Transcript CERN-Presentation_2013.10.24_Junjie_Tangx

Advanced Simulation tools in support
of the design of the Power System
architecture of the European Spallation
Source (ESS)
24.10.2013
Prof.Monti, Junjie Tang
E.ON Energy Research Center, RWTH Aachen University, Aachen, Germany
The E.ON Energy Research Center
 June 2006: the largest research co-operation in Europe between a private
company and a university was signed
 Five new professorships in the field of energy technology were defined across 4
faculties
 Research areas: energy savings, efficiency and sustainable power sources
24.10.2013 | ACS Automation of Complex Power Systems
Junjie Tang
|
E.ON ERC Infrastructure
24.10.2013 | ACS Automation of Complex Power Systems
Junjie Tang
|
Outlines
 Background
 Simulation Project Goal
 Simulation Approach and Modeling
 Test Scenarios and Results
 Conclusions and Future works
24.10.2013 | ACS Automation of Complex Power Systems
Junjie Tang
|
Background
 ESS: European Spallation Source
≡ A joint European project, which has partners from 17 European countries
≡ Trend for reduction of energy consumption and greenhouse gas (GHG)
emission
≡ Design proposal: establish an energy concept for demanding energy targets
to be
= RESPONSIBLE – 20 per cent reduction in energy consumption
= RENEWABLE – 100 per cent utilisation of renewable energy
= RECYCLABLE – 60 per cent recycling of utilized energy
 Grid simulation project
≡ Sustains for 2 years
≡ Partners include:
24.10.2013 | ACS Automation of Complex Power Systems
Junjie Tang
|
Background
Fig.2: Process of generating neutron
Fig.1: Structure of ESS and operation
[source:http://europeanspallationsource.se/photos-images]
 Internal ESS distribution grid, connected with several sections
 Linear accelerator (LINAC), the key component in ESS facility
24.10.2013 | ACS Automation of Complex Power Systems
Junjie Tang
|
Simulation Project Goal
 Develop a simulation model for the regional electricity grid, including
ESS grid
 Evaluate the mutual impact between newly integrated ESS and the
regional grid:
≡ predict disturbances that ESS can cause to the regional grid
≡ predict disturbances in the operation of the ESS that can be caused by the
regional electricity grid
Lund Grid
145 kV
D
G
D
G
D
G
T1-ESS
Wind farms
Other loads
in Lund
23 kV
G
T2-ESS
ESS energy solution
Evolution
24.10.2013 | ACS Automation of Complex Power Systems
|
G
Regional
power plant
ESS grid
Conventional energy solution
Junjie Tang
D
G
The RT Simulation Lab
24.10.2013 | ACS Automation of Complex Power Systems
Junjie Tang
|
Co-simulation Approach—Introduction
 Real time simulation and Hardware In the Loop methods are
essential tools for the development of future complex power
systems
 Given the complexity of such a scenario the use of a single tool
is unfeasible
 Co-simulation approaches have to be developed, so that:
≡
≡
≡
≡
Dedicated tools and library can be shared
Different expertees can be capitalized
Facilities at different geographical locations can be interconnected
Multi-rate execution can be perfomed
24.10.2013 | ACS Automation of Complex Power Systems
Junjie Tang
|
Co-simulation Approach
 RTDS has a good ability to simulate electromagnetic transient, while
Opal RT is compatible with the models built in MATLAB/Simulink
 The idea is to remain Lund grid models in RTDS, while to model ESS
grid in Opal RT including the power electronics of LINAC
 Connection between two simulators for co-simulation by analog interface
24.10.2013 | ACS Automation of Complex Power Systems
Junjie Tang
|
Modeling Lund Grid in RTDS
Rack 1
Current source
controlled by
Opal RT signal
Rack 2
Fig.: Lund grid modeling in RTDS
24.10.2013 | ACS Automation of Complex Power Systems
Junjie Tang
|
Modeling ESS internal grid in MATLAB/Simulink
Modeling
Power electrical part
Transplant
Download
I/O ports
Exchange
with RTDS
Power eletronics
part
24.10.2013 | ACS Automation of Complex Power Systems
Junjie Tang
|
Modeling ESS internal grid in RT-LAB
Subsystem 2 of
ESS grid
Subsystem 3 of
ESS grid
Subsystem 1 -equivalent voltage source
and LINAC loads
Voltage source
controlled by RTDS
signal
Subsystem 4 of
ESS grid
24.10.2013 | ACS Automation of Complex Power Systems
Junjie Tang
|

