摘要
火电厂过热汽温控制系统具有大惯性、大延迟和时变等特性,采用常规串级PID控制方法的过热汽温系统难以获得满意的控制效果。本文设计了基于RBF神经网络辨识的可在线调整PID参数的过热汽温控制系统。在仿真实验的基础上,对基于神经网络的过热汽温控制和最优常规PID控制进行了比较和分析。仿真结果表明,基于神经网络控制策略能够充分利用神经元自学习、自适应的能力使系统的控制品质提高。
Since fresh steam temperature control systems of fossil power plants are characterized by large inertia, long tune lag and variation with tune, satisfying control effect can not be achieved with fresh siemn temperature systems which use the normal cascade PID control methods. This article designed a fresh steam temperature control system based on RBF neural network which can optimizes PID parameters on-line. On the basis of simulation, a compare was made between neural network control and normal cascade PID control. The results of simulation dedicates that the nueral network contrul strategy can make full use of neurons self-learing, self-adaptivity to improve the control quality of the system.
出处
《微计算机信息》
2010年第16期4-5,11,共3页
Control & Automation
关键词
大惯性
神经网络
PID控制
过热汽温
large inertia
nueral network
PID control
fresh steam temperature