for roll-out deployment


The Italian Regulator’s approach to demonstration pilot s

• Demonstration pilot: real operations in real grid (no lab)

• Regulatory attention to both effectiveness (performance) and efficiency (cost): pilots are paid by all customers….

• Transparency of the rules: procedures, evaluation methods and criteria, etc., known ex-ante

• Knowledge development with the involvement and the support of the best expertise (RSE and University like Politecnico MI)

• Continuous monitoring in the medium and long term: cost benefit analysis for the whole life-time of the new components

• Output disclosure: because demonstration pilots are paid by all customers > results must be public (no patents)

• Replicability and dissemination of the best-practices


05/05/2018

13








Requirements for smart grid demonstration pilot projec ts

• Focus on MV networks: 75% of DG rated power

• Active grids only: at least reverse power-flow for 1% of timefrom MV to HV

• Real time control system at MV level: the selected MV network has to be controlled (voltage limits / anti-islanding)

• Open grid: non-proprietary communication protocols required, in order to minimize interface costs for network users

• Selection process with both quantitative benefit/cost ratio and qualitative score – time consuming process

• Awarded with extra-WACC (+2%) for 12 years (“input-based”)

• Dissemination: report to the Regulatory Authority every six months for selected projects, published on Regulator’s website


05/05/2018

14

KPI approach for smart grid pilot projects evaluati on

Synthetic indicator IP used to assess ex-ante the expected performance of the demonstration projects

C

AjP

IP

m

ij

smart ∑=

⋅=

8760EIEI

P prepostsmart

−=

IP: priority index

Aj: project benefits [point score]

C: project costs [€]

P-smart: increase in DG-produced electricity / hour [MW]

EI-post: DG-produced electricity that can be injected in the network after the project in safe conditions [GWh]

EI-pre: DG-produced electricity that can be injected in the network before the project without reverse flow [GWh]

Aj: score for size, innovation, feasibility, replicability

Quantitative cost and mainbenefitindicator

Further benefits qualitative scoring


P-smart concept: hosting capacity in safe conditions

MV

HV

MVMV

LVLV

REVERSE POWER-FLOW

TIME: 1% of year Vcontrol

Psmart is the increase in DG-production (PDG) that can be connected to the grid in safe conditions (voltage, currents, frequency) thanks to smart investments on the grid

Passive network: NO flow from MV to HV

Smart network – latency 10-20 s remote voltage regulation

Smart network – latency 200 msremote intertrip (no islanding)*

Smart network – latency 200 ms: + storage (regulatory issues)

Min. Load

PDG

PDG= 0

Psm

art

* Very critical in Italydue to fast reclosure (400ms) for network automation


05/05/2018

15

«in safe conditions for voltage, currents and frequency»

• Power modulation: modulating active power

Related to Current (thermal) limits > latency: minutesCostly solution due to loss of revenue for generators (not implemented)

• Voltage regulation: modulating reactive power

Related to Slow Voltage Variations > latency: some tens of secondsVery low-cost solution for voltage network constraints

• Intertrip: combined dialogue between DSO and DG in order to avoid inslanding in case of network fault and secure the system

Related to both Frequency perturbation and fast Voltage Dips > latency hundreds of millisec (from 2012 a local solution has been implemented)

• Keep alive: in case ICT layer is not available, DG rolls back in old setting in order to avoid risks (check every 1 second)


Smart distribution system functionalities (tested i n pilots)


1. TSO-DSO integrationMeasurement collection, DG production forecasting and data transmission towards TSO systems - better observability

5. Fast MV fault IsolationDetection on isolation of MV fault sections without the tripping of the breaker at the line departure(requires extra-performant tlc )

6. Electricity storageUsage of electrochemical storage as a mean for improving QoSand managing RES intermittency – further evaluation needed

2. Voltage ControlParticipation of MV Distributed Generation to Voltage regulation on MV feeders – higher hosting capacity

3. Active power modulationParticipation of MV distributed generation to ancillary services market (not yet disciplined for units below 10 MVA)

4. Anti – IslandingDetection of possible islanding condition on MV Network – no longer needed after local solution implemented in 2012

05/05/2018

16

Dissemination activity of pilot projects


Acknowledgements (Annex 1 to cons.paper 255/2015/R/ee l)

