Minimum-Learning-Parameters-Based Adaptive Neural Fault Tolerant Control With Its Application to Continuous Stirred Tank Reactor
针对执行器故障的多输入多输出系统，提出一种基于最小学习参数的分散神经网络输出反馈容错控制方法，并应用于连续搅拌釜式反应器，保证闭环系统信号半全局一致最终有界。
In this paper, a decentralized neural network (NN) output feedback fault tolerant control (FTC) problem is addressed for a class of multi-input multi-output systems with actuator fault. In order to avoid the noncausal problem, the original system is transformed into an input-output expression in accordance with the diffeomorphism theory. Then, in order to establish a quick response to the fault, the fault tolerant controller with minimum learning parameters has been designed such that the semiglobal uniform ultimate boundedness of all the variables in the resulting closed-loop systems can be guaranteed. Finally, the output feedback FTC approach is applied to the interconnected CSTRs, and the comparisons with existing methods are provided to show the effectiveness of the proposed strategy.