Study On Sensor Failure Detection And Diagnoses Of Control System

Summarizes the achievements up to the minute in the sensor fault diagnoses field in recent years, and makes deeply study on the two basic diagnose methods. Realizes the sensor fault diagnoses of power plant main-steam system by using observer principle, and using MATLAB studied the neural network observer group by simulation. Presents a new sensor fault diagnoses method, the twice predictive diagnose method of control system sensor failure based on the radial basis functions networks, and makes an overall performance analysis based on a complex nonlinear system model. At last makes a particular discussion about the application issue on the sensor failure diagnoses method.FANG Fang (Control Theory and Control Engineering)Directed by Prof. SUN Wan-yun and Prof. MU Yu-guang.