modeling and real time simulation of wind power … 2- real-time simulation 3- rt-lab platform in...
TRANSCRIPT
KHIAT Mounir, BENAAMA Kamel, KELKOUL IlyesLaboratory SCAMRE, ENPO-MA , Oran, Algeria
Modeling and Real Time Simulation of Wind Power Systems Using Rt-Lab
Platform
1- Introduction
2- Real-Time Simulation
3- RT-LAB Platform In SCAMRE Laboratory
4- Modeling and Real Time Simulation of Wind Power Systems
5- Results and Discussion
6- Conclusion
Plan of presentation
1- Introduction
In response to the increasing demands of electrical systems for performance,reliability and cost, The control and protection equipments has becomeincreasingly sophisticated.
It is essential to validate these equipment before they are installed on the realpower system.
To accelerate the development and validation cycle of these equipments, toreduce costs and risks, the current trend is to test these equipments with areal-time digital simulator.
2- Real-Time Simulation
The objective of the real - time simulator is to test the differentelectrical equipments in the most natural possible conditions: asif they were connected to the real physical systems associatedwith them.
Therefore, the real-time simulator must reproduce as closely aspossible the dynamic behavior of the electrical system undercontrol.
The real-time simulation of the electrical system to becontrolled passes first through:
1- A modeling phase that consists in the putting of equation of
the system.
2-Then a phase of conception of an algorithmic specification(choice of sampling period, discretization andquantification)
3- And finally, a phase of real - time implantation.
Among the real-time digital simulators used in the world, are:
RT-LAB of OPAL-RT Technologies (CANADA).
ARENE URT (EDF R&D, France).
RTDS (RTDS Inc.,Manitoba/Canada).
NETOMAC (SIEMENS, Allemagne).
Etc.
3- RT-LAB Platform In SCAMRE Laboratory
The real-time simulator used in our SCAMRE laboratory is the RT-LAB digital simulatordeveloped by OPAL-RT Technologies (Montreal, Canada).
RT-LAB Platform in SCAMRE Laboratory
AC / DC / AC converter (Triphase), 17 KW, connected to the real-time simulator.
3.1 Hardware Architecture
The hardware architecture installed within our laboratory SCAMRE, it iscomposed of two connected simulators, the Wanda 4u and the OP 5600 . Thetarget is equipped with two CPU processers (Intel Xenon six-core, 3.33 GHZ,12 M, 6.4 GT/s), including 2 cores enabled and 16 I/O, for Wanda and twoCPU processers (Intel Xenon six-core, 3.46 GHZ, 12 M, 6.4 GT/s) including 2cores enabled and 16 I/O, for OP5600
The target is responsible for the execution of models. Development, edit,verification, and compilation of models are all done on the host computer;moreover, it works as a console or command station in charge of control andobservation during simulation. Ethernet is used to communicate between hostsand targets. Communication among nodes is used by shared-memoryarchitecture. The host computer is a general PC.
System Hardware Overview for op5600 and Wanda 4u System Connection Overview
3.2. Software Architecture
The figure represents the software architecture on the host. All the studied models are
developed under the Matlab/Simulink environment. RT-LAB is a real-time GUI platform
and it is designed to make the real-time simulation of the Simulink models on the
clusters. RT-LAB builds parallel tasks from the original Simulink models and run them
on each core of the multi-core CPU computer or on the separate computers.
RT-LAB
(Real Time Simulation GUI platform)
MATLAB / SIMPOWER
(Model Development Environment)
eMEGASim With Artemis
(Fixed-Step Solvers For Real Time
Simulation)
Software Architecture
In the RT-LAB simulation platform, a solver named Artemis which is designed
specifically for power system can improve the simulation speed greatly, and the multi-
processor operating mode makes it available to do real-time simulations on RT-LAB
platform by separating a complex system to some simple subsystems and do parallel
operations in multiprocessor.
RT-LAB can also connect physical devices to the simulation system to make the
simulation closer to the reality and to get more convincing results.
