Nonlinear System Fault Diagnosis And Fault-tolerant Control Based On Neural Network

The main work of this thesis is fault diagnosis for nonlinear systems based on neural network, at the same time, there is some introduction of fault-tolerant control based on fault diagnosis.First, the thesis introduces the technology of fault diagnosis, including the meaning, the methods, the contents and the task. Then, the development, the current and the foreground of fault diagnosis are summarized.After a simple introduction of neural network technology, radial-basis function neural network and the simulation of its arithmetic are researched.The present methods of nonlinear systems fault diagnosis are researched and summarized, and the disadvantages are pointed. Based on them, the way of this thesis is ascertained. Then, there are simulations of an example to test the way presented.At last, there is a simple introduction of active fault-tolerant control.The results of the simulations indicate the validity of the fault diagnosis arithmetic based on neural network observer.