ESS grid is split to 4
subsystems

Each subsystem using
one CPU/core for realtime simulation with
parallel technology
Modeling power electronics for LINAC in RT-LAB
Power electronics part
amplify to
200 times
P: variation in range of [75, 310] kW
P: variation in range of [15, 62] MW
 Depends on control of power electronics, the power consumption is not
very close to constant
 Simulate 200 LINACs with power electronics in real-time is infeasible
 Equivalent modeling with a controllable load and a amplified signal, is
used to simulate power amplification
24.10.2013 | ACS Automation of Complex Power Systems
Junjie Tang
|
Testing Scenarios—Short circuit
Test scenarios
Location
Type
Conditions, purpose
ESS_in
three phase L-G
with different fault clearing times:
0.05 s, 0.1 s and 0.3 s
ESS_in
three phase L-G
with different impedances of T1-ESS and
T2-ESS: 11.5%, 15% and 20%
ESS_in
three phase L-G
with different short circuit impedance:
0.01 Ω, 0.1 Ω and 1Ω
ESS_in
Phase A and B L-L
compare with three phase L-G
ESS_in
Phase A L-G
compare with three phase L-G
ESS_out
three phase L-G
with different impedances of T1-ESS and
T2-ESS: 11.5%, 15% and 20%
ESS_out
three phase L-G
with different fault clearing times:
0.05 s, 0.1 s and 0.3 s
ESS_out
Phase A and B L-L
compare with three phase L-G
ESS_out
Phase A L-G
compare with three phase L-G
SEE
three phase L-G
compare faults at different buses
SEE
Phase A and B L-L
compare with three phase L-G at SEE
SEE
Phase A L-G
compare with three phase L-G at SEE
VKA
three phase L-G
compare faults at different buses
Short circuit
 To check the impacts when various types of short circuits happen at different
locations, as well as impedances of transformers and grounding
24.10.2013 | ACS Automation of Complex Power Systems
Junjie Tang
|
Testing Scenarios—Loss of components
Test scenarios
Tripping of
power source
Loss of branch
and transformer
Location
Type
HVA
HVB
HVDC
Wind farm
Örtofta and Helene
simultaneously
ÖVT Steam and Gas
simultaneously
(SEE-ESS) and (ESS-Lund)
simultaneously
(SEE-ESS) and (HVA-BBK)
simultaneously
(SEE-ESS) and (SEE-VKA)
simultaneously
T2-ESS
tripping
tripping
tripping
tripping
tripping
Purpose
compare such kind of faults
at different buses
tripping
loss
loss
compare such kind of faults
at different buses
loss
loss

As common fault types - loss of components
 With cascading faults occurring more frequently than in the past, different
combinations of component loss are relevant
 Loss of transformers connecting ESS and Lund grid - possible disruptive impact
on the operation of ESS
24.10.2013 | ACS Automation of Complex Power Systems
Junjie Tang
|
Testing Scenarios—Normal operations
Test scenarios
Location
Connection
ESS_in
Disconnection
ESS_in
Charging
schemes
ESS_in
compare with the impact caused by charging scheme I and
III on grid
ESS_in
analysis for realistic data based harmonics
Harmonics
check the impact of such operations on the grid
ESS_in
SEE
Maxlab
Wind farm
Purpose
Lillgrund
analysis for the threshold of THD 8% (as defined in the grid
code of E.ON Sverige AB) based harmonics
check the impact of wind farm with different generation
ratio: 0%, 50%, 100%, in aspect of power flow

