基于神经网络的自适应电力系统稳定器  

A NEURAL NETWORK BASED ADAPTIVE POWER SYSTEM STABILIZER

作　　者：谭振宇[1] 程时杰[1] 魏璇 李江[1] 文劲宇[1] 

机构地区：[1]华中科技大学,武汉430074

出　　处：《电力系统自动化》2000年第15期1-5,36,共6页Automation of Electric Power Systems

基　　金：国家重点基础研究专项经费资助项目!(G1998020319)

摘　　要：摘要 :为彻底解决常规自校正控制器中受控系统参数难以识别的问题 ,提出了一种可以根据受控系统运行工况自动调整控制参数的新型自适应控制。在这种自适应控制中 ,人工神经网络被用来对受控系统的运行状态进行辨识。根据系统辨识的结果 ,控制器可按照传统 PID控制器的设计原则自动调整控制器参数 ,达到最佳的控制效果。根据所提出的原理 ,设计了一种变参数自适应 PID型电力系统稳定器 (NNPIDPSS)。将设计的 NNPIDPSS用于单机—无穷大电力系统和具有多模振荡特性的多机电力系统 。For overcoming the shortcoming that parameters of the controlled syste m can't be properly identified inconventional self-adjustable controller, a new adaptive control is proposed. It can self-adjust the control parameters accordingto the operating conditions of the controlled system. In the proposed control, a neural network is used to identify theparameters of the transfer function of the controlled system associated with the operating conditions. According to theidentification results, the proper control parameters are acquired based on a PI D control strategy. An self-adaptive PID PSScyNPIDPSS) is designed by use of the above--mentioned method. The applications o f this NNPIDPSS in a single-machineinfinite--bus power system and a multi--machine power system with multi--mode os cillations are simulated. The simulationresults show the excellent performance of NNPIDPSS.This project is supported by National Key Basic Research Special Fund of China ( No. G1998020319).

关 键 词：电力系统稳定器 PID控制 神经网络 

分 类 号：TM44[电气工程—电器]