L’Autorità esprime profondo apprezzamento per il lavoro svolto e l’impegno profuso da tutte le persone coinvolte dalla sperimentazione. Nell’impossibilità di ringraziare individualmente, si desidera in particolare segnalare

• il personale delle imprese di distribuzione che ha seguito i progetti pilota tenendo i contatti con l’Autorità (Fiori e Francucci di A.S.SE.M., Pisciotta e Tallei di ASSM, Fasciolo, Lucchini, Martinazzi e Pluda di A2A Reti Elettriche, Carta, Liotta e Zendri di ACEA Distribuzione, Loperfidoe Paulucci di ASM Terni, Bianchin e Perron di DEVAL, Alagna, Consiglio, Di Lembo, Di Napoli, Lombardi, Massimiano e Petroni di ENEL Distribuzione),

• i professori e ricercatori universitari che hanno fornito contributi tecnici e scientifici sia in fase di valutazione che di disseminazione (La Scala del Politecnico di Bari, Borghetti dell’Università di Bologna, Pilo dell’Università di Cagliari, Delfino dell’Università di Genova, Pelacchi dell’Università di Pisa, Turri e Fellin dell’Università di Padova, Ippolito dell’Università di Palermo, Capone, Delfanti, Falabretti, Merlo, Olivieri del Politecnico di Milano),

• nonché altri esperti (Timò del CEI, Sica di FederUtility, De Nigris e Celi di RSE, Carlini e Giannuzzi di Terna, e Denti, Graditi, Noia e Pigini coinvolti come esperti nella fase di predisposizione dei progetti)

• e tutte le altre persone, anche delle imprese fornitrici di apparati e servizi, che sono state a vario titolo coinvolte nella realizzazione dei progetti e di cui non è possibile indicare qui i nomi

Allegato 1, documento di consultazione 255/2015/R/eel, 29 maggio 2015


05/05/2018

17








PROJECT ASSESSMENT: lessons learned from abroad…

“Many technical and economic features of the Smart Grid, Distributed Generation and Demand Response providediffuse benefits to the customers that are hard to put a value on.

[US] regulators must engage in lenghty proceedings to set methods of measuring the value and then utilities must administer them under the critical eye of regulators…”

P. Fox-Penner, Smart Power, 2010

In ARERA strategic vision, output-based, selective incentive regulation is the key progress towards «smart regulation»


05/05/2018

18

INNOVATION: from pilot projects to output-based incent ives

CEER Guide-lines (2010)

Smart grid Pilot projects (launched 2011)input-based regulation

Dissemi-nation events (2013-14),public disclosure on website

Lesson learnt (public consultation 255/2015)

Output-based regulation (2016-19) for 2 smart functionalities: Observabilityand Voltage smart regulationon MV networks

Luca Lo Schiavo, ARERA 35

ARERA-AGCOMcollaboration on

M2M issues(2014>)

!

ERRA 08.05.18

Initial thoughts for output-based incentives (paper 255/2015)


Learning from pilots

Bal.Marketrules notready yet

DSO revenues stillRAB-based

Detailed analysis of smart functionalities

Valuable results for some functionalities

Without specific balancing market rulesfor DG, network users are not interested

but first results may be achieved by DSOseven without involving network users

New TOTEX approach envisagedfor eliminating perverse incentives

Output-based regulatory incentivesfor a few smart functionalities

High RES penetration

Selectivity of smart investments: priorityin areas with higher level of RES

Still to be defined: financial supportfor «early bird» network users

05/05/2018

19

Example of «complexity levels»: voltage control (255/2 015)


Caso Base Programma P3 Programma P3 + IRE

Linea Limite termico [MW]

Limite tensione [MW]

Limite termico [MW]


Aumento rispetto caso

base (%)


Limite termico [MW]

Aumento rispetto

caso base (%)

Carpinone 7.402 - 7.402 - 0% - - 0%

Sessano 7.647 - 7.647 - 0% - - 0%

Colle Breccione 11.594 - 11.594 - 0% - - 0%

Pesche - 7.258 8.658 8.658 19% 8.658 8.658 19%

Pescolanciano - 7.624 8.233 8.233 8% 8.233 8.233 8%

Fontecurelli - 3.296 - 4.406 34% 4.623 4.623 40%

Polverone - 4.519 - 5.365 19% 5.431 5.431 20%

Pescorvara 4.386 - 4.386 - 0% - - 0%

S. Domenico - 4.164 - 5.023 21% 5.512 5.512 32%

Santa Maria - 5.049 - 5.232 4% 5.532 5.532 10%

TOTALE [MW] - 31.91 - 36.917 16% 37.989 - 19%

«Caso Base»: withoutany smart functionality

«Programma P3»: first level of complexity (no communication with DG)