4- Modeling and Real Time Simulation of Wind Power Systems
4.1 Wind turbine system configurationThe dominant technology for utility-scale applications is the horizontal axis windturbine. Typical ratings range from 500 kW to 5 MW. It must be noted that the poweroutput is inherently fluctuating and non-dispatchable. A typical wind turbineconsists of the following subsystems:-Rotor (consists of blades and hub);- Drive-train (shafts, gearbox, couplings, mechanical brake, and electrical generator) ;- Nacelle and main-frame (housing, bedplate, and yaw system);- Tower and foundation;- Electrical system (cables, switchgear, transformers, and power electronicconverters if present).
Modern wind turbine diagram
This presentation, describes the detailed modeling and simulation of a wind
farm based on Doubly-Fed Induction Generators (DFIG), integrated in to
the power grid.
The main advantage of the DFIG is the possibility of operating at variable
speed, For it, the system of DFIG adjusts the speed of the rotor depending on
the wind speed.
Indeed, the DFIG allows to run in hyposynchronous and hypersynchronous
generator. Thus we arrive to extract maximum possible power.
The interest of the variable speed for a wind turbine is to be able to run over a
wide range of wind speeds and to get the maximum power possible for each
wind speed.
The DFIG configuration in which the voltage on the stator is applied from the
grid and the voltage on the rotor is induced by the rotor-side converter. The
frequency difference between the mechanical frequency (rotor side) and the
electrical frequency (stator side) is compensated through a converter that
injects a variable frequency rotor current both in normal and fault condition
operation.
4.2 DFIG Model
Today, 80% of new aerogenerators contain Doubly Fed Asynchronous (with rotorcoil).
One characteristic of the DFIG-based wind turbine system is the bidirectionalpower flow of the rotor. When ωr >ωs the power flows from the rotor to the powernetwork. When ωr <ωs , the rotor absorbs the energy from the power network
System Diagram of a DFIG-Based Wind Turbine System
4.3 Implementation
The model of the power system used in this study, is Composed of a 12 DFIG
connected to a system of transmission to high voltage 220KV with 19 buses,
power stations of generation (synchronous machines and regulators), and 11
loads. The nominal operating frequency of the network is 50 Hz.
B19
SM 3
L4L5
L1
L24
L22
L28L27 L26
B6
B5
B4
B2B3
Idealsource
B15
B16 B14
B11
B17
B7
B12 B13
B8
B9
B10
B1L25
L19
L31 L32
L30
L20 L21 L23
L13 L14
L15 L16
L17 L18
L2
L6 L3
L7 L8
L9 L10
L11
L12
SM 1
SM 2
SM 4
SM 5
L29
B18
Wind Farm (12 DFIG) UnitsGrid
Single-line diagram of the power system under investigation
4.3.1 Model description
The transmission grid is divided into two subsystems namely
"SS_NETWORK_A" and "SS_NETWORK_B". SS NETWORK_A consists of
the swing bus, buses (07) to (17), seven loads and three power stations of
generation. Buses (1) to (6) and buses (18) and (19) are located in
SS_NETWORK_B which contains also two power stations of generation. This
subsystem also consists of four loads and a tie in to the wind farm. The regulation
and control of generators are placed in SM _Control subsystem. Thewind farm
subsystem is designated by SS_Wind Farm and it is connected to the transmission
system.
The WANDA and OP5600 digital simulators are limited by the number of
processors enabled on the simulation target. In the case of this study, the
maximum number of active processors (CPU) licensed was four, hence the power
system could be broken into a maximum four subsystems, as shown in Figure .
The first CPU takes in charge the synchronous generators and their regulator
simulation. The 19 buses are distributed on CPU 2 and CPU 3 as shown in figures
15 and 16, respectively; the last CPU is reserved to the simulation of twelve
DFIGs with their controls.
Model Implementation of Park Aerogenerators Integrated Network.
