基于混沌演化的蒸汽动力装置负荷控制优化

为提高蒸汽动力装置负荷控制系统适应机组负荷变化的能力,将混沌理论和演化算法相结合,基于系统辨识提出一种PID控制参数的在线优化方法.该方法利用演化算法的全局最优性和混沌的遍历性实现对控制器参数的在线整定,避免了传统PID控制算法受控制对象影响较大,只有在各参数整定良好的前提下才能得到满意控制效果的缺点.仿真试验证明:采用混沌演化优化后的控制系统在保留传统PID控制结构简单、易于实现、可消除静差等优点的同时,还具有超调小、响应迅速、抗干扰能力强等特点,适用于工况频繁变化场合下的自适应控制.

Optimization of load controller for steam power plant using evolutionary algorithm and chaos

In order to improve the load control effect of steam power plant units, an on-line PID control parameters optimization method that using evolutionary algorithm and chaos was proposed based on system identification. This method adopts the global optimization of evolutionary algorithm and the ergodicity of chaos to achieve the on-line optimization for PID parameters, conquering the shortcoming that the traditional PID tuning method is greatly influenced by the controlled objects and can only obtain satisfying control effect when every paramete has been set correctly. The simulation shows that the optimized control system adopting evolutionary algorithm and chaos retains the excellences of the traditional PID control such as simple structure, easily implemented, no steady error and so on, and possesses low overshoot, fast response and good anti-jamming capability at the sametime, so it is suitable for the self-adaptive control of the units which need frequent load changes.