On-line voltage stability assessment using radial basis function network model with reduced input features

In recent years, voltage instability has become a major threat for the operation of many power systems. This paper presents an artificial neural network (ANN)-based approach for on-line voltage security assessment. The proposed approach uses radial basis function (RBF) networks to estimate the voltage stability level of the system under contingency state. Maximum L-index of the load buses in the system is taken as the indicator of voltage stability. Pre-contingency state power flows are taken as the input to the neural network. The key feature of the proposed method is the use of dimensionality reduction techniques to improve the performance of the developed network. Mutual information based technique for feature selection is proposed to enhance overall design of neural network. The effectiveness of the proposed approach is demonstrated through voltage security assessment in IEEE 30-bus system and Indian practical 76 bus system under various operating conditions considering single and double line contingencies and is found to predict voltage stability index more accurate than feedforward neural networks trained by back propagation algorithm and AC load flow. Experimental results show that the proposed method reduces the training time and improves the generalization capability of the network than the multilayer perceptron networks.     

Wide-area measurement-based voltage stability sensitivity and its application in voltage control

This work proposes a measurement-based voltage stability index, namely wide-area measurement-based voltage stability sensitivity based on L index. The sensitivities of L index with respect to nodal reactive power (Q) and real power (P) injections are first derived. The derived L-Q and L-P sensitivities analyze the impact of nodal injection to nodal voltage stability and can help determine the reactive power compensation and emergency load shedding amount for voltage stability control. To improve the computational efficiency, a simplified L index, L′, along with its sensitivities with respect to nodal reactive and real power injection (L′-Q and L′-P sensitivities) are derived which makes the proposed approach suitable for the practical large-scale systems. Moreover, a control strategy for voltage stability is proposed based on the L-Q, L-P, L′-Q, and L′-P sensitivities. The proposed sensitivities and control strategy are tested on the New England 39 bus system and the IEEE 118 system. Test results on both systems verify the proposed sensitivities and the control strategy by demonstrating their accuracy and robustness in voltage stability assessment and control. In conclusion, the proposed measurement-based sensitivities can be applied to voltage stability assessment and control by using the wide-area measurements.     

自组织映射神经网络用于动态电压稳定分析的新方法

介绍了一种利用人工神经网络(ANN)进行动态电压稳定分析的新方法。这种多层自组织网络(SHNN)综合利用了自组织映射网络(文中使用Kohonen网络)和多层感知机网络（MLP）。Kohonen网络把输入样本按运行条件的相似性进行聚类,从而使MLP网络的性能得到提高。使用2个SHNN模型,一个用于判定电力系统是否处于动态稳定,另一个预测动态稳定系统的PQ节点的电压幅值。通过动态模拟得到训练样本。最后对WSCC 9节点系统和New England 39节点系统进行数字仿真,证明了该方法的有效性。     

基于功率传输路径的在线电压稳定性评估新方法

电压失稳本质上是一种局部现象,系统的电压稳定程度可以由最易于电压失稳的功率传输路径的电压稳定性来表征。文中应用局部电压稳定性指标确定薄弱负荷节点,并借助电气距离信息,从无功、有功传输路径两方面确定路径参与节点和关键功率源点,得到系统功率传输的薄弱路径集,通过等值判别最弱功率传输路径,将该路径的电压稳定性指标与关键发电机的无功储备指标相结合,评估整个系统的电压稳定程度。新英格兰10机39节点算例系统的仿真结果表明,该方法速度快、精度高,适合于在线应用。     

分布式电源接入配网对其静态电压稳定性影响多角度研究

分布式电源(Distributed Generation, DG)接入配网后,将对配网静态电压稳定性造成影响。基于IEEE33节点配电网络和Matlab仿真测试软件平台,考虑分布式电源的接入数量、接入位置、接入容量和功率因数4个影响因素。应用电压稳定性判定指标L,对比单个DG接入单分支线路和多个DG接入多条分支线路两种情况,得出各自对配网电压分布和电压稳定性造成的影响。对比测试所得L指标,结果表明多DG接入的方式更有利于配网电压稳定。该研究对配网中多DG的布局和规划具有积极的指导意义。     