Length: 75km
Length: 75km
Length: 75km
Length: 75km
Discrete,Ts = 5e-005 s.
powergui
control-WF
result-WF
A1
B1
C1
A2
B2
C2
A3
B3
C3
A4
B4
C4
SS_Wind_Farm
Cotrol-Bresult-B
A5
B5
C5
A6
B6
C6
A7
B7
C7
A8
B8
C8
B6a1
B6b1
B6c1
B6a2
B6b2
B6c2
SS_Network_B
control-A
result-A
B7a1
B7b1
B7c1
B7a2
B7b2
B7c2
SS_Network_A
fault
result-A
result-B
result-WF
result
control-A
control-B
SM_Control
resultcontrol-WF
fault
SC_Scoop
DPL
Length:75km
DPL
Length: 75km
DPL
DPL
DPL
DPL
ARTEMiS GuideTs=50 usSSN: OFF
5 Results and discussion
The system is running at the real time step of 50 μs. The parameters of the DFIG based wind
turbine are shown in next table. As can be seen from next figure 18, the adopt wind profile varies
between 8m/s and 12m/s for turbines in Wind-Farm subsystem.
0 10 20 300
5
10
Time [s]
speed [
m/s
]
Wind speed profile.
Parameter Value
Nominal power 1 MVA
Gear ratio 80.27
Line to line voltage 400 V
Frequency 50 HZ
DC-bus voltage 1250 V
Pole pairs 1
Stator resistant 0.0256294 pu
Rotor resistant 0.0100649 pu
Stator inductor 0.0998644 pu
Rotor inductor 9.62367 pu
Mutual inductance 3.47857 pu
Generator inertia 0.62 kg m2
Filter inductor 0.001 H
DC-bus capacitor 0.025 F
Kp and Ki for Pitch control PI 150 , 50
Kp and Ki for Pitch compensation PI 3 , 4
Kp and Ki for torque control PI 3 , 0.6
Kp and Ki for RSC (current regulation) 1.896 , 60.9162
Kp and Ki for RSC (voltage regulation) 0.87946 ,
60.905
Kp and Ki for GSC (current regulation) 4.158 , 500
Kp and Ki for GSC (DC regulation) 1.25 , 200
Wind turbine (DFIG) parameters
5.1 Wind turbine performanceSince the wind turbines are similar and present the same behavior, it is enough topresent the performance of the first wind turbine.
Figure 1 and Figure 2 show the behavior of the three-phase currents and voltagesgenerated by each DFIG. It is clear that the voltages and currents are perfectly stable andsinusoidal during steady-state operation.
Figure 3 shows the active and reactive powers generated by the wind turbine. Thesepowers are controlled by the rotor side converter (RSC).
Figure 4 shows the DC bus voltage, thanks to the grid side converter control (GSC) theDC bus is almost a constant value, in which the DC voltage can be maintained at 1250V.A stable voltage at the DC bus could help the GSC to control the active and reactivepowers.
The turbine and electromagnetic torques of the DFIG, as shown in Figures 5 and 6,respectively are following the wind speed profile in the aim to extract the maximumpower. The sign of torques is as function of the slip sign. When the slip is negative(over-synchronism) the electromagnetic and mechanical torque are negative obviouslyin steady-state.
0 5 10 15 20 25 30-400
-200
0
200
400
600
800
Time [s]
U [
V]
U1
U2
U3 10 10.01 10.02
-500
0
500
0 5 10 15 20 25 30
-0.2
0
0.2
0.4
0.6
Time [s]
I [A
]
Ia
Ib
Ic
×104
0 5 10 15 20 25 30
-0.4
-0.2
0
0.2
0.4
0.6
Time [s]
P[M
W],
Q[M
VA
r]
P Q
1- Wind generator voltage power 2-Wind generator currents 3- Wind generator active and
reactive
10 10.01 10.02-0.4-0.2
00.2
×104
0 0.51245
1250
0 5 10 15 20 25 300
500
1000
1500
Time [s]
Udc [
V]
0 0.51245
1250
0 5 10 15 20 25 30
-0.25
-0.2
-0.15
-0.1
-0.05
0
Time [s]
Tm
[kN
.m]
0 5 10 15 20 25 30-0.4
-0.3
-0.2
-0.1
0
0.1
Time [s]
Te [
kN
.m]
4- DC-bus voltage 5-Turbine torque 6- Electromagnetic torque.