«Programma P3+IRE»:highest level of complexity(with communication with DG)

A relevant benefit can be extracted evenat less complex levels

NEW INCENTIVES FOR DSOs: output-based, selectivity

1. Observability of DG(and TSO-DSO data exchange)

2. Improved voltage control on MV networks

Benefits Less reserve due to better forecast; less control action due to better knowledge of system conditions � 5% reduction of fast resources, with a 50 M€/year saving

Higher hosting capacity (HC): deferred distribution investments

� based on experimental data, 16% HC increase, with a saving for each PS equal to 50 M€/year

Benefit/cost ratio

Under conservative assumptions: 3.3

Under conservative assumptions: 2.5

Output Rated power [MW] of DG connected to each MV/HV Primary Substation accurately observable

Transformation capacity [MVA]of each HV/MV Primary Station with improved voltage control

Yearly incentives

20 Eur/MW x PRES x months / 12(OSS-1: 20 sec. measurement)

250 Eur/MVA x SHV/MV

(REGV-1: not yet local control)


output-based incentives for smart functionalities are granted only in area with RPT > 1%

ERRA 08.05.18

05/05/2018

20

39

«Output-based» incentive regulation mechanisms: summ ary

OUTPUT DISTRIBUTION TRANSMISSIONSAIDI (cust. minutes lost)

Rewards/penalties (unplannedfrom 2000; planned from 2017)

-

ENS (energy notsupplied)

Rewards/penalties (2008)Mitigation (reduction rewards)

SAIFI+MAIFI (long and short interruptions)

Rewards/penalties (from 2008) -

GS* multiple interruptions

Customer Compensation (MV only) (2004)

Contribute to customercompensations

GS* very long interrupt. Customer compensation and penalties (2007, ren. 2016/17)

Contribute to customercompensations and penalties

GS* commercial quality 11 guaranteed standards (6 in 2000, then enlarged)

-

Smart grid Rewards only, 2 innovation outputs(observab.MV, hosting capacity)

Target capacityin critical T-grid sections

Additional capacity (in the limit of target capacity, CBA approach)

* GS: guaranteed standards

IMPORTANT STEPS AHEAD IN «SMART REGULATION»…

1. Output-based Regulation is the key step towards«smart regulation» as it combines efficiency and effectiveness


2. Selective priorities are needed to ensurevalue-for-money deployment of smart infrastructures

3. Cost/benefit analysis provides ground for system-wide incentives that transfer part of positive externalities to DSOs

4. Learning process is the way to build robust regulation, due to the uncertainty and risks of new technology applications

5. Flexibility is still a target that requires new balancingmarket rules, standardisation and advanced TLC services

ERRA 08.05.18

05/05/2018

21

FULLY IN LINE WITH ACER/CEER “BRIDGE TO 2025” VISION

Regulatory framework must adapt to the new challenges:

• Enable DSOs to take on the role of a neutral market facilitator and avoid DSO foreclose competition

• Accompany the development of new flexibility markets (on both sides: demand as well as distributed generation)

• Review distribution network tariff structure in order to develop focused incentives for smarter networks

• Clearly define DSOs’ and TSOs’ respective roles, establishing coordination requirements (system security)

• Enable customer awareness on his/her own energy footprint through smart metering data on consumption

Towards a future-proof power system…


INNOVATION: the relevance of independent regulation


Regulation should … endeavor to find approaches that

facilitate technological change, experimentation,

and the possibility of disruptive business models that

defy the classic approaches of incumbents.

As the technology evolves, the following issues merit

further evaluation and study. How smart is smart enough?

It merits investigation whether there are less expensive

alternatives that could provide nearly equal benefits.