An approach for real power scheduling to improve system stability margins under normal and network contingencies

In the present day power system planning and operation, considerable interest is being shown in system security and stability analysis. Pattern of load sharing/generation scheduling that results in heavy flows tend to incur greater losses, threaten stability, security and ultimately making certain generation patterns undesirable. Generation schedules mainly based on economic criteria may lead to lower reserve margins and therefore diminished reliability is a serious concern for the systems. With increased loading of existing power transmission systems, the problem of voltage stability and voltage collapse has also become a major concern in power system planning and operation. While the voltage stability is more dependent on the reactive power sources/voltage profile in the system, it is also a function of real power flows. In this paper, network sensitivity between load voltages and source voltages to compute voltage stability index (L), is used as the basis to evaluate desirable load sharing for improving stability margins. The proposed method has been tested on typical sample systems and also on a practical 24-bus equivalent power system, and results are presented to illustrate the proposed approach.     

An economical generation direction for power system static voltage stability

This paper proposes an economical generation direction for static voltage stability. The proposed generation direction minimizes the total operating cost at any loading level, up to the point of collapse. The proposed approach, named as EGD approach, is based on the economic load dispatch with load flow. Two alternative methods to identify the economic generation direction are proposed. Economical generation direction is used in the continuation power flow process to obtain the static voltage stability margin. Static voltage stability margin with economical generation direction is compared with the margins resulting from other existing generation directions in the modified IEEE 14-bus test system under various system operating conditions. Cost of providing loading margin as well as PV curves and losses under different generation directions resulting from various generation direction approaches are also studied and compared. The proposed approach provides an alternative way for utilities to obtain the lowest total operating cost with voltage stability constraint by using any existing commercial software.     

基于粒子群算法并计及静态电压稳定性的电力系统无功规划

提出了基于改进粒子群优化算法的无功规划方法,该方法有效提高了初始粒子的质量,消除了传统粒了群优化算法的边界振荡现象,利用电压稳定裕度指标对系统负荷节点进行排序,选择其中电压稳定性较薄弱的节点作为无功补偿装置的安装点,将电压稳定裕度最大作为无功规划的目标函数之一,以改善系统的静态电压稳定性。该规划方法在IEEE30节点系统的应用结果表明其在降低网损的同时能够有效提高负荷节点的电压稳定裕度。     

Online voltage stability assessment of load centers by using neural networks

This paper presents a neural network based method for evaluating online voltage stability conditions for a selected load center of an electric power system. Starting with a dynamic model of the system, a suitable index is defined to evaluate the proximity of the power network to voltage collapse. Then, a three-layer feedforward neural network is trained to give, as output to a prespecified set of inputs, the expected value of the voltage stability index. For this purpose, two different neural network architectures are proposed. The error back-propagation algorithm is used in this paper to train the chosen neural network structure. Moreover, it is shown that a good estimate of the real power margin of the selected load center can also be obtained using the value of the output of the designed neural network. To demonstrate the effectiveness of the proposed neural network based approach for voltage stability monitoring, a sample power system is considered. Test results show that neural networks can yield, in real time, an accurate assessment of voltage stability conditions.     

Power system transient stability margin estimation using neural networks

This paper proposes a methodology for estimating a normalized power system transient stability margin (ΔV_n) using multi-layered perceptron (MLP) neural network with a fast training approach. The nonlinear mapping relation between the ΔV_n and operating conditions of the power system is established using the MLP neural network. The potential energy boundary surface (PEBS) method along with a time-domain simulation technique is used to obtain the training set of the neural network. Results on the New England 10-machine 39-bus system demonstrate that the proposed method provides a fast and accurate tool to evaluate online power system transient stability with acceptable accuracy. In addition, based on the examination of generators rotor angles after faults, a method is presented to select the power system operating conditions that most effect the ΔV_n for each fault.     

