GROWDERS & PLATOS, session 1 Optimal use of storage systems Petra de Boer Project coordinator GROWDERS Roger Cremers Developer PLATOS tool Gabriël Bloemhof Consultant decision making models 14 February 2011 16:00 – 17:30 17 February 2011 16:00 – 17:30
Development and use of Renewable Energy Sources is one of the key elements in European Electricity Research. However, connecting energy sources such as photovoltaics and wind turbines to the electricity grid causes significant effects on these networks. Bottlenecks are stability, security, peaks in supply & demand and overall management of the grid. Energy storage systems provide means to overcome technical and economic hurdles for large-scale introduction of distributed sustainable energy sources. The GROW-DERS project (Grid Reliability and Operability with Distributed Generation using Flexible Storage) investigates the implementation of (transportable) distributed storage systems in the networks. The project is funded by the European Commission (FP6) and the consortium partners are KEMA, Liander, Iberdrola, MVV, EAC, SAFT, EXENDIS, CEA-INES and IPE.
In this project 3 storage systems (2 Li-ion battery systems and a flywheel) have been demonstrated at different test locations in Europe. Additionally, a dedicated software tool, PLATOS (PLAnning Tool for Optimizing Storage), has been developed by KEMA to optimize the energy management of electricity networks using storage. For each network, the location, size and type of storage systems is evaluated for all possible configurations and the most attractive option is selected.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
GROWDERS & PLATOS, session 1
Optimal use of storage systemsPetra de Boer Project coordinator GROWDERS
Roger Cremers Developer PLATOS tool
Gabriël Bloemhof Consultant decision making models
14 February 2011 16:00 – 17:30
17 February 2011 16:00 – 17:30
2
Introduction Workshop Leaders
Petra de Boer Roger Cremers Gabriël Bloemhof
Energy storage
Electric vehicles
Smart Grids
Power Factory
Developer PLATOS
Energy systems
Grid integration
Optimization
3
Agenda
14 February 2011, 16:00 – 17:30– Introduction of GROWDERS project
– Benefits of grid connected storage
– 4 field tests in Europe using storage
– Decision making model in PLATOS
17 February 2011, 16:00 – 17:30 – Main features of PLATOS
– Demonstration with the PLATOS tool
– Examples, results
4
Introduction of GROWDERS
EC funded under 6th Framework Programme Coordinated by KEMA
Goal: To demonstrate the technical and economical
possibilities of existing electricity storage
technologies.
– Realization of Transportable Flexible storage systems
– Realization of an Assessment tool for optimal distribution network
management
– Description of conceptual directions for EU regulatory framework
Grid Reliability and Operability with Distributed Generation using Transportable Storage
Zutphen, The Netherlands Zamudio, Spain Chambery, France Mannheim, Germany
12
Test site Bronsbergen (Zutphen)
13
Test site Bronsbergen (Zutphen)
LS RAIL 40.401.00
-150.20
GE HKL0.401.00
-150.20
Sectie Kast 20.401.00
-150.20
Sectie Kast 1 0.401.00
-150.20
MS rail 4 10.001.000.00
V
VDC STOR_002
0.000.000.00
V
VDC STOR_001
0.000.000.00
STO
R_0
02
0.00-0.000.00
STO
R_0
02
0.000.000.00
STO
R_0
01
0.00-0.000.00
STO
R_0
01
0.000.000.00
STO
R_0
03
0.00-0.000.00
STO
R_0
03
0.000.000.00
V
VDC STOR_003
0.000.000.00
Koppeling
45.4111.9367.82
Koppeling
-45.41-11.9367.82
Link
1Li
nk 1
Link
4Li
nk 4
Trf 3
92
45.4112.092.71
Trf 3
92
-45.41-11.9367.82
Line
73
(Gro
en)
6.191.739.29
Line
73
(Gro
en)
-6.19-1.739.29
Line
226
(Paa
rs)
12.973.11
19.27
Line
226
(Paa
rs)
-12.97-3.1119.27
Line
39
(Roz
e)
15.094.23
22.64
Line
39
(Roz
e)
-15.08-4.2222.64
Line
3 (B
lauw
)
11.152.8616.63
Line
3 (B
lauw
)
-11.15-2.8616.63
Link 402Link 402
Link
403
Link
403
Roelofs
45.41 kW12.09 kvar
0.97
DIg
SIL
EN
T
14
Holiday parc Bronsbergen in Zutphen
200 homes ~300 kW solar panels
15
Flywheel system
16
Results Zutphen
Flywheel system passed all tests All initial problems were solved (like communication