Normal operation of ESS may also impact the whole grid
 Charging schemes of LINAC mainly determine the load characteristics of ESS
 Variability of wind power generation impacts power flow and state variation in
Lund grid
24.10.2013 | ACS Automation of Complex Power Systems
Junjie Tang
|
Test Result 1
3-phase L-G short circuit at ESS_in with different transformers impedances
0
0.1
0.2
0.3
0.4
0.5
t [s]
0.6
0.7
0.8
Phase A
Phase B
Phase C
0.9
1
50
0
-50
0
0.1
0.2
0.3
0.4
0.5
t [s]
0.6
0.7
50
0
0.2
0.3
0.4
0.5
t [s]
0.6
0.7
ESS__out
ESS__in
Maxlab
SEE
50.5
50
49.5
0
0.1
0.2
0.3
0.4
0.5
t [s]
0.6
0.7
50
0
0.1
0.2
0.3
0.4
0.5
t [s]
0.6
0.7
0.8
0.9
50
0
0.1
0.2
0.3
0.4
0.5
t [s]
0.6
0.7
0.8
0.9
0
1
1

0.1
0.2
0.3
0.4
0.5
t [s]
0.6
0.7
0
0.8
0.9
1
ESS__out
ESS__in
Maxlab
SEE
0
0.1
0.2
0.3
0.4
0.5
t [s]
0.6
0.7
0.8
0.9
1
ESS__out
ESS__in
Maxlab
SEE
100
1
ESS__out
ESS__in
Maxlab
SEE
51
49
0.9
ESS__out
ESS__in
Maxlab
SEE
50.5
49.5
0.8
0
100
Angle [radians]
0.1
200
0
Angle [radians]
0
200
ESS__out
ESS__in
Maxlab
SEE
100
Angle [radians]
-50
|Voltage| [kV]
0.8
Phase A
Phase B
Phase C
0.9
1
0
-50
200
|Voltage| [kV]
0.8
Phase A
Phase B
Phase C
0.9
1
|Voltage| [kV]
Voltage of buses
50
Frequency of buses
Frequency [Hz]
Frequency [Hz]
Frequency [Hz]
Iabc [kA]
Iabc [kA]
Iabc [kA]
Short circuit current with transformer impedance 11.5% , 15% , 20%
0
0.1
0.2
0.3
0.4
0.5
t [s]
0.6
0.7
0.8
0.9
1
0.7
ESS__out
ESS__in
Maxlab
SEE
0.8
0.9
1
0.7
ESS__out
ESS__in
Maxlab
SEE
0.8
0.9
1
Angle of buses
2
0
-2
0
0.1
0.2
0.3
0.4
0.5
t [s]
0.6
2
0
-2
0
0.1
0.2
0.3
0.4
0.5
t [s]
0.6
ESS__out
ESS__in
Maxlab
SEE
2
0
-2
0
0.1
0.2
0.3
0.4
0.5
t [s]
0.6
0.7
0.8
0.9
1
With different impedances of transformers 11.5%, 15% and 20%
 Maximum of AC short circuit current are 30.16, 26.54 and 21.59 kA individually
 Higher impedance of the transformers can restraint the short circuit if such kind
fault happens at ESS internal
24.10.2013 | ACS Automation of Complex Power Systems
Junjie Tang
|
Test Result 2
Loss of transformer T2-ESS
Voltage when loss of T2-ESS
Frequency when loss of T2-ESS
142
50.05
140
50
138
ESS__out
ESS__in
Maxlab
SEE
134
49.95
Frequency [Hz]
|Voltage| [kV]
136
132
130
49.9
49.85
128
ESS__out
ESS__in
Maxlab
SEE
126
49.8
124
122
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
t [s]
Three phase voltage of ESS-in
49.75
1
0
0.5
0.4
0.3
0.2
0.1
t [s]
Three phase voltage of ESS-out
0.6
0.7
0.8
0.9
1
Phase A
Phase B
Phase C
20
0
-20
100
Voltage [kV]
Voltage [kV]
40
Phase A
Phase B
Phase C
50
0
-50
-100
-40
0.05
0.1
0.15
0.2
t [s]
Three phase voltage of Maxlab
0.25
0.05
0.1
0.15
0.2
t [s]
Three phase voltage of SEE
0.25
Phase A
Phase B
Phase C
10
0
-10
-20
0.05
0.1
0.15
t [s]
0.2
100
Angle [radians]
Voltage [kV]
20
0.25
Phase A
Phase B
Phase C
50
0
-50
-100
0.05

