On Coordinated Control of OLTC and Reactive Power Compensation for Voltage Regulation in Distribution Systems With Wind Power
2016-12-06 16:19:29 0 举报
AI智能生成
本文研究了风力发电系统中的有载调压变压器(OLTC)和无功功率补偿器的协调控制策略,以实现电压调节。随着风力发电在配电网中的广泛应用,其对电压稳定性的影响日益显著。因此,提出了一种基于OLTC和无功功率补偿器的协调控制策略,以提高系统的电压稳定性。该策略通过优化OLTC的操作参数和无功功率补偿器的补偿容量,实现了对配电网电压的有效调节。实验结果验证了所提策略的有效性和优越性,为风力发电系统在配电网中的稳定运行提供了有力保障。
作者其他创作
大纲/内容
Introduction
Background: Voltage rise problem caused by the integration of DG
Traditonal solutions, such as grid reinforcement
Not cost effective
Active management schemes, such as RPC & OLTC
Aim: maintain the voltage within a given deadband
Only use OLTC
1. Voltage set point changes according to measurements obintained from critical locations
Identification of the critical points is hard
2. Assumation of the availability of voltage measurements from every node
The availability of voltage measurements is rare in exiting systems
Solutions
SE based on real-time measurement along with pseudo-measurement isused to determine the voltage level and to control the target voltage of AVC relays
load estimation based on customer class curves with measurement data at the substation and from remote DG are used to determine the maximum and minimum voltage in the network.
3. separate local measurements on feeders with load only and on feeders
that contains generation are made
that contains generation are made
Power output from DG is estimated and used to determine the voltage setting required to mitigate voltage rise at DG terminal based on measurement data and load sharing knowledge
Voltage setting required to mitigate undervoltage in the system is determined based on the load measurement in the feeder
Only use RPC
1. PI controller with the use of deadband
Make sure the PI controller works only when the voltage is above the max or below the min voltage level
2. PI controller with power factor tracking method & Fuzzy logic
Make sure the PI controller works only when the voltage is above the max or below the min voltage level
3. Aim: eliminate the voltage rise introduced due to active power injection from the wind turbine
keep the voltage at various points at the same level
increase the power loss in the system
4. Droop based control method
Fuzzy logic based location adaptive droop method to coordinate RPC from multiple DG for voltage mitigation
RPC using droop control function to mitigate voltage rise in multiple PV installations
Local voltage measurements are used to ensure the proper operation of the droop based compensator
Coordination of OLTC & RPC
Contribution of this paper
A new control strategy for the voltage rise mitigation by using RPC from wind turbine and OLTC of the substation transformer
A stragety that changes the voltage deadband instead of the voltage set point of the AVC relay of the tap changing transformer
Coordination of reactive power from multiple wind turbines for voltage regulation
Coordination of the voltage regulation from OLTC & RPC using a higher voltage deadband in the case of OLTC than RPC
The control of OLTC
Background
Problem
1. Voltage depends on the level of load & wind power in the system which cannot be known beforehand
2. Voltage information data are not available in the frequency that is required for voltage control
Solution
The applicability of using few real-time measurements together with pseudo-measurement to determine the voltage level in the network using SE & provide acceptable level of voltage regulation quality
A. The state estimation algorithm
Basic principle of SE: minimize the weighted error between measured values and calculated values
STEP 1: develop measurement functions that relate voltage angle & magnitude with measurement data available at each point
STEP 2: develop the Jacobin of the measurement function using (3) and the measurement covariance as given in (4)
STEP 3: set all bus voltage magnitudes equal to 1 p.u. except buses that measurements are available and bus voltage angles to 0. Using (2) to calculate next estimates of state vector
STEP 4: cycle repeats until the objective function J(x) is below a given threshold or the state vector change is below a certain small positive value
STEP 5: once the final estimate of the state vector is determined, use (6) to calculate the covariance Cx of the state vector
B. The control algorithm
1. The calculation of output signal of the voltage level analyzer block
2. Change the voltage deadband instead of voltage set point of the AVC relay
Advantage of this replacement: the calculation of the voltage set point is not straight forward
Normal operation: the voltage deadband is changed to ± 5%;
AVC relay sends the Tap-up or Tap-down signal to the OLTC when the voltage obtained from the voltage level analyzer block is outside the given deadband for a given time delay
AVC relay sends the Tap-up or Tap-down signal to the OLTC when the voltage obtained from the voltage level analyzer block is outside the given deadband for a given time delay
Communication-failure operation: the deadband can be changed to the default value;
The voltage set point can be changed to the voltage level at the secondary side of the transformer at the moment of communication failure or convergence problem.
