Fault diagnosis method for wireless security monitoring system in chemistry and chemical engineering laboratory

For the safety problem of chemistry and chemical engineering laboratory, a kind of wireless security monitoring system was designed. It can transmit locale sensor parameters and compressed images via wireless communication module, and realize the function of laboratory safety remote wireless monitoring and surveillance. The wireless communication module in monitoring system consumes much energy and it prone to malfunction in long term running. The fault detection method for wireless communication module was also complicated. In order to detect faults of wireless communication module on line and identify their types as well as ensure the reliability of wireless safety monitoring system, the current characteristics of the wireless communication module was studied, and a fault diagnosis model based on fuzzy neural network was established. It can diagnose faults of the wireless communication module in different status. Experimental results show that the fault diagnosis method of wireless communication module base on fuzzy neural network has many advantages as compared to BP neural network, for instance short training time, fast convergence, small training error and validated error, high diagnostic accuracy, and a variety of faults types for wireless communication module can be detected on line. Therefore, the proposed diagnose method improves the reliability of wireless security monitoring system and has good practical value.