problems) Flywheel and inverter work well Harmonic compensation worked perfectly Reactive power supply was OK Islanding mode worked perfect
Noisy system Power losses of the system
17
Battery test site Zamudio and Chambery
18
Transportable storage systems
19
Results Chambery and Zamudio
System is transportable All initial problems were solved (like communication problems) Battery and inverter worked well together Energy management system is developed and tested in
Chambery Load alleviation is proven in these field tests
Communication problems caused a delay in these field tests Energy management system is a critical component of the system
20
Final field test Mannheim
3 storage systems at one location (2 Li-ion batteries
and 1 flywheel) A large PV solar power unit is installed at this location January 2011 – June 2011
21
Lessons learned GROWDERS
Transportation of storage systems is possible, but not always
easy to realize (heavy, strict regulations)
Technology is still in the demonstration phase (e.g. the energy
management system, communication between the components)
Financial benefits are for different stakeholders– this is not organized yet
Regulations are not clear and differ per country
Number of locations available for the grid operator is often limited
22
The need of storage in the grid
GROWDERS demonstrated that Transportable
(containerized) storage can be used on LV networks
when Distributed Energy Resources (DER) are
installed:– To alleviate potential network overloads
– For voltage control to avoid over/undervoltage
– To mitigate voltage dips
– To improve Power Quality (PQ) (Flicker, harmonics etc.)
Demand and DGNetwork Storage
23
Questions
Questions that need to be answered:
– What type and size of storage is required?
– Where on the network should storage be installed?
– When should the storage system be charged / discharged?
– What are the costs and benefits?
Demand and DGNetwork Storage ?? ?
24
Planning tool for optimizing storage (PLATOS)
Tool that assist network planners to optimize the
location, size and types of energy storage systems in
an electrical power system
LS RAIL 40.401.00
-150.20
GE HKL0.401.00
-150.20
Sectie Kast 20.401.00
-150.20
Sectie Kast 1 0.401.00
-150.20
MS rail 4 10.001.000.00
V
VDC STOR_002
0.000.000.00
V
VDC STOR_001
0.000.000.00
STOR
_002
0.00-0.000.00
STOR
_002
0.000.000.00
STOR
_001
0.00-0.000.00
STOR
_001
0.000.000.00
STOR
_003
0.00-0.000.00
STOR
_003
0.000.000.00
V
VDC STOR_003
0.000.000.00
Koppeling
45.4111.9367.82
Koppeling
-45.41-11.9367.82
Link
1Lin
k 1
Link
4Lin
k 4
Trf 3
92
45.4112.092.71
Trf 3
92
-45.41-11.9367.82
Line
73 (G
roen
)
6.191.739.29
Line
73 (G
roen
)
-6.19-1.739.29
Line
226
(Paa
rs)
12.973.11
19.27
Line
226
(Paa
rs)
-12.97-3.1119.27
Line
39 (R
oze)
15.094.23
22.64
Line
39 (R
oze)
-15.08-4.2222.64
Line
3 (B
lauw)
11.152.8616.63
Line
3 (B
lauw)
-11.15-2.8616.63
Link 402Link 402
Link
403
Link
403
Roelofs
45.41 kW12.09 kvar
0.97
DIgS
ILEN
T
•Weather forecast
•Trade price forecast
Generation & Load forecast
Network
Storage
specifications
Optimal
Storage
Solution
Genetic
Optimization
Calculate costs
and benefits
Load flow
analysis
3
1
4
Prediction module
Modeling module
Optimization module
Costs / benefits module
1
2
2
3
4
Complexity of Decision Making
Optimizing the use of storage
Decision methodology introductionfor distribution system planning
Focus: Application to storage
Implementation in GROWDERS (PLATOS)
26
Setting expectations
Learning goals:– Getting insight in overview of decisions
about storage applications
– Setting foundation for understanding implementation
for GROWDERS
General description of decision process– Introduction to decision process only
– Pitfall: not everything can be explained
Focus on implementation of principles and
choices made for GROWDERS (PLATOS)
27
Scenario 1
Scenario 2
Scenario n
Distribution system planning in general
Decision
process (Sub)Scenarios Optimal solutions
Optimal distribution network
Bottlenecks
PowerFactory 13.2.337
Projec t:
Graphic : Grid
Date: 4/17/2008
Annex:
Load Flow Balanced
Nodes
Line-Line Voltage, Magnitude [V]
Voltage, Magnitude [p.u.]