0.1
0.15
t [s]
0.2
0.25
Maximum power consumption of ESS 47 MW, while capacity of transformer 40 MVA
 Once any one of the transformers is out of work, there is a potential risk of overload for
the other one
 For the voltage of ESS_in, the decrease is over 10%
24.10.2013 | ACS Automation of Complex Power Systems
Junjie Tang
|
Test Result 3
Short circuit at ESS_in with different fault types
Three phase voltage of ESS-in
Three phase voltage of ESS-out
Three phase voltage of ESS-in
-20
Phase A
Phase B
Phase C
50
0
-50
-100
0.1
0.15
0.2
t [s]
Three phase voltage of Maxlab
0.25
0.05
Phase A
Phase B
Phase C
20
0
-20
100
Voltage [kV]
0
100
Voltage [kV]
20
-40
0.05
Phase A
Phase B
Phase C
50
0
-50
-100
0.1
0.15
0.2
t [s]
Three phase voltage of SEE
-40
0.05
0.25
20
0.1
0.15
0.2
t [s]
Three phase voltage of Maxlab
0.25
0.05
0.1
0.15
0.2
t [s]
Three phase voltage of SEE
0.25
-10
Phase A
Phase B
Phase C
50
0
-50
Phase A
Phase B
Phase C
10
0
-10
-100
-20
0.05
0.1
0.15
t [s]
0.2
0.25
0.05
0.1
Single phase A L-G
Three phase voltage of ESS-in
0.15
t [s]
0.2
-20
0.05
0.25
0.1
0.15
t [s]
0.2
0.25
100
Angle [radians]
0
100
Voltage [kV]
20
Phase A
Phase B
Phase C
10
Angle [radians]
Voltage [kV]
Three phase voltage of ESS-out
40
Phase A
Phase B
Phase C
Voltage [kV]
Voltage [kV]
40
Phase A
Phase B
Phase C
50
0
-50
-100
0.05
0.1
Phase AB L-L
0.15
t [s]
0.2
0.25
Three phase voltage of ESS-out
Phase A
Phase B
Phase C
20
0
-20
100
Voltage [kV]
Voltage [kV]
40
Phase A
Phase B
Phase C
50
0
-50
-100
-40
0.05
0.1
0.15
0.2
t [s]
Three phase voltage of Maxlab
0.25
0.05
0.1
0.15
0.2
t [s]
Three phase voltage of SEE
Fault type
Va (kV)
Vb (kV)
Vc (kV)
Single phase A L-G
0.3778
28.04
31.48
Phase AB L-L
8.94
7.48
16.37
Three phase L-G
2.14
2.06
2.39
0.25
Phase A
Phase B
Phase C
10
0
-10
100
Angle [radians]
Voltage [kV]
20
Phase A
Phase B
Phase C
50
0
-50
-100
-20
0.05
0.1
0.15
t [s]
0.2
0.25
0.05
0.1
Three phase L-G
0.15
t [s]
0.2
0.25

