multi-level management 4200 - 465 porto of electricity storage©2006 3 introduction • energy...

Campus da FEUPRua Dr. Roberto Frias, 3784200 - 465 PortoPortugal

T +351 222 094 000F +351 222 094 [email protected]

© 2006

MultiMulti--level Management level Management of Electricity Storageof Electricity Storage

J. A. Peças Lopes([email protected])

23 April 2008 - BrugesSmart and Sustainable Electricity Systems

2© 2006

Introduction - New Paradigmas

• The vision of the future

From the SmartGrids document - EU Commission

Storage is a key issue in this new vision!

3© 2006

Introduction

• Energy storage is a way to transfer energy!

• Storage management strategies can be oriented by:– pure economic reasons

– technical reasons, to cope with global system constraints or network constraints.

• The most promising technologies are based on:– Electrochemical devices: batteries, Flow batteries; Fuel Cells

– Mechanical devices: flywheel, pumped hydro, compressed air

– Electronic devices: super-capacitors, SMES

– Thermal storage

4© 2006

Electric storage on a time frame basis

• Short term storage– Short-term storage provides rated power for a few seconds

up to a minute. Can be used to provide power quality to energy.

• Medium term storage– Medium-term . Applications for this type of storage capability

include renewable energy management, customer energy management, area control/frequency regulation, and rapid reserve management.

• Long term storage– Long-term storage typically bulk energy storage used to take

advantage of the energy price difference between peak and off-peak periods.

5© 2006

Multi-level Storage Management

• Storage should be managed at different levels:

System level

Distribution network level

Low voltage distributionnetwork

Power quality and reliabilityAvoid branch congestion

Peak shavingReserves management

Generation level Market participation

6© 2006

The need for more centrally managed storage systems

• The Iberian peninsula electric power system

7© 2006


• Deployment of wind generation in the Iberian Peninsula -Evolution of the wind power installed capacity in Spain andPortugal

0

5.000

10.000

15.000

20.000

25.000

MW

2000 2001 2002 2003 2004 2005 2006 2007 2011

Installed Wind Power Capacity (MW) Installed WindPower Capacity(PT)Installed WindPower Capacity(SP)

4500 MW

20000 MW

8© 2006


• Large scale wind deployment requires central storagemanagement, involving several storage facilities:– To tranfer renewable energy– To deal with reserves management

Prospective generation allocation in a winter windy wet day (2011)

Wind Generation of 14 th February 2007Source: REN (Portuguese TSO); http://ww.ren.pt

Storage is neededReservemanagement

9© 2006

Wind ParkWater Pump

Station

Hydro GeneratorLower Reservoir

Upper Reservoir

Combined storage operation strategies - definition and modelling

10© 2006

Output Power

HydroGeneration

Pump Consumption

Wind Generation

Output Bus

Wind GeneratorBus


11© 2006

•• NeededNeeded data:data:

– , reservoir storage capacity;

– , efficiency of water pump station and water pipes network;

– , efficiency of the water reservoir and hydro generator;

– and , initial and final levels of the reservoir;

– and , lower and upper production power limits of the hydro generator;

– and , lower and upper physical power limits of the pump station, respectively.

UE

ηp

ηh

1espE +1

espnE

LPh UPh

LPp UPp

data about the remuneration of energy delivered to the grid and costs of storing energy for each time interval of the period under analysis are required.


12© 2006

• An optimization problem for a given time horizon (eg: 24h)

Where the inputs of the problem are:the hourly wind power generation forecastsa vector of hourly active power pricesa vector of pump operation cost for the same period. Under market environments a forecast of the energy prices for each time step needs to be obtained for the same period.


