关键词: 故障诊断 , 容错控制 , BP 神经网络 , 粒子群优化算法 , 连续搅拌釜式反应器

Abstract: A fault-tolerant control method combining fault diagnosis and fault-tolerant control was proposed for sensor faults of a class of nonlinear system. A BP neural network based on particle swarm optimization algorithm was used to estimate system states and fault parameters of the constructed model for sensor faults. The estimated fault parameters were processed by the modified Bayes classification algorithm to achieve sensor faults detection, separation and estimation online, and fault-tolerant control was realized by compensation algorithm. Simulation results for continuous stirred tank reactor (CSTR) show good convergence of the approach and strong fault-tolerant ability for sensor faults.

Key words: Fault diagnosis, Fault-tolerant control, BP neural network, PSO, CSTR