Voltage, Angle [deg]
Branches
Active Power [kW]
Reactive Power [kvar]
Current, Magnitude [A]
E
A
R
P
2-Winding..
199.9
496.9
36.4
122.2
2
-199.9
4-9
3.9
7312.8
922.2
2
STOR_002
0.000.000.00
STOR_003
0.000.000.00
STOR_001
0.000.000.00
5
14.0
06.7
823.3
1
18
-b
14.0
06.7
823.1
4
18
-o
18. 0
08.7
229. 7
5
4
19.009.2030.97
14.1
26.7
923.3
123.3
1
-14.0
0-6
.78
23.3
123.3
1
47.3
822.4
476.2
076.2
0
-46.1
1-2
2.2
976.2
076.2
0
19.219.2330.9730.97
-19.00-9.2030.9730.97
67.4
031.9
2107.1
7107.1
7
-66.5
9-3
1.6
6107.1
7107.1
7
Ex
tern
al
Gri
d
199.9
4..
96.9
3 .
.0.9
0
19.219.2330.6130.61
-19.00-9.2030.6130.61
3
19.0
09.2
030.6
2
87.9
541.5
8137.7
8137.7
8
-86.6
1-4
1.1
5137.7
8137.7
8
13
0.0
00.0
00.0
0
12
11.255.4518.75
0.0
00.0
00.0
00.0
0
-0.0
0-0
.00
0.0
00.0
0
-11.25-5.4518.7518.75
11.325.4618.7518.75
-18.00-8.7229.7429.74
18.198.7429.7429.74
14
16.2
07.8
527.3
1
#
17. 1
08.2
828. 8
4
10
18.008.7229.74
8
16.207.8525.97
16.367.8627.3027.30
-16.19-7.8427.3027.30
17.2
88.3
028.8
328.8
3
-17.0
9-8
.28
28.8
328.8
3
34.3
216.2
456.1
356.1
3
-33.6
3-1
6.1
656.1
356.1
3
50.1223.3879.7879.78
-48.72-23.2279.7879.78
24.2
711.6
039.6
339.6
3
-23.9
2-1
1.5
639.6
339.6
3
43.5
120.4
669.3
669.3
6
-42.4
6-2
0.3
469.3
669.3
6
111.9
952.3
8175.1
1175.1
1
-109.8
3-5
1.6
9175.1
1175.1
1
0
12.606.1020.88
6
14.406.9723.65
407.6
50.9
9-0
.69
20000.0
01.0
00.0
0
28
Changing balance of objectives
(Risk management)
Avoid
ControlIntuition
Well funded(Net present value, Life cycle costing)
Operation Immediate to short time Direct effects Assets are given Configuration may change Actual parameters can be set Operational values are known,
alterations in procedures are possible
Projects
Strategy
Planning
Operations
36
Problem definition
A well described problem definitioncontains:– Uncertainties– Decision variables– Objectives– Constraints– Space– Time
General problem definition:What should I do (choice of decision variables) to obtain the objective(s) and satisfy the constraints, given the uncertainty, time and space?