13© 2006

= 1,....,i n

( )−∑ i i ii

c P cp PpMax.Output Wind-Hydro Generationand Pump Costs

= +i i iP P w P h Output Bus Balance

= + +i i i DLiPv Pw Pp P Wind Bus Balance

ηη+

⎛ ⎞= + −⎜ ⎟

⎝ ⎠1

ii i p i

h

PhE E t Pp Energy Reservoir Balance

≤ ≤0 MiE E Stored Energy Reservoir Limits

( )≤ + ≤m Mi iPg Pw Pp Pg

η⎛ ⎞≤ ≤ ⎜ ⎟⎝ ⎠

min ,m M ii h

EPh Ph Pht

≤ ≤m MiPp Pp Pp

Hydro Generation Limits

Wind Park Generation Limits

Pump Station Limits

≤ ≤L Ui i iP P P Output Generation Limits

=1 1e s pE E

+ +=1 1e s p

n nE E

Initial and Final Reservoir Conditions


14© 2006

0 1 0 2 0 3 0 4 00

2

4

6

8

Win

d P

ower

[MW

]

h o u r s

Forecasted Available Wind Power

0 1 0 2 0 3 0 4 05 0

6 0

7 0

8 0

9 0

1 0 0

1 1 0

Pric

e [€

/MW

]

h o u r s

Active Power Price(Decreto-Lei Nº 168/99 and Decreto-Lei Nº 339-C/2001)

PgM

[MW]PhM

[MW]PpM

[MW]ηL cp

[€/MWh]11 2 2 0.75 2

EM

[MWh]E1

esp

[MWh]En+1

esp

[MWh]PL

[MW]PU

[MW]

22 0 0 0 6.0


Network limit

15© 2006

Storage Levels in the Upper Reservoir

In low price periods, the upper reservoir always increases its storage level. In the second part of first high price period, the storage can be increased because restrictions in the output generation are reached.

0 1 0 2 0 3 0 4 0 5 00

5

1 0

1 5

2 0

2 5

h o u r s

Res

ervo

ir [M

Wh]

R e s . L e v e l


16© 2006

Wind and Hydro Active Power Outputs

0 1 0 2 0 3 0 4 0 5 00

2

4

6

8

1 0

h o u r s

W a

nd H

Pow

ers

[MW

]

WH

The hydro generator (dotted line) only operates in high price periods.


17© 2006

Managing Multiple storage systems

Solve a daily wind-hydro plant operation optimization problem.

Input data:- network data - wind parks power production- hydro with pumped storage stations data- market issues

Output data:- hourly active power to be generated by the hydro and wind generators for 24 h.

- storage levels and the pump operational strategy in the period

- combined wind-hydro economic gains

18© 2006

Multi-microgrids – MV distribution network of the future

• Microgrids

– DFIM– Fuel Cell– Microturbin

e– Storage

(VSI)

– PV

• Storage device-VSI

• Large DFIM

• Hydro

• CHP

• Small Diesel

• SheddableLoads

HV Network

VSI

Diesel

DFIM

MicroGrid

MicroGrid

MicroGrid

CapacitorBank

Hydro

CHP

SheddableLoads

Storage

19© 2006

Multi-microgrids – MV distribution network of the future

• Storage devices in MV (and LV) networks can be used for:

– Peak shaving purposes (controled from a local control structure – CAMC)

20© 2006 20

PV

Wind Gen

MicroGrid: A Flexible Cell of the Electric Power System

Microturbine

Fuel Cell

Storage DeviceMGCC

MC

MC

MC

MC

MC

LC

LC

LC

LC

LC

MG Hierarchical Control:

• MGCC, LC, MC

• Communication infrastructure

21© 2006

MicroGrid and or Multi-Microgrid Operation andControl

• Storage device’s active power output is proportional to the MG frequency deviation (droop control) Islanding operation is possible

Correct permanent frequency deviations during islanding operation

22© 2006

Results from Simulations (in islanding conditions)

• Frequency and VSI/Storage device Active Power output

0 50 100 150 200 25049.2

49.4

49.6

49.8

50

50.2

Freq

uenc

y (H

z)

0 50 100 150 200 250-20

-10

0

10

20

30

40

50

Time (s)

VSI A

ctiv

e Po

wer

(kW

)

23© 2006

Conclusions

• Intelligent management of storage will allow the increase of renewable power sources penetration in the power system;

• Arbitrage function provided by storage could annihilate most of extra network reinforcement and reduce balancing expenses;

• Storage active control will allow the development of the Microgrids concept and will bring more reliability to final consumers;

• Plug-in hybrid vehicles will contribute to management of electricity demand with major impacts on network control.

multi-level management 4200 - 465 porto of electricity storage©2006 3 introduction • energy...

Documents