Recordings about the instantaneous voltage at ESS_in, ESS_out, Maxlab and SEE
 Single phase L-G short circuit introduces overvoltage to the other two phases of ESS_in
 Two-phase L-L short circuit lead voltage decrease in the two phases of ESS_in
 Three-phase L-G cause serious undervoltage for each phase of ESS_in
24.10.2013 | ACS Automation of Complex Power Systems
Junjie Tang
|
Test Result 4
3-phase L-G short circuit at different locations in the Lund grid
Bus_Name
AIE
BBK
BFO
ELVS
ESS
HVA
HVB
HVDC
Helene
Lund ÖM
MRP
Maxlab
ÖVT
Örtofta
SEE
SÖV
VKA
VPE
HRGD
Hörby
Lillgrund
SSY
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Fig.: Maximum instantaneous amplitudes of three-phase L-G short circuit current of each bus
when ESS is connected to Lund grid
 Three-phase L-G short circuit is the most detrimental to the
system
 Effects of short circuit faults highly depend on their locations in
the grid
 The fact that the ESS is connected or disconnected from the
Lund grid has a minor impact on the short circuit current
24.10.2013 | ACS Automation of Complex Power Systems
Junjie Tang
|
Test Result 5
Impact from different charging schemes of LINAC of ESS grid
Active Power Consumption [MW]
55
Charging scheme I
Charging scheme III
50
45
40
35
30
25
20
15
0
0.2
0.4
0.6
0.8
1
t [s]
24.10.2013 | ACS Automation of Complex Power Systems
Junjie Tang
|
Test Result 5
Impact from different charging schemes of LINAC of ESS grid
Scheme I:
|Voltage| [kV]
600
ESS__out
ESS__in
Bus 12
Bus 15
400
200
0
0
0.1
0.2
0.3
0.4
0.5
t [s]
0.6
0.7
0.8
0.9
1
Scheme II:
|Voltage| [kV]
145
140
ESS__out
ESS__in
Bus 12
Bus 15
135
130
125
0
0.1
0.2
0.3
0.4
0.5
t [s]
0.6
0.7
0.8
0.9
1
 Voltage: the Charging Scheme I yields high flicker to the other loads,
especially for Bus 12 (another research facility around)
 Frequency of the voltage pulses is identical to charging frequency of
Scheme I (constant current)
 With respect to the voltage, Charging Scheme I is worse than Charging
Scheme II (constant power)
24.10.2013 | ACS Automation of Complex Power Systems
Junjie Tang
|
Test Result 6
Wind farm with different generation ratios
from 0% to 50%
from 50% to 100%

Due to the power being mostly provided by transmission network, the impact
aroused by the wind farms is limited, from the point view of power flow distribution
24.10.2013 | ACS Automation of Complex Power Systems
Junjie Tang
|
Test Result 7
Uncertainty analysis for the variation of loads and wind farm generation
Load condition
Maximum Minimum
Substation
(MW)
(MW)
AIE
400
45
BBK
75
25
BFO
15
4
ELVS
55
10
Heleneholm
25
10
Lund ÖM
35
15
Maxlab
20
15
MRP
110
25
SEE
235
100
SÖV
15
2
VKA
15
4
VPE
110
50
ÖVT
10
2.5
ESS
(Charging
45.7
31.7
Scheme II)
Test scenario
Uncertainty source
1
Load variation of Bus AIE, MRP, SEE and VPE
2
Intermittence of the two wind farms
3
Load variation of ESS
4
Load variation of the four buses in test scenario 1
together with ESS, and intermittence of the two wind
farms