OECD, DAF/COMP(2010)10, p.185

ERRA 08.05.18

05/05/2018

22

Thank you for your attention [email protected]

www.arera.itPlease visit:

Suggested reading on the Italian case (smart grid)CHANGING THE REGULATION FOR REGULATING THE CHANGEInnovation-driven regulatory developments in ItalyICER Distinguished regulatory scholar Award 2012http://www.iern.net/portal/page/portal/IERN_HOME/ICER_HOME/ABOUT_ICER/Distinguished_Scholar_Award_2012


Suggested reading on the Italian case (smart metering)SMART METERING: AN EVOLUTIONARY PERSPECTIVEGuidelines and lessons learnt from the Italian regulatory experienceHighly Acknowledged Paper ERRA regulatory research Award 2017http://erranet.org/knowledge-base/erra-regulatory-research-award/#winner2017

Annex 1Output-based regulatory incentives

for Quality of Service


05/05/2018

23

FUNDAMENTALS

• QoS is a worthy good

• The regulator wants to provide regulated companies with economic signals in order to deliver the optimal level of QoS

• It’s up to the regulated company to define the most effective actions (e.g. investments, resource organization, automation, etc) to minimise the total cost


Incentive regulation in practice is considerably more complicated than incentive regulation in theory.

Paul Joskow“Incentive Regulation in Theory and Practice:

Electricity Distribution and Transmission Networks”, CEEPR Paper 05-014, Sept. 2005

QoS INCENTIVE REGULATION IN PRACTICE

• Most EU regulators adopt an output-basedincentive regulation scheme fo QoS (esp. Continuity of Supply, both for distributionand transmission)

• CEER regularly publishes a Benchmarking of QoS Report (first comparison of incentive schemes for QoS in the 3rd CEER Report)

• Basic elements of most incentive schemes:


1. QoS metrics (clear and fair, trackable and auditable)

2. QoS baseline and improvement valuation (WtP, WtA)

3. Toolkit for managing uncertainty (acceptable risk)

05/05/2018

24

• Rewards and penalties are basedon net indicators

THE REGULATORY EXPERIENCE IN ITALY: 1. METRICS


• Internationally-used QoS indicators (SAIDI, SAIFI, MAIFI)

• Separately per kind of outage (unnotified vs notified)

• Separately per «district» (each internally homogeneous)

• Separately per voltage level and causeSAIDI-net

SAIFI-net

• Rewards and penalties are basedon net indicators

• Regulatory guidancefor recording QoSevents (incl. SCADA)

• Audits in field(sanctions)

THE REGULATORY EXPERIENCE IN ITALY : 2a. BASELINE


INCENTIVE REGULATION ADJUSTMENT FROM 1ST TO 2ND REGULATORY PERIOD

Referencestandards

0

10

20

30

40

50

60

70

80

90

0 1 2 3 4 5 6 7 8 9

CU

ST

OM

ER

MIN

IUT

ES

LO

ST

(n

et o

f in

terr

uptio

nsno

tattr

ibut

able

todi

strib

utio

nco

mpa

nies

)

TARGETS ACTUAL LEVELS

1998/99 1999/00 2000/01 2001/02 2002/03 2003/04 2004/05 2005/06 2006/07

1° REGULATORY PERIOD


REFERENCE STANDARDS

Referencestandards

0

10

20

30

40

50

60

70

80

90

0 1 2 3 4 5 6 7 8 9

CU

ST

OM

ER

MIN

IUT

ES

LO

ST

(n

et o

f in

terr

uptio

nsno

tattr

ibut

able

todi

strib

utio

nco

mpa

nies

)


1998/99 1999/00 2000/01 2001/02 2002/03 2003/04 2004/05 2005/06 2006/07



REFERENCE STANDARDS

Example: URBAN AREAS

Referencestandards

Referencestandards

0

10

20

30

40

50

60

70

80

90

0 1 2 3 4 5 6 7 8 9

CU

ST

OM

ER

MIN

IUT

ES

LO

ST

(n

et o

f in

terr

uptio

nsno

tattr

ibut

able

todi

strib

utio

nco

mpa

nies

)


1998/99 1999/00 2000/01 2001/02 2002/03 2003/04 2004/05 2005/06 2006/07



REFERENCE STANDARDS

Referencestandards

0

10

20

30

40

50

60

70

80

90

0 1 2 3 4 5 6 7 8 9

CU

ST

OM

ER

MIN

IUT

ES

LO

ST

(n

et o

f in

terr

uptio

nsno

tattr

ibut

able

todi

strib

utio

nco

mpa

nies

)


