Fault Diagnosis of Coal Mine Fan Based on PSO-BP Neural Network

Aiming at the problem that it is difficult to detect the early safety hazard of coal mine ventilator in time, a mechanical fault diagnosis method of ventilator based on vibration and noise is proposed. Firstly, the fan fault test bench is built, and four kinds of fan faults are simulated. Secondly, the collected fault noise is decomposed into multiple eigenfunctions by EEMD, and the eigenindexes of each layer of eigenfunctions are analyzed to obtain the fault eigenvector table. Finally, the faults are identified and predicted by PSO-BP neural network. The experimental results show that the fault characteristics of fan can be extracted effectively by using EEMD method. The kurtosis index, spectrum maximum value and reciprocal power of normalized signal are input into PSO-BP neural network for fault identification, and the fault recognition rate is high.