摘要
火电厂主汽温具有大惯性、大迟延等特性,其动态特性随负荷而变化,采用常规的按照典型工况整定的固定参数PID串级控制难以获得满意的控制效果。为此,提出一种用RBF网络整定的PID串级主汽温控制策略,将RBF神经网络和常规PID串级控制相结合构成的智能PID控制器不仅具有常规PID控制器的特性,而且具有智能控制器的自学习能力,增强了系统对不确定因素的适应性。仿真研究结果表明:系统动态品质明显优于通常的PID串级控制,能适应对象参数的变化,具有较强的鲁棒性和自适应能力。
Fresh steam temperature variations in thermal power plants are characterized by large inertial time-delay. The dynamic characteristics vary with the unit's load. Satisfactory control effects can hardly be obtained with conventional PID cascade control method tuned at typical operating conditions. Therefore, a novel PID cascade control strategy with RBF network tuning is being proposed, according to which an intelligent PID controller is formed by combining RBF neural network with conventional PID cascade control, inheriting thus not only the advantages of conventional PID cascade controlling, but also gaining the self-study ability of intelligent controllers in addition. The system's adaptability to uncertainties is herewith increased. Simulation results evidently show that the system has a dynamic performance is superior to that of the conventional PID cascade control and can readily accommodate itself to variations of the object' s parameters. It is featured by strong robustness and self-adaptability.
出处
《动力工程》
EI
CSCD
北大核心
2006年第1期89-92,134,共5页
Power Engineering
关键词
自动控制技术
RBF网络
PID控制
主汽温控制系统
串级控制
参数整定
automatic control technique
RBF network
PID control
fresh steam temperature control system
cascade control
parameter self-tuning