A new voltage control equipment using fuzzy inference

Power system control equipment needs higher sensitivity and operational reliability. Advanced voltage control equipment is needed for reducing the frequency of tap changes and improving the characteristics (the relationship between the actual voltage and reference voltage) of the voltage to meet today's power system requirements. However, these objectives are in a trade-off relationship. Studies of voltage control derived from a knowledge base suitable for electric power systems can satisfy these objectives using fuzzy inference. Compared with corresponding conventional equipment, the new equipment improved the deviation of 30 min average voltage of 30 percent. This paper describes the design concept of new voltage control equipment using fuzzy inference. In addition, field test results are described along with rules of fuzzy inference, membership functions, and the deviation of 30 min average voltage through detailed simulation.     

Stabilizing control of power system using fuzzy control

This paper presents an application of fuzzy control to enhance power system stability. The proposed control consists of the controller for large disturbance (FU 1), the fuzzy controller for small disturbance (FU 2), and the fuzzy judgment mechanism (FU 3). FU 1 is determined based on the fuzzy controller [FU 1(F)] is determined according to the control rules and its input signals, i.e., speed deviation and acceleration at every sampling time of the machine. FU 2 consists of two controllers, namely, FU 2-ω and FU 2-P; FU 2-ω has the same mechanism as FU 1, while the output signal of FU 2-P is determined according to the rules together with the change of error of electrical power and terminal voltage. To obtain the optimal desired control signal during both the large and the small disturbances, the operations of FU 1 and FU 2 are judged by FU 3, where the magnitude of speed deviation is chosen as its input signal. The determined control signal is fed to AVR of the machine. The implementation of the proposed control is simple due to the small amount of calculations and required data. The effectiveness of the proposed control is demonstrated by the one-machine infinite-bus system model and very good system performance is obtained throughout all the simulations.     

Voltage and reactive power control using recurrent neural networks

Many controllers using neural networks have been recently developed. The hybrid and direct types are two major categories of such controllers. While the first type tunes the parameters of the conventional controller by means of neural networks, the latter constructs the controller by learning the inverse dynamics of the control target. Electric power systems require voltage and reactive power (VQ) control to avoid voltage collapse. The conventional VQ control, however, meets this requirement unsatisfactorily because the control is only approximate. In this paper, we will propose a new algorithm for VQ control using recurrent neural networks (RNNs), which have the ability to deal with the system's controlled target by means of RNNs. Second, we will apply this algorithm to VQ control. We will call this controller “neuroVQC.” Finally, the usefulness of the neuroVQC will be shown by comparison with the conventional VQ controller.     

Control strategy of leveling load power fluctuations based on fuzzy neural networks by tuning coefficients of learning rate with genetic algorithm

The effective usage of the power facilities can be realized by leveling the fluctuating active power and compensating the reactive power. The fuzzy and fuzzy neural network control strategy of superconducting magnetic energy storages (SMES) was proposed for this purpose. The control results depend on the values of coefficients of learning rate in fuzzy neural networks. Therefore, it is desirable to obtain better control results by tuning the coefficients of learning rate to their optimum values. In this paper, the control strategy based on an autotuning of scaling factors with neural network and tuning of coefficients of learning rate of neural network with genetic algorithm is proposed for leveling load fluctuations. Encoding and decoding of coefficients of learning rate and selection, crossover, and mutation within genetic operations are shown, and crossover rate and mutation rate are discussed. Through these methods, we can achieve a better leveling of load power fluctuation by using fuzzy neural network with genetic algorithm. © 1998 Scripta Technica, Electr Eng Jpn, 125(1): 65–72, 1998     

基于模糊神经网络的机器人自学习控制

研究了一种模糊神经网络与传统PD控制相结合的机器人学习控制系统,该控制系统具有自学习、自适应、控制精度高等特点。     

基于神经网络辨识的模糊预测函数控制

针对生产过程中存在的滞后性、时变性、不确定性和变工况等特点及预测函数控制中模型失配的影响的情况,提出了基于神经网络辨识参数、通过模糊推理对控制量进行补偿的解决方案。并将基于神经网络辨识的模糊补偿预测函数控制应用于锅炉燃烧控制系统,通过连续系统仿真,结果表明这种控制器具有较强的鲁棒性。     

基于模糊RBF神经网络自学习的不确定机器人模糊自适应控制

针对机器人系统的不确定、非线性特点，设计了模糊RBF神经网络控制器学习机器人系统的不确定性上界，并利用模糊推理机产生的分目标学习误差进行训练，避免了采用系统直接输出反馈误差进行训练所存在的权值饱和与过调整问题。此外，在反馈回路还设计了固定比例增益控制器FC，起着监督的作用，对系统实施渐近稳定的控制。仿真结果表明这种控制方案实现了对机器人系统的高精度控制。     

