23
Practical Coordination of DERs at Scale using Packetized Energy Management Paul Hines, University of Vermont, Packetized Energy Collaborators: Jeff Frolik, Mads Almassalkhi Sumit Paudyal (MI.Tech), Adil Khurram, Mahraz Amini, et al.

Practical Coordination of DERs at Scale using Packetized ... · 300 electric hot water heaters 11 0 50 100 150 200 0 200 400 600 800 1000 Power (kW) 0 50 100 150 200 −5 0 5 T −

  • Upload
    others

  • View
    0

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Practical Coordination of DERs at Scale using Packetized ... · 300 electric hot water heaters 11 0 50 100 150 200 0 200 400 600 800 1000 Power (kW) 0 50 100 150 200 −5 0 5 T −

Practical Coordination of DERs at Scale using

Packetized Energy Management

Paul Hines, University of Vermont, Packetized Energy

Collaborators: Jeff Frolik, Mads AlmassalkhiSumit Paudyal (MI.Tech), Adil Khurram, Mahraz Amini, et al.

Page 2: Practical Coordination of DERs at Scale using Packetized ... · 300 electric hot water heaters 11 0 50 100 150 200 0 200 400 600 800 1000 Power (kW) 0 50 100 150 200 −5 0 5 T −

2

©BP, 2018

Non-Hydro (wind/solar) energy consumption

Million tonnesoil eq

uiv.

Page 3: Practical Coordination of DERs at Scale using Packetized ... · 300 electric hot water heaters 11 0 50 100 150 200 0 200 400 600 800 1000 Power (kW) 0 50 100 150 200 −5 0 5 T −

And probably a lot more to come

3

0

10

20

30

40

50

60

2010 2020 2030 2040 2050

Actual GtCO2e VRE Other

Without mitigation

With mitigation from renewables

With mitigation from efficiency, reforestation,

etc.

After mitigation

~10 TW of renewables + ~8-10 TW of flexibility. Mean load today ~ 3 TW

Perhaps2 TW of flexible supply5 TW of storage + 2 TW of flexible demand.

Page 4: Practical Coordination of DERs at Scale using Packetized ... · 300 electric hot water heaters 11 0 50 100 150 200 0 200 400 600 800 1000 Power (kW) 0 50 100 150 200 −5 0 5 T −

2 TW of flexible demand

4

1 TW of flexible large C&I load 1 TW of flexible

small c&R load ~2 billion devices

Page 5: Practical Coordination of DERs at Scale using Packetized ... · 300 electric hot water heaters 11 0 50 100 150 200 0 200 400 600 800 1000 Power (kW) 0 50 100 150 200 −5 0 5 T −

So:

How do we coordinate billions of connected devices to solve grid

problems?

5

Page 6: Practical Coordination of DERs at Scale using Packetized ... · 300 electric hot water heaters 11 0 50 100 150 200 0 200 400 600 800 1000 Power (kW) 0 50 100 150 200 −5 0 5 T −

Let’s borrow ideas from something that already connects billions of devices

6

Page 7: Practical Coordination of DERs at Scale using Packetized ... · 300 electric hot water heaters 11 0 50 100 150 200 0 200 400 600 800 1000 Power (kW) 0 50 100 150 200 −5 0 5 T −

Packetization

7

Randomization

Page 8: Practical Coordination of DERs at Scale using Packetized ... · 300 electric hot water heaters 11 0 50 100 150 200 0 200 400 600 800 1000 Power (kW) 0 50 100 150 200 −5 0 5 T −

Simple state machines for EV charging

8Rezaei, Frolik, Hines, “Packetized Plug-in Electric Vehicle Charge Management”, IEEE TSG, 2014

EV requests charge based on the state of the vehicle.

Urgent/non-urgent, response to prior charges

Page 9: Practical Coordination of DERs at Scale using Packetized ... · 300 electric hot water heaters 11 0 50 100 150 200 0 200 400 600 800 1000 Power (kW) 0 50 100 150 200 −5 0 5 T −

How much does it cost to avoid fancy optimization?

9Rezaei, Frolik, Hines, “Packetized Plug-in Electric Vehicle Charge Management”, IEEE TSG, 2014

Page 10: Practical Coordination of DERs at Scale using Packetized ... · 300 electric hot water heaters 11 0 50 100 150 200 0 200 400 600 800 1000 Power (kW) 0 50 100 150 200 −5 0 5 T −

What about a water heater?

10

TOO HOT

TOO COLD

Turn off

Turn on

Stochastically request packets of energy at a rate proportional to the need for energy

Tem

per

atur

e

Virtual Battery server replies to requests based on grid conditions

Virtual Battery solves grid problems

Page 11: Practical Coordination of DERs at Scale using Packetized ... · 300 electric hot water heaters 11 0 50 100 150 200 0 200 400 600 800 1000 Power (kW) 0 50 100 150 200 −5 0 5 T −

300 electric hot water heaters

11

0 50 100 150 2000

200

400

600

800

1000

Power

(kW

)

0 50 100 150 200−5

0

5

T−

Tset(

◦C)

Time (min)

wind productiondemand

Start choreography

Demand

Pow

er (k

W)

RenewableSupply

Time (minutes)

Almassalkhi, Frolik, Hines, “Packetized energy management...,” ACC, 2017

Page 12: Practical Coordination of DERs at Scale using Packetized ... · 300 electric hot water heaters 11 0 50 100 150 200 0 200 400 600 800 1000 Power (kW) 0 50 100 150 200 −5 0 5 T −

12

Conventional thermostat (long on/off times)

Time (minutes)

Black = Device is OFF

Packetized! (multiple short on/off times)

De

vic

e ID

De

vic

e ID

Conventional vs. Packetized

Page 13: Practical Coordination of DERs at Scale using Packetized ... · 300 electric hot water heaters 11 0 50 100 150 200 0 200 400 600 800 1000 Power (kW) 0 50 100 150 200 −5 0 5 T −

Diversity makes stuff work better

13Almassalkhi, Duffaut-Espinosa, Hines, JFrolik, Paudyal, Amini, “Asynchronous coordination of DERs...,” in Energy Markets and Responsive Grids, Springer, 2018

VPP

Pow

er (M

W)

Time (minutes)

Balancing signalEVs + Batteries + WHsWater heaters only

Page 14: Practical Coordination of DERs at Scale using Packetized ... · 300 electric hot water heaters 11 0 50 100 150 200 0 200 400 600 800 1000 Power (kW) 0 50 100 150 200 −5 0 5 T −

At scale?

