基于二阶泛模型的无模型自适应控制及参数整定

Model free adaptive control and parameter tuning based on second order universal model

首先充分利用无模型自适应控制(MFAC)边建模、边控制的特点,推导基于二阶"泛模型"的改进无模型自适应控制方法,并推导伪偏导数及控制律的迭代公式,与基于一阶泛模型的MFAC方法相比,改进策略可以使每次迭代的泛模型更加准确,从而进一步提高控制精度。接着,针对改进MFAC的参数整定问题,提出基于优化技术的控制器参数整定方法,运用辨识出的近似模型针对不同的目标函数进行优化,使得其实用范围更加广泛。通过大量仿真实验对比可以看出:经过Jeu-tr型性能指标进行参数优化的改进MFAC控制器动态响应最好,且优化迭代次数较少。因此,控制效果得到显著改善。

An improved model free adaptive control（MFAC） based on second-order universal model was derived,which can greatly improve the model and control precision.The control law and pseudo-partial-derivative were iteratively derived.For the parameter tuning of improved MFAC,a parameter optimization algorithm was presented.Using the approximate identified model,the optimal controller parameters were obtained for several different objective functions,which had wide scope of application.The Jeu-tr performance index makes the system possess better dynamic response,and less iteration times.The simulation results show the effectiveness of improved MFAC control strategy and parameter tuning method.