Fault Detection and Accommodation for Nonlinear Systems Using Fuzzy Neural Networks

A fault detection and accommodation method based on fuzzy neural networks was presented for nonlinear systems. The fault parameters was designed to detect the fault, adaptive updating method was introduced to estimate and tracking fault, fuzzy neural networks was used to adjust the fault parameters and construct automated fault diagnosis, and the fault compensation control force, which given by fault estimation, was used to realize fault accommodation. This framework leaded to a simple structure and an accurate detection. The simulation results in brushless DC motor showed that it was still able to work well with high dynamic performance and control precision under the condition of motor parameters´ variation fault and load torque disturbance