Research On Fault Diagnosis Of Mine Ventilator Bearing Based On Date Drive

Ventilator is one of the most important equipment in the production of coal mine enterprises,the bearing is a key component to maintain steady operation of the ventilator,which plays a role in driving and reducing friction,the research on monitoring status and fault diagnosis of the bearing has a great important significance.In this paper,the ventilator bearing is taken as the research object,according to the non-linearity and non-stationary characteristics of bearing vibration signal,a fault diagnosis method based on LCD and IPSO optimized probabilistic neural networks is studied.In this paper,firstly,the wavelet packet improved threshold noise reduction method is used to preprocess the bearing fault signals.Then,the LCD method is used to extract the time-frequency characteristic value of ventilator bearing vibration signal,compared with EMD method,the LCD method has the optimal decomposition accuracy and the minimum reconstruction error,which is more suitable for feature extraction of ventilator bearing vibration signal.Based on the acquired signal fault characteristic parameters,a probabilistic neural network(PNN)model for identifying ventilator bearing faults is established.Aiming at the randomness of the parameter values of smoothing factor in the PNN network,an improved particle swarm optimization(IPSO)is used to optimize smoothing factor,the optimal smoothing factor value obtained by the improved particle swarm optimization algorithm is 0.25,which is applied to the probabilistic neural network model for training and testing.At the same time,the simulation experiment is carried out in MATLAB environment,the simulation results showed that the accuracy of bearing fault diagnosis of the IPSO-PNN network model is 94%,which is 17.5%higher than probability neural network and 2.17%higher than PSO-PNN network;the diagnosis time is 76.19%lower than that of probabilistic neural network and 55.72%lower than that of PSO-PNN network.Therefore,the diagnosis method of IPSO-PNN network not only improves the accuracy of ventilator bearing fault diagnosis,but also ensures the diagnosis rate,which has good engineering practical value.