Set borders Time
Complexity
New development
Generate alternative solutions
Technical evaluation per solution (check constraints)
Optimize the use of storage in a distribution grid– Decide about number, location, type, size
– Maintaining all grid constraints
– Calculate (optimal) total costs / benefits
Post processing:– Compare result with conventional solutions
– Decide about (economic) duration of (optimal) storage solution
38
Problem definition in GROWDERS (2/2)
Solve grid problems– Under/over-voltages
– Overloaded components
– Mitigation of voltage dips
Postpone investment– Reduce uncertainty
Trading– Subject to grid constraints
Generic: long(er) term planning process with operational
(short term) constraints
39
Generating alternatives (general)
Alternative = possibility to solve the problem Brainstorm:
– Any potential idea is (initially) OK
– Selection will be done later
Set up a not too small number of alternatives– When there is less creativity in alternatives this can lead to no solution
or the lack of the optimal solution
Choose the best version per set of alternatives If needed, use phases
“DO NOTHING”is also an Alternativeto be considered!
40
Storage alternatives within GROWDERS
Use of Storage– Number
– Type(s)
– Size(s)
– Location(s)
Combination of multiple storage systems All simultaneously Number of potential solutions is combinatorial
Future:– Use of electric cars (V2G Vehicle to Grid)
– Mobile storage (short term mobility)
41
Added complexity through storage (1/2)
Storage is both demand and generation
Pattern not given or estimated but (second stage) result of controls after “normal” demand & generation
Has both operational & planning characteristics– And others, like economics, space
Possible conflicting interests (different stakeholders)– E.g. peak-shaving for grid operator, peak enhancing due to trading
company. Solved in GROWDERS by maximizing profit by trade, while keeping minimal grid constraints
42
Added complexity through storage (2/2)
Storage adds many new decisions When choosing 3 locations (begin/middle/end) in 4
feeders for small or larger storage (single type), this leads to theoretically 412 = 16.777.216 combinations
More realistic: 80 homes, available storage in 2 types, each 3 sizes: theoretically 680≈2*1062 options
When storage operations have to be optimized along the pattern, each trial solution requires a loadflow sweep for e.g. 8760 hours per year– loadflow means matrix inversion
Thus analytically impossible to solve
43
Other alternatives within GROWDERS“conventional” solutions Operational
– Change ratings, setting or use dynamic monitoring
– Demand side (& generation) management
– Load or generation curtailment (fined)
Planning (investments)– Add cables, lines
– Add transformer capacity
Reconfigure system (not within PLATOS) In GROWDERS (PLATOS) results are compared, not
simultaneously optimized
44
New development
Generate alternative solutions
Technical evaluation per alternative solution (check constraints)
Per alternative solution:Define optimal investment phasesEvaluate expected objectives (costs, reliability, image, …)
Usual loadflow and short circuit checks Time patterns through “sweep” function Uncertainty modeled through prediction error: Neural
Network used to determine prediction error of demand
Prediction error
Demand
Historical data from Bronsbergen (NL) used. Neural Network predicts on basis of:- Time of the day- Day of week- Which month- Public holidays- Meteorological data (temperature, cloudiness)
47
Different objectives
Objectives DSO Performance Indicators For example
– Economical (costs, profits) CAPEX + OPEX
– Reliability
– Social environment
– Risk (financial)
– Image
n
tt
t
iCF
NPV0 )1(
CF = Cashflow
i = interest
t = time (year)
n = period
n
tt
t
iCF
NPV0 )1(
CF = Cashflow
i = interest
t = time (year)
n = period
48
Objectives within GROWDERS
Net Present Value of all costs Trading: Maximized within grid limits Performance indicators translated to costs, using
prices– Overloading
– Over/under-voltage
– Voltage dipsStorage Solution Generator
Alleviation Algorithm:
Overloading
Alleviation Algorithm:
Over/undervoltage
Alleviation Algorithm:
Voltage Dips
Trading Algorithm
Solution Assessment
Output data processing
Performance Indicator 1
Performance Indicator 2
Performance Indicator 3
Performance Indicator 4
+ + +
Non-storage Solution Generator
Performance Indicator
49
New development
Generate alternative solutions
Technical evaluation per alternative solution (check constraints)
Per alternative solution:Define optimal investment phasesEvaluate expected objectives (costs, reliability, image, …)