考虑执行器饱和的改进无模型自适应控制

摘要

Model free adaptive control (MFAC) is a data-driven based control approach. The advantages of this method lie in low computational complexity, strong robustness and no-need of modeling during its design progress. However, actuator saturation is a problem which is not yet considered in all of the existing MFAC methods. In this paper, a novel improved MFAC method is proposed to deal with the constrains of actuator. Hildreth method is used to solve control output by introducing constraint condition for the critical function of control input, which simplifies the programming progress and reduces the computing load. After that, the stability of the closed-loop system is proved through rigorous analysis. At the end, taking Wood/Berry distillation as the plant, a series of comparative simulation is conducted and the result shows a better performance by using the proposed controller than traditional MFAC methods when actuator saturation exists.%无模型自适应控制(Model free adaptive control, MFAC)是一种数据驱动的控制方法,具有计算简单、鲁棒性强、无需建模等优点。目前无模型自适应控制方法普遍未考虑可能出现的执行器饱和问题。本文针对这一问题,对执行器执行能力存在上限的情况设计了改进算法。该算法通过对控制输入准则函数引入约束条件,使用Hildreth 方法进行数值求解,具有编程简单、计算量小的优点。在此基础上分析并证明了闭环稳定性。最后以蒸馏塔模型为控制对象,通过对比仿真实验验证了算法的有效性。