一种基于VRFT的全格式无模型自适应控制

针对无模型自适应控制中的时变参数拟梯度向量估计值信息没得到充分利用的问题,提出一种基于VRFT（虚拟参考反馈调整方法）的无模型自适应控制方法.该方法利用VRFT中的参考模型思想,首先将非线性系统经全格式线性化后的线性模型与无模型自适应控制器构成的闭环传递函数作为参考模型,再以参考模型的输出与系统期望输出的误差作为控制器的输入,从而将拟梯度向量在过去时刻的估计值引入到无模型自适应控制设计中,提高了伪偏导数估计值信息利用率.同时消去控制律中的权重因子,避免因人为赋值而影响控制效果.

A Full-format Model Free Adaptive Control Based on VRFT

Considering the fact that in model free adaptive control（MFAC）,the estimated value mtormation in the proposed time-varying parameter＇ s gradient vector can＇ t be fully used,a model free adaptive control method based on virtual reference feedback tuning（VRFT） was proposed.According to the theory of VRFT,both fully-formatted linear model of the nonlinear system and the closed-loop transfer function constituted by model free adaptive control were taken as the reference model;and the error between reference model output and system＇ s desired output was taken as the control input,in this way,the estimated value of gradient vector in the past time was introduced to the model free adaptive control to have utilization rate of the estimated value of pseudo partial derivative improved,meanwhile,the weighting factor of the control law can be eliminated so that the influence on control effect from artificial assignment can be avoided.