Loads vary with uniform distributions, while the
Weibull distribution for wind speed of wind farms
 Uncertainty sources include five selected load
buses with large loads and two wind farms
 RTDS simulation co-operated with Monte Carlo
(MC) method is adopted to investigate the
uncertainty issue, (10000 MC simulations)
24.10.2013 | ACS Automation of Complex Power Systems
Junjie Tang
|
Test Result 7
Uncertainty analysis for the variation of loads and wind farm generation
Scenario 1
Scenario 2
Scenario 3
Scenario 4
 Ranges of voltage variation in scenarios are different
 Some voltages even drop to an unacceptable range in test scenario 2
and 4 due to the integration of wind farms
 Voltage is more volatile at the buses located with wind farm or heavy
loads
24.10.2013 | ACS Automation of Complex Power Systems
Junjie Tang
|
Test Result 7
Uncertainty analysis for the variation of loads and wind farm generation
0.06
0.4
0.04
0.02
0
137
0.2
Bus 15
Density
Density
scenario 1
scenario 2
scenario 4
0.3
0.15
0.1
0.05
137.5
138
Voltage [kV]
0
140.5
138.5
0.4
141
141.5
Voltage [kV]
0
136
0.3
0.2
0.1
137
138
139
140
Voltage [kV]
141
142
Fig.1: Frequency histogram of voltage at Bus 5 in test
scenario 1, 2, 4 respectively.
0
141
Bus 21
Density
0.1
142
0.2
Bus 14
Density
Density
0.2
Bus 12
0.15
0.1
0.05
141.5
142
Voltage [kV]
142.5
0
140.5
141
141.5
Voltage [kV]
Fig.2: Frequency histogram of voltage at Bus 12, 14, 15
and 21 in test scenario 3

Uncertainty merely from wind farm generations does not yield a large variation
to the voltage Bus 5 (ESS), while variations from the four largest load buses
lead to a broader range of possible voltage values
 Uncertainty of wind farms can mitigate the impact from stochastic load
variations
 In terms of voltage variation, Bus 12 (Maxlab) is larger in comparison with Bus
14 (power plant), Bus 15 (big load consumer) and Bus 21 (wind farm)
24.10.2013 | ACS Automation of Complex Power Systems
Junjie Tang
142
|
Test Result 8
Uncertainty analysis for the variation of loads in future scenarios
2013
Inauguration
Now
300
250
250
250
150
100
50
Frequency histogram [p.u.]
300
200
200
150
100
50
0
136
137
138
139
Voltage of ESS [kV]
140
141
0
136
Time
Fully in operation
300
Frequency histogram [p.u.]
Frequency histogram [p.u.]
2025
2019
200
150
100
50
137
138
139
Voltage of ESS [kV]
140
141
0
136
137
138
139
Voltage of ESS [kV]
140
 Load demand is uniform distributed, and its growth rate is assumed 5%
per year
 Errors in load prediction following a normal distribution (0, 0.01), the
incrementals are 0, 30%*(1±0.03) and 60%*(1±0.03) individually
 Uncertainty increases along with time, the impact will be bigger and
more chanllenging in future scenarios
24.10.2013 | ACS Automation of Complex Power Systems
Junjie Tang
|
141
Conclusions
 Main transformers connecting ESS grid and Lund grid highly decide the
status of ESS grid
 Load characteristics of ESS grid mainly depends on the charging
schemes of LINAC, which is possible to disturb the power quality of
Lund grid
 Integration of ESS brings a minor impact on short circuit current of the
buses in Lund grid
 Involvement of wind farms has a slight influence on the operation of
ESS grid and Lund grid
 A deeper awareness about how uncertainty from the renewable
sources and loads affects the operation of ESS grid and Lund grid is
obtained
24.10.2013 | ACS Automation of Complex Power Systems
Junjie Tang
|
Future works
 Heat energy can be also considered in ESS grid, as well as Lund
district heating system, will be modeled in RT-LAB to evaluate the
energy efficiency from a global view in co-simulation
 A platform is under construction for combining real-time simulation and
uncertainty quantification, to provide a possibility to extend such
analysis even to Hardware in the Loop and Power Hardware in the
Loop tests
24.10.2013 | ACS Automation of Complex Power Systems
Junjie Tang
|
Thanks for your attention!
Any question?
24.10.2013 | ACS Automation of Complex Power Systems
Junjie Tang
|