Power oscillation damping by superconducting magnetic energy (SMES) storage unit

With the increase in the size and capacity of electric power systems and the growth of widespread interconnections, the problem of power oscillations due to the reduced system damping has become increasingly serious. Since a Superconducting Magnetic Energy Storage (SMES) unit with a self-commutated converter is capable of controlling both the active (P) and reactive (Q) power simultaneously and quickly, increasing attention has been focused recently on power system stabilization by SMES control. This paper describes the effects of SMES control on the damping of power oscillations. By examining the case of a single generator connected to an infinite bus through both theoretical analyses and experimental tests (performed with a SMES unit with maximum stored energy of 16 kJ and an artificial model system), the difference in the effects between P and Q control of SMES is clarified as follows:

• 1 In the case of P control, as the SMES unit is placed closer to the terminal of the generator, the power oscillations will decay more rapidly.
• 2 In the case of Q control, it is most effective to install the SMES unit near the midpoint of the system.
• 3 By comparing the P control with Q control, the former is more effective than the latter based on the conditions that the SMES unit location and the control gain are the same.

     

改进模糊控制的光伏发电系统最大功率点跟踪

研究了光伏发电系统最大功率点跟踪的问题,由于其存在着随机性,且往往不够充分与准确,容易导致系统稳态剧烈震荡或无法准确跟踪。鉴于传统人工总结模糊控制规则难度高,提出了模糊神经网络控制算法,将T-S模糊推理方法与神经网络理论相结合,选择混合法作为训练方法,网格法作为生成算法,由实测数据自动生成模糊控制规则,将其嵌入到模糊控制器当中,从而实现了MPPT控制功能。仿真结果表明,采用该方法生成的模糊规则实用准确,系统稳态性能与动态性能均十分优越。实验证明人工神经网络法与模糊控制技术相结合,实现光伏发电MPPT高效准确。     

A decentralized control system for stabilizing multimachine power systems

A decentralized control system is studied for stabilizing multimachine power systems. A longitudinal power system with three areas, each having one machine, is considered in this study. A decentralized control design method is proposed, which is based on the optimal regulator theory. First a centralized control system is designed without any consideration on whether state variables are all available or not. Second a pseudo-decentralized control system is designed by omitting control gains corresponding to state variables which give hardly any effects on the power system stability. It is found that only one variable of phase angle of each machine is absolutely necessary for the pseudo-decentralized control system. This leads to an idea based on power system engineering, that is to say, new variables of tieline power flow are introduced in the decentralized control system design to substitute for the phase angle of each machine. Thus a decentralized control system for power system stability can be designed using the new variables of tieline power flow. It is demonstrated from simulation studies that the decentralized control system improves even longitudinal power system stability as well as the centralized control system.     

基于神经网络的模糊控制在变电站综合控制中的应用

针对目前变电站电压 无功综合控制存在的不足 ,将基于神经网络的模糊控制应用于传统的系统分区 ,充分利用模糊控制与神经网络的各自优势 ,进行基于模糊无功边界的综合控制 ,有效减少了有载分接头调节次数 ,提高了系统的调节性能和电压质量。     

Enhancement of transient stability by fuzzy logic-controlled SMES considering communication delay

This paper presents a fuzzy logic-controlled superconducting magnetic energy storage (SMES) for the enhancement of transient stability in a multi-machine power system. The control scheme of SMES is based on a pulse width modulation (PWM) voltage source converter (VSC) and a two-quadrant DC–DC chopper using gate-turn-off (GTO) thyristor. Total kinetic energy deviation (TKED) of the synchronous generators is used as the fuzzy input for SMES control. Communication delays introduced in online calculation of the TKED are considered for the actual analysis of transient stability. Global positioning system (GPS) is proposed for the practical implementation of the calculation of the TKED. Simulation results of balanced fault at different points in a multi-machine power system show that the proposed fuzzy logic-controlled SMES is an effective device for transient stability enhancement of multi-machine power system. Moreover, the transient stability performance is effected by the communication delay.     

Adaptive control of Wiener-type nonlinear systems using neural networks

This paper proposes new adaptive control schemes with neural networks for Weiner-type nonlinear systems which have output nonlinearity. First, by adopting a robust adaptive control law and a functional link network (FLN), we present an adaptive linearizing scheme as a primary step for a model reference control scheme, where the FLN compensates the output nonlinearity. Second, we analyze the stability of the adaptive linearizing scheme by using a robust adaptive control technique, and demonstrate that all of the parameters are bounded and that the boundedness of all of the signals in the closed loop is guaranteed under some reasonable conditions. Third, based on the linearizing scheme, we present a new direct model reference adaptive control scheme by choosing the reference output appropriately. The stability of the system is guaranteed under several conditions in a similar manner. Finally, we illustrate the effectiveness of the proposed scheme through some numerical examples. © 1998 Scripta Technica. Electr Eng Jpn, 122(1): 37–48, 1998     

