摘要
对于参数可变的时变系统和非线性复杂系统,常规PID控制器不能获得理想的控制效果,针对复杂非线性对象的神经网络PID控制不失为1种有效的控制策略.根据神经网络初始权值的选取影响控制器性能的特点,提出了基于遗传算法优化参数的神经网络PID控制器,实现了基于实数编码的GA参数优化.仿真结果证明了该算法的有效性.
For the nonlinear and time-varying parameters systems, the traditional PID controller can't get good control effect. So the neural network PID controller is one of effective methods. The initial weight values of neural network PID affected the performance of controller. A neural network PID controller optimized by a new improved Genetic Algorithim(GA) based on the real number code is proposed in this paper. The simulation results show the validity of the proposed algorithms.
出处
《北华大学学报(自然科学版)》
CAS
2004年第5期462-465,共4页
Journal of Beihua University(Natural Science)
基金
国家自然科学基金项目(60274017)
国家教委博士点基金项目(20011045023)
沈阳市自然科学基金项目(1022033 1 07)