1998/99 1999/00 2000/01 2001/02 2002/03 2003/04 2004/05 2005/06 2006/07



REFERENCE STANDARDS

Example: URBAN AREAS

ReferencestandardsReferencestandards

),],([ avgENSttt PVATfQ −=±

Tt: targets (ex-ante, per district)

At: actual levels(ex-post, per district)

VENS: avg value of 1 hour interrupt. avoided for 1 kW(based on market survey - WTP)

Pavg: avg power of each district[kWavg]

Incentives for quality (year t)

At

Tt

05/05/2018

25

Ex-ante setting of standards, ex-post control of actual levels


Ex-ante for 4 years targets are set (baseline)

−= %2;1max

12

1

j

k

j LivPart

LivObαk: territ. density

)1(1,, jtjtj TT α−×= −

t0 1 2 3 4 … … … 12

Stdj,t [SAIDInet]

LivPart

LivObk

Tj,4

0TLivPart =

MaxQMin tj ≤≤ ,0%5 ,,, =⇒<− tjtjtj QAT

+×−= ∑

∈ 8760)( ,,,,

, ,domtjdomndomtjndom

Districtsj

tjtEncEncATQ tj

Ex-post actual levels are compared with relevant targets and reward/penalties applied

Aj,t

Qt [€]

Tj,t

-Min

+Max

[SAIDInet]

Unitary incentive/penalty [€/kWh-not-served]ex-ante based on WTP/WTA customer survey

FSR Cybersecurity 04.05.18

THE REGULATORY EXPERIENCE IN ITALY : 2b. VALUATION


DOMESTIC CUSTOMERS

Parameter Cdom

BUSINESS CUSTOMERS

Parameter Cndom

(BEST DISTRICTS) Below national ref.

7.2 €/kWh-ENS 14.4 €/kWh-ENS

(MEDIAN) From 1x to 3x nat.ref.


(WORST DISTRICTS) Above 3x national ref.


national reference (SAIDI-net): urban 25 min/cust/year, rural 60 min/cust/year

Values based on a Willingness-to-Pay (WtP) survey, commissioned by AEEG. For further details: • Bertazzi, Fumagalli, Lo Schiavo, The use of customer outage cost surveys in policy decision-making:

the italian experience in regulating quality of electricity supply, CIRED (2005) paper n. 300• CEER, Guidelines of Good Practice on Estimation of Costs due to Electricity Interruptions and Voltage

Disturbances, Ref. C10-EQS-41-03, December 2010.

VOLL: UNIT VALUE OF AVOIDED ENERGY NOT SUPPLIED (ENS)

05/05/2018

26

THE REGULATORY EXPERIENCE IN ITALY : a practical cas e


THE REGULATORY EXPERIENCE IN ITALY : RESULTS


• Huge improvement in QoSover 4 reg.periods

• Limited impact on tariff(less than 3 €/cust/year)

• Resilience issues in last years (out of DSO control)

• «Convergence» effectto recover existing gaps

among districts of the sameterritorial density

05/05/2018

27

Annex 2Smart metering (from 1st to 2nd generation)


ELECTRICITY SMART METERING: a synopsis for Italy

54

Italy has been the first Country with full roll-out of smartmetering:

• More than 35 million customers with smart meters

• More than 100 distribution companies operating their ownsmart metering systems (even the smallest ones)

• More than 400 million readings per year remotely operated

• More than 10 million operations per year for remote customer management

• More than 28 million customers with Time-of-Use tariff(households: 2 price-bands; small business: 3 bands)

• More than 15 years of experience in smart metering

Luca Lo Schiavo, AEEGSI (Italy)ERRA 08.05.18

05/05/2018

28

ELECTRICITY SMART METERING: European developmentson

BenchmarkingReport on

Smart metering

Eur. CommissionRecommendation

148/2012

EuropeanDirective

2009/72/CE

2009 2012

EuropeanDirective

2012/27/UE

2014

«WinterPackage»

2016

The pioneering Italian case for smart metering (launched in 2001 on initiative of three distributors, among which Enel)

has been a benchmark for Europe

55Luca Lo Schiavo, AEEGSI (Italy)ERRA 08.05.18

ELECTRICITY SMART METERING: Italy was a pionieer

56

Italy: Decree MiSE 60/2015

Periodical verification forlow-voltage electricity meters: 15 years (“MID meters”)