5.2 Wind farm and grid performances
In this part, the simulation results obtained at bus 6 are presented. The bus 6 is chosen asmeasurement bus because it is connected to the transformer that joints the wind farm to thepower grid. At that bus, the three-phase voltages and currents are shown in figures 6 and 7,respectively.
It is can be seen that the waveforms are perfectly stable, and take sinusoidal shapes atsteady-state.
The voltage takes the value of 200 kV after being increased by the transformer connected atbus 18 where the voltage was 25 kV.
Figure 8 presents the active and reactive powers measured at bus 6. It is interesting tonotice that the active and reactive powers are totally decoupled.
The active power takes a value about -102 MW produced by the machine SM1 and windfarm, taking into account the loads consumption.
0 5 10 15 20 25 30-2
0
2
4x 10
5
Time [s]
U [
V]
U
a
Ub
Uc 10 10.01 10.02
-2
0
2x 10
5
0 5 10 15 20 25 30
-400
-200
0
200
400
600
800
Time [s]
I [A
]
Ia
Ib
Ic
10 10.01 10.02-500
0
500
0 5 10 15 20 25 30-15
-10
-5
0x 10
7
Time [s]
P [
W],
Q [
VA
r]
P Q
0 10 20 30
-1.025
-1.02
-1.015
-1.01x 10
8
6- Bus 6 voltages. 7-Bus 6 currents. 8-Bus 6 power.
5.3 System performance under faulty conditions
In order to exhibit the behavior of the proposed power system under faulty
conditions the bus 6 is grounded at t=5s, and the applied three-phase fault lasted for
1s. The real simulations results under the bus 6 grounding are recorded at buses 1,
6, and line 5-6.
The following figures show the results obtained for buses 6,1 and line 5-6
under faulty conditions.
0 5 10 15-20
-15
-10
-5
0
x 107
Time [s]
P [
W]
0 5 10 15
-2
0
2
4
6x 10
5
Time [s]
U [
V]
U
a
Ub
Uc
11 11.01 11.02-2
02
x 105
0 5 10 15
-1
-0.5
0
0.5
1
1.5x 10
4
Time [s]
I [A
]
Ia
Ib
Ic
11 11.01 11.02-500
0
500
Voltages at Bus 6. Currents at Bus 6. Active power at Bus 6.
0 5 10 15-0.5
0
0.5
1
1.5
2
2.5x 10
6
Time [s]
P [
W]
106
0.5
-0.1
0.1
0.2
0.3
0.4
5 10 15 20 25 300
200
400
600
800
1000
1200
Time [s]
U d
c
0 5 10 15
-3
-2
-1
0
1
2
x 108
Time [t]
P [
W]
0 5 10 15
-2
0
2
4
6x 10
5
Time [s]
U [
V]
Ua
Ua
Ua 11 11.01 11.02
-2
0
2x 10
5
0 5 10 15
-1500
-1000
-500
0
500
1000
1500
2000
Time [s]
I [A
]
Ia
Ib
Ic 11 11.01 11.02
-200
0
200
0 5 10 15-4
-3
-2
-1
0
1
2x 10
8
Time [s]
P [
W]
5 5.5 6
-4
-2
0x 10
7
Voltages at Bus1 Currents at Bus1 Active power at Bus1.
Wind generator active power. DC-bus voltage Active power at line5-6.
5 6 71100
1200
1300
6. Conclusion
In this presentation, a detailed wind farm model based on DFIG connected to alarge power system is designed and simulated using eMEGASim, real-time digitalsimulator, In order to allow for a higher level of wind turbines integration inelectric grid without affecting the quality of the generated electric power.
Simulation results show clearly the good performance of the overall system basedon eMEGASim, real-time. Indeed, with much less time consuming, this simulatorhas demonstrated to be an effective way to study interactions between complexnonlinear power system components as in real operating conditions.
THANK YOU