Fault diagnosis for boiler drum water level sensor based on RBF network
SU Hong, TIAN Pei, CHANG Xiao-quan, SU Xi-feng (School of Control Science and Engineering, North China Electric Power University, Baoding 071003, China; Beijing University of Aeronautics and Astronautics, Beijing 100083, China)
Based on constructing RBF network estimator, a novel fault diagnosis method is suggested. The RBF network is trained by the real- time data of all kinds of correlative parameters. It is used to estimate the water drum level with high accuracy. The deviation between the output of the RBF (Radial Basis Function) network estimator and the actual value of the sensor is compared with the predetermined threshold. It is to detect the failures of boiler water level sensor. When a failure appears, the reconfiguration of monitor or control system of boiler water level can be achieved by using a switch. The fault diagnosis system can eliminate the fault effect. Simulation results show that the fault diagnosis monitor can achieve state monitor and diagnosis of boiler level sensors in time.