Investments for smart metering

Italy: 97 euro/point

France*: 135 euro/point

G. Britain**: 161 euro/point

Finland: 210 euro/point

Netherlands**: 220 euro/point

Sweden: 288 euro/point

Spain: not available

Source: Eur.Commission, SWD(2014) 189 final* roll-out on going ** roll-out on going, joint gas/electricity

Luca Lo Schiavo, AEEGSI (Italy)ERRA 08.05.18

05/05/2018

29

ELECTRICITY SMART METERING IN ITALY: benefits

57

1G METERING TARIFF (Euro/point, current values 2004-2016)

Luca Lo Schiavo (AEEGSI)Eurelectric 28.09.17

References to tariff decisions:Del. 5/04; Del. 275/06

Del. 348/07; Del. 199/11Del. 654/15

BENEFITS 1G

• Time-of-use energy pricing (mandatory)

• Monthly readings

• Almost no estimatedbills

• Easy switch (spot reading)

• Minimum “vital” service

• better service(prompt reconnection)

• energy balance in LV networks

Time-of-Use price (LV customers in Universal Supply Regime)


• In the free market customers can be supplied with «flat» prices(no ToU mandatory requirement for energy prices).

• Consumptions are however read separately for timebands evenfor free market customers (DSO is the metering operator).

Universal Supply Regime: Household & Small business customersthat do not choose their own supplier

USR energy prices regulated by AEEG are differentiated per time-bands on mandatorybasis:

• Household: 2 bandspeak vs. (mid-level+off-peak)

• Small business: 3 bands

M T W T F S S

peak

off-peak

mid-level

7.008.00

19.00

23.00

0.00

24.00

ERRA 08.05.18

05/05/2018

30

SMART METERING: did the “1 st generation” work?

59Luca Lo Schiavo (AEEGSI) 59

WHAT WE GOT OUT OF 1G … …AND WHY

High availability96% of remote readings properly

accomplished (end-to-end)

Very good reliabilityNo relevant cases for meter substitution

due to faults

Limited cases of interference

between PLC and inverters

PV inverters EM emissions reduce data

acquisition (prosumers counting <2%)

1 channel only, not available for

real-time data messages

Communication channel (via PLC band A)

dedicated exclusively to validated data

Very limited use for voltage dataBuffer for interruption events too shortVoltage measurement not compliant with EN 50160

No interoperability with 3rd party

In-Home Devices

No message encryption (launched in 2001),

non disclosed protocol (cyber-sec. reasons)

Slow reconfiguration process Overall firmware download: ≈9 months

ERRA 08.05.18

SMART METERING: “future-proof” design criteria (416/2015)

Ensure electro-magnetic immunity from disturbances

Allow for multi-channel communication (PLC and radio)

Ensure advanced cyber-security measures (eg. data encryption)

Improve integration between smart metering and smart grid roll-out

Minimise constraints of backward compatibility for next generation

Minimize occasions for system reconfiguration (firmware download)

Ensure hardware independency for any function

Keep separate communication resources for remote reading and for energy efficiency

Ensure interoperability with third party in-home devices

Ensure interchangeability with metering systems of other DSOs


05/05/2018

31

SMART METERING: system architecture from “1G” to “2G”

61

PLC (band A)

MV/LVtrasf.

Smart meters

Public TLC

Data Concentrator

AMMsystem(DSO)

web

Retail suppliers

Customisation (for suppliers)

New concept: distinctionbetween

• Chain 1: from meter to retailer (validated data), similar to 1G, empowered

• Chain 2: from meter to customer (directly, realtime data but non validated): new

Third partyin-home devices

“Chain 2”:instantaneous data from meterto customer

PLC (band C)

Remotereading andcustomer management

Reconfiguring the system(download)

Telemonitoring(upload)

Upgrade LTE

+backup RF

Luca Lo Schiavo (ARERA)ERRA 08.05.18

CHAIN 1: meter reading for billing, from 1 st to 2nd generation

Luca Lo Schiavo (AEEGSI) 62

1st

GENERATIONCapacity 1G Meter

reading1G reading content

Default.supplbilling

Households Any(typic 3kW)

Monthly 3 timebands 2 prices(mandatory)

Small business

Up to 55kW Monthly 3 timebands 3 prices(mandatory)

Medium business

Above 55 kW Monthly 96 quarter-hours per day

N/A (only free market)

2nd

GENERATION)Capacity 2G Meter

reading2G reading content

Default.supplbilling

All customers Any Daily 96 quarter-hours per day

N/A (only free market)

1G: fixed time bands, preloaded in the meter � need for massive reconfiguration

2G: time bands directly customizable by suppliers � no need for massive reconfig.

