神经网络预测-PID串级控制在同步发电机励磁控制中的应用  

Application of Neural Network Predictive Control in Excitation Control of Synchronous Generator

作　　者：陆桂明[1] 张运哲[1] 石瑞杰 

机构地区：[1]华北水利水电学院,河南郑州450011 [2]三门峡供电公司,河南三门峡472000

出　　处：《华北水利水电学院学报》2010年第4期72-75,共4页North China Institute of Water Conservancy and Hydroelectric Power

摘　　要：为提高发电机励磁控制系统的稳定性,分析了同步发电机的自并励励磁系统的结构和数学模型,介绍了神经网络预测控制的结构和算法,分别基于PID控制、神经网络预测控制和神经网络预测-PID串级控制算法对自并励励磁系统进行了仿真分析.通过仿真结果的对比分析,说明神经网络预测-PID串级控制在励磁控制中的应用提高了励磁系统的动态性、稳定性和抗干扰能力.The principle,structure and mathematical model of the self-shunt excitation control system of synchronous generator are analyzed, at the same time, the structure and algorithm of neural network predictive control are introduced in this paper. Self-shunt excitation system is simulated respectively based on PID control algorithm, neural network predictive control algorithm and neural network-PID cascade control algorithm. The comparative analysis of simulation results shows that neural network-PID cascade control algorithm used in the Excitation Control has potential and advantage.

关 键 词：神经网络 预测控制 励磁控制 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] TM341[自动化与计算机技术—计算机科学与技术]