论文部分内容阅读
将PID经验整定公式和自适应算子引入到微分进化(DE)算法中,在此基础上提出了改进的微分进化(IDE)算法,即采用PID经验整定公式指导DE算法中初始种群的产生,减少了DE算法寻优的随机性,加快了收敛速度且不依赖调试人员的经验,实现完全自适应获得PID参数.对具有严重参数不确定性、多扰动及大迟延的循环流化床(CFB)锅炉汽温控制系统进行仿真研究,并将IDE算法应用到某300 MW CFB锅炉主汽温控制系统中.结果表明:IDE算法寻优速度快、计算量小,对参数优化非常有效;与遗传算法(GA)整定PID参数的效果相比,IDE算法具有较好的调节品质和较强的鲁棒性,汽温控制效果得到明显改善.
PID empirical tuning formula and adaptive operator are introduced into differential evolution (DE) algorithm. Based on this, an improved differential evolution (IDE) algorithm is proposed, which uses PID empirical formula to guide the generation of initial population in DE algorithm, Which reduces the randomness of DE algorithm optimization, accelerates the convergence speed and does not depend on the experience of the commissioning personnel, so that the PID parameter can be obtained completely adaptively.For CFB with serious parameter uncertainty, multi-disturbance and large delay, ) Boiler steam temperature control system is simulated and the IDE algorithm is applied to the main steam temperature control system of a 300 MW CFB boiler.The results show that the IDE algorithm is fast in search speed and less in computation cost and is very effective in parameter optimization. Compared with the effect of GA algorithm in tuning PID parameters, IDE algorithm has better regulation quality and stronger robustness, and the effect of steam temperature control is obviously improved.