论文部分内容阅读
为解决模糊多变量控制中的“维数灾”问题 ,提出了一种 4输入 1输出的分层模糊控制器 ,推导了它的解析表达式。依据分层模糊控制器的解析表达式 ,提出了其设计参数的模型参考自学习方法 ,其中参考模型选取典型二阶系统 ,用梯度法设计自学习律 ,在线学习分层模糊控制器的可调参数 ,以达到最优值。为检验所提出控制策略的有效性 ,将其应用到板球系统的轨迹跟踪问题 ,并进行了仿真实验。仿真实验结果表明 ,这种分层模糊控制器的可调参数少 ,学习算法收敛快 ,控制效果好 ,具有较强的实用性。
In order to solve the problem of “dimensionality disaster” in fuzzy multivariable control, a 4-input-1-output hierarchical fuzzy controller is proposed and its analytical expression is derived. According to the analytic expression of hierarchical fuzzy controller, a model reference self-learning method of its design parameters is proposed, in which the typical second-order system is selected by the reference model, the self-learning law is designed by gradient method, the adjustable fuzzy controller of online learning hierarchical fuzzy controller Parameters, in order to achieve the optimal value. In order to test the validity of the proposed control strategy, this method is applied to the trajectory tracking problem of cricket system, and the simulation experiment is carried out. Simulation results show that the hierarchical fuzzy controller has fewer adjustable parameters, faster learning convergence, better control effect and strong practicability.