The Fault Diagnosis of Electromagnetic Valves Based on Driving Current Detection

The fault diagnosis of electromagnetic valves based on driving end current detection is proposed. The current characteristics of the faulted electromagnetic valves and the failure signal are analyzed. The four conditions of the valve are detected,including normal state,spring break state,spool seizure and un-resetting state. Variation trend is the character signal on terms of the direct-current characteristic. For the trait of the current signal,the wavelet packet decomposition is used to distill the corresponding frequent band energy as feature vector. The feature database is combined with the each frequent band energy which is produced after reconfiguration. The feedforward-back propagation network is used to identify the fault type of the electromagnetic valves. The result shows that the diagnosis method of energyfault can distinguish the different conditions of the electromagnetic valves,and the feedforward-back propagation network after training can identify the 3 fault conditions. The method is an effective assistant method for the maintenance of the electromagnetic valves,which can be widely used for the fault diagnosis of other electromagnetic valves.