ERRA 08.05.18

05/05/2018

32

SMART METERING: data collected, from 1 st to 2nd generation

Metering data 1G 2G

Active energy withdrawn 3 values per month [Wh]* every 15 min [Wh]

Active energy injected 3 values per month [Wh]* every 15 min [Wh]

Reactive energy withdrawn 3 values per month [Wh]* every 15 min [VARh]

Reactive energy injected not available every 15 min [VARh]

Active power withdrawn peak in 15 min [W](max value in the month)

peak in 15 min [W](max value in the day)

Active power injected not available peak in 15 min [W](max value in the day)

Instantaneuos power 2 sec [W] (last value on display)

up to 1 sec [W] (available via chain 2 in realtime)

Max/min voltage some measure availablebut non EN50160-compliant

per week [V](EN50160-compliant)

Interruption event

(voltage below 5% of Un)

(in practice not used

because of too short buffer)

on event occurrence: start, end, duration [min.ss]

63(*) Only for 1% of LV customers (those with rated capacity >55kW): 1 value every 15 min

SMART METERING: 2nd generation concept (decision 87/2016)

Luca Lo Schiavo (AEEGSI) 64

«Chain 1»

• Purpose: billing and network managem’ t

• Validated data, SLA

• Daily collection

• Operated by DSO

• Back-up channel

«Chain 2»

• Purposes: customerawareness and valueoffering for suppliers

• Real-time, notvalidated data

• Continuous flow

• Interoperable with 3rd party IHDs

• No back-up channel

ERRA 08.05.18

Under development

05/05/2018

33

CHAIN 2 (new with 2G): interoperable In-Home Device

Luca Lo Schiavo (ARERA) 65

• IHDs developed by third parties (integrated with home ecosystem)

• Standard communication protocol(independent of manufacturers)

• Developed by CEI (standardisationbody) with industry consensus: July-2017 (CEI TS 13-82/85)

• Start with physical layer PLC in CENELEC “band C”

• Monitoring phase under course,very good performance (end-to-end)

• Release 2.1: cooperation with Agcomon further options for chain 2 -consultation paper ARERA n.245/18

wireline/contact

wireline/wireless

Generic

Frequent- 1 retry

Rare - 4 retry

Avg 98,79% 98,23%

95 percentile 97,28% 100 %

99 percentile 50,64% 0 %% success in daily samples

ERRA 08.05.18

CHAIN 2 (new with 2G): real-time communication

• “frequent asynch data”: level of used power capacity changes very rapidly; limitation of contractual power can lead to disconnection

• Requirements: sampling up to every 1 second “frequent asynchdata” are not to be stored (strong privacy issues)

L.Lo Schiavo, A.Pitì (AEEGSI, IT)

1

2

3

3.3

PUSH PUSH PUSH PUSH PUSH PUSH

PUSH ALERT

P [kW]

T [sec]

011001…101

PUSH

66ERRA 08.05.18

05/05/2018

34

SUMMARY: from 1st to 2nd generation of smart meterin g

1G field experience including trials on IHDs

Functional requirements (decision 87/16) after public consultation(416/15)

Rules for cost allowance based on sharing matrix(decision 646/2016)

Roll-out approval: Regulator’s decision after public consultation of distribution company’s roll out plan(decision 222/2017)

Analysis of technical evolution for a more advanced version (“2.1”)of smart meter in cooperation with telecom Regulator (AGCOM)(consultation 245/2018)

Luca Lo Schiavo, AEEGSI 67ERRA 08.05.18

TRASPOR-TATION

Vertical regulation vs cross-sector innovation

68

ENERGY

TELECOM

ENERGY

Luca Lo Schiavo, AEEGSIERRA 08.05.18

05/05/2018

35

COOPERATION AMONG SECTORIAL NRAs: energy and TLC

AEEGSI contribute to AGCOM survey on M2M services

In particular, AEEGSI underlined the relevance of latency and proposed the following classification:

