论文部分内容阅读
针对某些具有可能故障先验知识的非线性系统,提出了一种基于径向基函数(RBF)网络的逆系统多模型内模主动容错控制方法.首先采用RBF对系统正常及各种先验故障情形的逆系统建模,并由此建立动态系统逆模型库,然后基于逆系统方法将逆模型与系统串联形成伪线性系统,并对其设计了具有良好鲁棒性能的内模控制器.系统实际运行时,监控决策机制依据系统性能容忍度指标和模型失配度指标的实时计算分析,诊断系统所处运行模式,调用与之匹配的RBF逆模型,使其通过始终与原系统模型的串联保持为不变的伪线性系统,从而在无需改变内模控制器的情况下达到对非线性系统主动容错控制的目的.仿真实例验证了所提方法的有效性.
Aiming at some nonlinear systems with possible prior knowledge of fault, this paper proposes a kind of inverse system multi-model active fault tolerant control method based on Radial Basis Function (RBF) network.Firstly, RBF is used to normalize the system and various a priori Inverse system modeling of fault conditions and dynamic system inverse model library are established. Then inverse system is used to form pseudo-linear system in series with the inverse system, and an internal model controller with good robustness is designed. According to the real-time calculation and analysis of system performance tolerance index and model mismatch index, the decision-making mechanism of the system diagnoses the operating mode of the system and invokes the matching RBF inverse model to make it pass through with the original system model The pseudo-linear system which is kept in series is invariable, so as to achieve active fault-tolerant control of nonlinear system without changing the internal model controller.The simulation results verify the effectiveness of the proposed method.