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.

中文翻译：

---

基于最小学习参数的自适应神经容错控制及其在连续搅拌釜反应器中的应用

在本文中，针对一类具有执行器故障的多输入多输出系统解决了分散式神经网络（NN）输出反馈容错控制（FTC）问题。为了避免非因果问题，根据微分同胚理论将原系统转化为输入输出表达式。然后，为了建立对故障的快速响应，设计了具有最小学习参数的容错控制器，以保证所得闭环系统中所有变量的半全局统一极限有界。最后，将输出反馈 FTC 方法应用于互连的 CSTR，并提供与现有方法的比较以显示所提出策略的有效性。