A.monitoring: remote data collection and configuration, without delay requirements

B.control: data collection and implementation commands with low delay requirements (1s)

C. protection: data collection and immediate reaction in difficult circumstances where speed is essential for safety or security reasons (<1s)

Smart meters(traditional)

Smart meters(advanced)

Smart grids(most

functionalities)

Smart grids(faster

functionalities)

69

The AEEGSI contribution to AGCOM survey has been published: www.autorita.energia.it/allegati/inglese/457-14eng.pdf (in English)

Luca Lo Schiavo, AEEGSIERRA 08.05.18

SMART GRID and SMART METERING: a synthesis


Smart grid Smart metering

BenefitsRES integrationSystem security

Cost efficiencyRetail competition

Latency issuesDown to 100 ms

(fast fault selection)Close to 1 s

(non validated data)

Size(number of points, IT)

0,6 M prosumers0,4 M MV/LV substations

35 M customers

Asset managementTLC layer relatively

independentTLC layer intimately

embedded

Cooperationenergy/TLC NRAs

Access to passive infrastructure (OF)

Technology neutralityvs best performance

ERRA 08.05.18

05/05/2018

36

Annex 3Electro-mobility and recharge infrastructure


EV recharge (1/2) (acknowledgement: ABB)

72

+

+

Parking- Parking garage- Residential- Office

Commercial- Daily shopping- Lunch- Medium distance trip

Highway- Long distance trip

Typical time spent 2-10 hours 30-90 min 10-20 min.

Power level < 20 kW ~50 kW 50-300 kW

Typical Performance

2010-2017 Small EV(~30 kWh)

Full charge Full charge To ~80% in 20 min.(Max. 50 kW)

>2018 Long range EV(>80 kWh)

50-100% charge 50-100% charge To 80% in 15 min.(150 - 300 kW)


05/05/2018

37

EV recharge (2/2) (acknowledgement: ABB)

73

2x4x

8x

First priority for operators: more outlets, more locations


EV recharge (in public places): main relationships

74

Charging station

Smartmeter

Distribution gridElectric vehicles(EVs)

CHARGING POINT OPERATOR (CPO)

ELECTRICITYSUPPLIER

EV DRIVERS

ELECTRICITYDISTRIBUTION COMPANY

1a

2

3

Displays (can be set to zero at eachtransaction)

4

CONTRACTUAL RELATIONSHIPS

1a. EV Driver – CPO (directly)

2. CPO – Elect.Supplier

3. Elect.Supplier – DSO

4. DSO – CPO (only for connection)

Chargingpoint

Chargingpoint

Chargingpoint

Other usesof electricity

• Interoperability issues are the focus of SGEMS (Sub Group for Electro-Mobility Services) within the Sustainable Transport Forum

Point of delivery

MOBILITY SERVICE PROVIDERS (MSP)

1b

1b

1b. EV Driver – CPO (intermediated by MSP)


05/05/2018

38

EV recharge at home: removing old «progressive» tariff

75

Yearly expenditure for network tariff [€]

0

50

100

150

200

250

300

0 900 1.800 2.700 3.600 4.500 5.400

Household consumption (P=3 kW)

D2-2015

D3-2015

TD

Resident(2015)

Non resident(2015)

New tariff(2017)

• In Italy a non-linear, progressive tariff for household was introduced in ’70s for energy efficiency

• The progressive tariff proved to be a barrier for electro-efficient new technologies

• ARERA launched a reform for a new cost-reflective tariff for household

• New network tariff has been fully introduced from 1.1.17

• New tariff is largely capacity based (see CEER guidelines, Ref. C16-DS-27-03, 23 January 2017)


EV recharge prices: an European example

76

€/kWh is a too simple approach:

price of EV charge is due to many factors, not only energy but also number of transactions, speed of charge, time-of-use - Further, price will

be influenced by demand aggregation (role of MSPs as wholesale buyer)

…prices charged by the operators of recharging points accessible to the public are reasonable, easily and clearly comparable, transparent and non-discriminatory

AFID, 4(10)


Documents

LoSchiavo ERRA Budapest 08May rev0€¦ · Staff of the Regulatory Authority have the duty to disclaim in public that only personal opinions are presented when speaking in public