Voltage stability enhancement using VSC-OPF including wind farms based on Genetic algorithm

This paper describes the impact of wind generation to enhance the voltage stability of power system in optimal power flow problem. In this work, the Voltage Stability Constrained Optimal Power Flow (VSCOPF) algorithm is used, which combines the economical aspects and the voltage stability control of power system networks. A voltage stability index called L-index has been utilized to identify the most sensitive node prone to voltage collapse. Here Improved Genetic Algorithm (IGA) with mixed form of representation has been used for multi-objective formulation. Real power setting and voltage magnitudes are represented as floating point numbers and transformer tap settings and capacitors as integers. For effecting genetic processing, crossover and mutation operator, which can be directly deal with floating point number and integers, are used. The IGA has been implemented on IEEE 30 bus system to study the impact of wind farm on voltage stability. The obtained simulation shows the effectiveness of Improved Genetic algorithm on voltage stability of the power systems.     

Model reference adaptive sliding mode control using RBF neural network for active power filter

In this paper, a model reference adaptive sliding mode (MRASMC) using a radical basis function (RBF) neural network (NN) is proposed to control the single-phase active power filter (APF). The RBF NN is utilized to approximate the nonlinear function and eliminate the modeling error in the APF system. The model reference adaptive current controller in AC side not only guarantees the globally stability of the APF system but also the compensating current to track the harmonic current accurately. Moreover, a sliding mode voltage controller based on an exponential approach law is designed to improve the tracking performance of DC side voltage. Simulation results demonstrate strong robustness and outstanding compensation performance with the proposed APF control system. In conclusion, MRASMC using RBF NN can improve the adaptability and robustness of the APF system and track the given instructional signal quickly.     

电力系统电压安全灵敏度的分析方法研究

针对目前的电力系统薄弱节点识别方法普遍存在难以适应互联大电网电压稳定计算和分析的问题,提出一种基于当前状态下的电压安全灵敏度分析方法。利用雅可比矩阵数学模型建立了评价节点薄弱程度的指标,对补偿前后的节点系统电压偏移量进行比较分析,依据薄弱节点的排序来确定系统的无功补偿方案。以IEEE14节点系统为例进行电压安全灵敏度计算分析,结果表明,所提出的薄弱节点指标能有效提取系统的薄弱节点,在薄弱节点进行无功补偿时系统电压幅值提高最大。     

用正规形方法研究负荷节点对电压稳定性的影响

不同的负荷节点对电压稳定性具有不同的影响程度。应用向量场正规形理论,分析电力系统潮流方程。提出以节点电压非线性参与因子作为依据,衡量负荷节点对电压稳定性的影响程度。所提出的方法可计及电力系统非线性特性对电压稳定性的影响,与线性化分析方法相比,该方法在系统具有强非线性特性的条件下,仍能准确识别负荷节点的重要程度。将所提出的方法用于New England 39节点系统,研究系统中负荷节点对电压稳定性的影响,通过对系统电压稳定性指标的比较,验证所提出方法的有效性。     

Recurrent modified Elman neural network control of PM synchronous generator system using wind turbine emulator of PM synchronous servo motor drive

The recurrent modified Elman neural network (NN) controlled a permanent magnet (PM) synchronous generator system, which is driven by wind turbine emulator of a PM synchronous motor servo drive, is developed to regulate output voltage of rectifier (or AC to DC power converter) and inverter (or DC to AC power converter) in this study. First, the wind turbine emulator of a closed loop PM synchronous motor servo drive is designed to produce the maximum power for the PM synchronous generator system. Then, the rotor speed of the PM synchronous generator, the output DC bus voltage and current of the rectifier are detected simultaneously to yield maximum power output of the rectifier through DC bus power control. Because the PM synchronous generator system is a nonlinear and time varying dynamic system, the online training recurrent modified Elman NN control system is developed to regulate DC bus voltage of the rectifier and AC line voltage of the inverter in order to improve the control performance. Furthermore, the online training recurrent modified Elman NN control system with the variable learning rate is derived based on Lyapunov stability theorem, so that the stability of the system can be guaranteed. Finally, some experimental results are verified to show the effectiveness of the proposed recurrent modified Elman NN controlled PM synchronous generator system.     