5000 water heaters

14

Page 15: Practical Coordination of DERs at Scale using Packetized ... · 300 electric hot water heaters 11 0 50 100 150 200 0 200 400 600 800 1000 Power (kW) 0 50 100 150 200 −5 0 5 T −

15

Page 16: Practical Coordination of DERs at Scale using Packetized ... · 300 electric hot water heaters 11 0 50 100 150 200 0 200 400 600 800 1000 Power (kW) 0 50 100 150 200 −5 0 5 T −

How do we make it work in the real world?

16

Step 1: Start a company

Step q: Partner with utilities

Step n: Build a device and a lot of software

Page 17: Practical Coordination of DERs at Scale using Packetized ... · 300 electric hot water heaters 11 0 50 100 150 200 0 200 400 600 800 1000 Power (kW) 0 50 100 150 200 −5 0 5 T −

Step π: build a business case

17

Wholesale energy arbitrage~$30/device/year Time

Pric

e Buy Sell

Avoided generation capacity (ICAP)~$100/device/year

Avoided T&D CapEx~$20/device/year

Frequency regulation~$30/device/year

Supply Demand

Total = $200/device/year

Page 18: Practical Coordination of DERs at Scale using Packetized ... · 300 electric hot water heaters 11 0 50 100 150 200 0 200 400 600 800 1000 Power (kW) 0 50 100 150 200 −5 0 5 T −

18

In part because of renewables,electricity is usually super cheap

83% of hours <4c/kWh

Page 19: Practical Coordination of DERs at Scale using Packetized ... · 300 electric hot water heaters 11 0 50 100 150 200 0 200 400 600 800 1000 Power (kW) 0 50 100 150 200 −5 0 5 T −

Which means that utilities should do this

19

Page 20: Practical Coordination of DERs at Scale using Packetized ... · 300 electric hot water heaters 11 0 50 100 150 200 0 200 400 600 800 1000 Power (kW) 0 50 100 150 200 −5 0 5 T −

LoadShaper: Automated Energy Arbitrage

20

Page 21: Practical Coordination of DERs at Scale using Packetized ... · 300 electric hot water heaters 11 0 50 100 150 200 0 200 400 600 800 1000 Power (kW) 0 50 100 150 200 −5 0 5 T −

PeakCrusher: Automated Peak Management

21

0

20

40

60

80

Pow

er (k

W)

Set pointActual

14:00 15:00 16:00 17:00 18:00 19:00 20:00 21:00 22:00 23:00Time Aug 06, 2018

0

100

200

LMP

($/M

Wh)

Pre-positioning 6-hour peak eventPost-eventrecovery

Return to LoadShaper

0

20

40

60

80

Powe

r (kW

)Set pointActual

14:00 15:00 16:00 17:00 18:00 19:00 20:00 21:00 22:00 23:00Time Aug 06, 2018

0

100

200

LMP

($/M

Wh)

Page 22: Practical Coordination of DERs at Scale using Packetized ... · 300 electric hot water heaters 11 0 50 100 150 200 0 200 400 600 800 1000 Power (kW) 0 50 100 150 200 −5 0 5 T −

22

The most important step: build a great team

• Advisors

BOB ZULKOSKIVermont WorksCo-founder Greenlots

LAURY SALIGMANVernal VenturesEnvironmental leader

MICHAEL BEERMarketing expert

HEMAL SHAHFacebook/InstagramFocus: UI/UX

BJOERN SIMONAltair EngineeringFocus: Back-end/ Architecure

SCOTT JOHNSTONECEOEnergy leaderFormer CEO of $100M Vermont Energy Investment Corp. (VEIC)

MADS ALMASSALKHICO-FOUNDERSerial entrepreneurSaaS, real-time controls Michigan PhD

JEFF FROLIKCO-FOUNDERCommunication systems, patentsMichigan PhD

PAUL HINESCO-FOUNDERElectricity systems/software, electricity policy/marketsCarnegie Mellon PhD

ANDREW GIROUXIoT/HardwareFormerly: UTC, GMBuilds Evs for fun

KATE DESROCHERSInitiative Lead, MBAFormerly: energy consultant

JOHN SLINKMANSaaS/AWSFormerly: Director at POLCO

FORREST WALLACEEmbedded SystemsFormerly: DEKA

Page 23: Practical Coordination of DERs at Scale using Packetized ... · 300 electric hot water heaters 11 0 50 100 150 200 0 200 400 600 800 1000 Power (kW) 0 50 100 150 200 −5 0 5 T −

But then again, if we can do this, it’s worth it.

23

0

10

20

30

40

50

60

2010 2020 2030 2040 2050

Actual GtCO2e VRE Other

Without mitigation

With mitigation from renewables

With mitigation from efficiency, reforestation,

etc.

After mitigation

~10 TW of renewables + ~8-10 TW of flexibility. Mean load today ~ 3 TW

Perhaps2 TW of flexible supply5 TW of storage + 2 TW of flexible demand.