Development of peak load forecasting system using neural networks and fuzzy theory

This paper presents a peak load forecasting system using multilayer neural networks and fuzzy theory. Electric load forecasting in power systems is a very important task from the perspective of reliability and economic operation. Daily peak load forecasting is one of the basic operations of generation scheduling for the following day. Therefore, many statistical methods have been developed and used for such forecasting even though it has been difficult to construct a proper functional model. The developed system is applied by neural network and fuzzy theory to forecast for daily, weekly and monthly peak load. The system consists of an engineering workstation (EWS) and a personal computer (PC). The EWS is for learning and data-bases, and the PC is for man-machine interface such as forecasting operation. The system has been used since June 1993. The result evaluated with an absolute mean error is 1.63 percent for 10 months. From the results shown here, the system applied by neural network and fuzzy theory has high validity.     

二级倒立摆的递阶模糊神经网络控制

为了表明模糊神经网络控制器比较适合于控制快速、多变量、强非线性、绝对不稳定系统，可以克服用模糊神经网络控制多变量系统时的规则组合爆炸问题。本文提出用递阶模糊神经网络控制二级倒立摆。这种方法可以有效地减少多变量输入的模糊神经网络控制器的规则数，有利于利用专家的控制经验初始化网络参数，从而有利于下一步利用遗传算法对其进行优化。实验结果表明：与线性最优控制相比，本文方法的控制效果好、鲁棒性强。     

Intelligent control using multiple models and neural networks

Most adaptive control algorithms for nonlinear discrete time systems become invalid when the controlled systems have non‐minimum phase properties and large uncertainties. In this paper, an intelligent control method using multiple models and neural networks (NN) is developed to deal with those problems. The proposed control method includes a set of fixed controllers, a re‐initialized neural network (NN) adaptive controller and a free‐running NN adaptive controller. The bounded‐input‐bounded‐output (BIBO) stability and performance convergence of the system are guaranteed by the free‐running adaptive controller, while the multiple fixed controllers and the re‐initialized adaptive controller are used to improve the transient response. Simulation results are presented to demonstrate the effectiveness of the proposed method. Copyright © 2007 John Wiley & Sons, Ltd.     

基于神经网络的油田l电力网无功功率管理技术

油田电力网在传输无功功率时产生了巨大的员耗。针对这一问题,本文应用一扩展Ｈｏｐｆｉｅｌｄ神经网络－耦合梯度网络,建立了油田电力网无功功率管理的全局优化的数学模型。仿真结果表明,通过这一网络模型的优化计算,可以获得可观的经济效益。     

基于模糊神经解耦控制的双馈水轮发电机系统仿真

双馈水轮发电机系统是一个涉及水力、水轮机和发电机的综合复杂系统。针对系统具有多变量、非线性、强耦合和参数不确定性的特点，本文提出了一种两级串联结构的自适应模糊神经网络解耦控制策略，前级为基于智能权函数规则的自调整模糊控制器，后级为基于动态耦合特性的自适应神经网络解耦控制器，并从理论上证明了学习算法的收敛性。为了验证所提出控制策略的有效性和正确性，本文对双馈水轮发电机系统在水力、水轮机和发电机参数变化时的鲁棒性分别进行了仿真研究。与常规PID控制的仿真结果比较表明，提出的解耦控制策略能较好地克服参数变化和对象模型结构变化对运行性能的影响，具有鲁棒性好，解耦能力强的优点。     

基于模糊神经网络PID算法的液位控制系统研究

在工业生产中,液位控制系统得到了广泛应用,但是对于这种大滞后、非线性的复杂控制系统,传统的PID控制方法存在着参数整定困难,控制效果不理想的缺陷。在对传统的PID算法、模糊控制算法和神经网络算法研究的基础上,提出了一种将模糊神经网络PID算法应用到液位控制系统中去的解决方案,并采用MATLAB对液位对象控制进行仿真实验,同时采用A3000型水箱实验平台对仿真实验结果进行验证。研究结果表明,基于模糊神经网络的PID算法的液位控制系统在调整时间和超调量上都优于传统的PID算法,控制效果和抗干扰能力更强,克服了传统PID算法的不足。     

基于模糊神经网络的交流伺服系统

提出了一种基于神经网络的模糊控制交流伺服系统,将神经网络与模糊逻辑结合起来,输入信号先模糊化,然后通过构建的神经网络,在线调整其权值和变化的控制参数,使系统的输出具有更好的动、静态性能,提高了系统的鲁棒性。仿真实验证明了这种基于神经网络模糊控制方法在交流伺服系统中应用的可行性和可靠性。