Current state of neural networks applications in power system monitoring and control

For over two decades Neural Network (NN) has been applied to power system monitoring and control. Conventional controllers suffer from certain limitations which NN as an Artificial Intelligence (AI) technique is able to overcome. Therefore, many researchers prefer to use NN technique in the monitoring and control of power systems. This paper reviews published recently schemes for control and monitoring based on NN. The performance of various NN controllers is compared with one another as well as to the performance of other types of controllers. This review further reveals that the design of a proper NN control can maintain first-swing stability, damp oscillation, ensure voltage stability and the reliable supply of electric power.     

采用正规形方法确定重要负荷节点排序

电网结构是电力系统安全稳定运行的一个重要因素。为研究不同负荷节点对电压稳定性所具有的不同影响程度,首先应用向量场正规形理论分析电力系统潮流方程,提出了以节点电压非线性参与因子作为依据衡量负荷节点影响电压稳定性的程度的方法。该方法可计及电力系统非线性特性对电压稳定性的影响,因此与线性化分析方法相比,该方法在系统具有强非线性特性的条件下,仍能准确识别负荷节点的重要程度。然后用该方法研究了New England 39节点系统中不同负荷节点对电压稳定性的影响,通过对系统电压稳定性指标的比较,验证了所提出方法的有效性。     

Simple and efficient method for steady-state voltage stability assessment of radial distribution systems

A new static voltage stability index of a radial distribution system is proposed to faithfully evaluate the severity of the loading situation, thereby predicting for voltage instability at definite load value. The proposed index includes different parameters which affect the steady-state voltage stability of distribution systems, therefore it gives accurate results. The maximum value of 1 of that index denotes the point where the system reaches the point of collapse whereas a minimum value of 0 shows the state of no load. The performance of the new index was tested on two radial distribution systems consisting of 33 and 85 buses. Comparison between the results of the new index and those of previous indices showed that the new index yielded reliable results in predicting voltage stability condition of the system. The new index overcomes the problem which faces many previous indices especially as the load approaches the critical point. Analysis of the two-bus equivalents of the test systems under different scenarios is also presented. A new P–Q plane of stability is introduced based on the equation of the proposed index. The active, reactive and apparent power margins are then directly determined from the voltage stability boundary.     

STRONG CONTROLLABILITY AND OBSERVABILITY AND THEIR EFFECTS ON TRANSIENT STABILITY OF POWER SYSTEMS

ABSTRACT

An algorithm is developed for identifying bus clusters called control areas that have strong local controllability and observability property for measurements and controls at the buses or on generators within the control area. Two types of control areas are identified; one that is based on weak boundaries of the reactive power voltage jacobian and is called voltage control area; the other is based on the jacobian of both real and reactive power balance equations with respect to voltage and angle at network buses. The two control areas are identical under light load conditions, but are different as the system approaches voltage collapse or poorly damped low frequency oscillations. The understanding of strong controllability and observability in terms of real power angle dynamics and reactive power voltage dynamics (voltage control areas) and the strong controllability and observability of voltage and angle dynamics (control areas) helps understanding of voltage instability and inter-area oscillations.     

A neural network-based method for voltage security monitoring

In this paper, a neural network-based method is proposed for monitoring the online voltage security of electric power systems. Using a dynamic model of the system, voltage stability is measured totally, considering a suitable stability index for the whole system, and locally, by defining appropriate voltage-margins for detecting the area of the system where the instability phenomenon arises. A three-layer feedforward neural network is trained to give, as outputs to a pre-defined set of input variables, the expected values of the above defined indices. The neural network is designed by using a fast learning strategy that allows the optimal number of hidden neurons to be easily determined. Moreover, it is shown that, in the operation mode, the system power-margin and the bus power-margins can be easily evaluated using the value of the voltage stability index given by the designed NN. The effectiveness of the proposed approach has been demonstrated on the IEEE 118-bus test system