The voltage set point can be changed to the voltage level at the secondary side of the transformer at the moment of communication failure or convergence problem.
Voltage regulation with reactive power compensation (RPC)
A. The control algorithm
Basic idea: RPC consumes reactive power when the voltage at wind turbine terminal is above the allowed level and injects reactive power when the voltage is below the min level.
The control logic for wind turbines regulating local voltage by RPC
The calculation of reactive power reference is shown in Fig. 3
The switching logic of Switch RPC on/off is shown in Fig.4. The switch of RPC can avoid unnecessary use of reactive power
The control logic for wind turbines regulating remote voltage by RPC
The introduction of
The modification of the measured voltage input of control loop to incorporate the voltage control of a remote wind turbine
Question:
In the situation that Vremote >1 > Vlocal, and Vremote-1>=1-Vlocal, according to Fig.6, the Vmeasure= Vmax =Vremote, and the local wind turbine will engage into the RPC control of remote voltage, it may consume more reactive power to decrease the remote voltage.
However, as the Vlocal is below the base value, the consumption of reactive power may futher decrease the Vlocal and make the situation worse for the local voltage.
In the situation that Vremote >1 > Vlocal, and Vremote-1>=1-Vlocal, according to Fig.6, the Vmeasure= Vmax =Vremote, and the local wind turbine will engage into the RPC control of remote voltage, it may consume more reactive power to decrease the remote voltage.
However, as the Vlocal is below the base value, the consumption of reactive power may futher decrease the Vlocal and make the situation worse for the local voltage.
This approach minimizes the amount of reactive power used to mitigate an overvoltage as it ensures that the remote wind turbine reactive power is only used when the local reactive power is fully utilized
B. Design of the PI controller
The cordination of voltage regulation through the SE based OLTC control & RPC
Summary
If the wind turbine regulates the voltage at its terminal, the voltage regulation of the rest of the network can be handled by the OLTC
By exempting the OLTC from regulating the highly variable voltage at the terminal of the wind turbine, one is protecting the OLTC from rapid wear and tear that would happen due to frequent tap changes
In the cases with & without data from wind turbines
if the voltage measurements from the terminals of the wind turbines are available in the SE process
the OLTC control algorithm knows exactly
the voltage level at the wind turbines. Due to the inherent
time delay present in the OLTC based voltage control, the
RPC from the wind turbine mitigates an overvoltage before the
OLTC takes any action
the voltage level at the wind turbines. Due to the inherent
time delay present in the OLTC based voltage control, the
RPC from the wind turbine mitigates an overvoltage before the
OLTC takes any action
if the voltage measurements from the terminals of the wind turbines are not available in the SE process
Thus, to coordinate the two voltage controllers
even when voltage measurement are not available from the
terminals of the wind turbines, the wind turbines should be
able to control the voltage at their terminals even at the worst
system condition
even when voltage measurement are not available from the
terminals of the wind turbines, the wind turbines should be
able to control the voltage at their terminals even at the worst
system condition
Case study system
A. Network and data descriptipon
B. Simulation set up
C. The results of analysis with the proposed control strategies
1. Voltage regulation using OLTC of the substation transformer
2. Voltage regulation using reactive power compensation from the wind turbines
3. Voltage regulation with both OLTC and reactive power compensation
Conclusion
1. The SE-based OLTC voltage control with relaxed
deadband shows a good voltage regulating capability and its
implementation is much simpler than adjusting the set point
of the OLTC
deadband shows a good voltage regulating capability and its
implementation is much simpler than adjusting the set point
of the OLTC
2. Reactive power from a wind turbine can be
used to mitigate an overvoltage that occurs locally as well
as at the terminal of any remote wind turbine on the same
feeder when the remote wind turbine has a limited reactive
power capability to mitigate overvoltage at its terminal
used to mitigate an overvoltage that occurs locally as well
as at the terminal of any remote wind turbine on the same
feeder when the remote wind turbine has a limited reactive
power capability to mitigate overvoltage at its terminal
3. In the proposed way, unnecessary tap regulation is avoided. Unnecessary
tap regulation is further avoided when the SE algorithm is
adjusted to reflect the effect of reactive power control from
the wind turbines in the distribution system
tap regulation is further avoided when the SE algorithm is
adjusted to reflect the effect of reactive power control from
the wind turbines in the distribution system
0 条评论
下一页