主汽温串级控制系统PSO优化方法的工程实践

Engineering Practice of Cascade Control System for Main Steam Temperature Using PSO Optimizing Method

常用的串级PID主汽温控制策略的控制品质在很大程度上取决于PID参数的整定。为获得最优控制参数,本文结合现场实际,给出了使用粒子群优化算法PSO对串级控制回路中主、副PI控制器同时进行优化的方法,指出了使用该方法时必须注意传递函数的选择、PID正反作用的判断及目标函数的确定3个方面的问题。仿真结果表明,优化后的控制器能有效克服原控制系统中由于参数设置不当导致的调节时间过长和执行机构频繁动作的缺点。实际投运效果也表明了参数优化后的主汽温控制效果取得了明显改善。

The cascade PID control is a commonly used strategy to control main steam temperature in the engineering field and its control quality depends largely on the parameters tuning.On the basis of practical situation,a method of simultaneously optimizing the parameters of main and secondary PI controllers using PSO（Particle swarm optimization）is proposed in this paper to acquire optimizing control parameter.In addition,three issues,which are the choice of transfer function,judgment of positive and negative PID and determination of objective function,are pointed out respectively when using this kind of method.Simulation results show that the optimized controller can overcome shortcomings of long settling time and frequent action of actuator caused by improper parameters setting in the original control system.Practical operation effects also show the control effect of main steam temperature is improved greatly after parameter optimization.