Research On Fault Diagnosis Method Of Rocker Drive System Of Shearer Based On Data Drive

The coal mining machine is one of the key equipments for the complete coal mining system.High power,high reliability,high intelligence and easy maintenance are the trend of shearer research at this stage.Rocker arm,as an important part of shearer to complete cutting task,will directly affect the working efficiency of shearer in case of failure.Its reliability is very important to the safe and efficient production of coal mine.Therefore,the research on the fault diagnosis method of rocker arm has become the most important task in the maintenance of shearer.In view of the above problems,this paper takes the shearer rocker arm transmission system as the research object,on the basis of analyzing the actual working condition and the fault mode,studies the extraction method of fault characteristic vector under different time scales.An intelligent diagnosis model based on improved depth confidence network is established.First of all,the structure of the shearer rocker arm transmission system is analyzed,and combined with the actual working condition environment,the various failure forms of the shearer rocker arm are analyzed,and the classification of the fault form,the division of the test area and the location of the measuring point are studied.Finally,the vibration characteristics of each,component are analyzed,which lays a foundation for feature extraction and intelligent diagnosis.Secondly,aiming at the non-linear and non-stationary vibration signal of rocker arm transmission system during coal cutting,the multi-threshold wavelet packet de-noising method is studied,and different threshold methods are chosen to de-noising different frequency signals.Furthermore,the local optimality of multi-threshold wavelet packet de-noising is elim,inated by EMD decomposition,and several IMF signal components are obtained.Combined with the entropy value of Shanan information,the characteristic checking of multiple IMF components is carried out,and the simulation experiment of analog signals is carried out.Then,the fault diagnosis method of rocker arm transmission system based on supervised learning network is proposed,the training process of supervised learning model is analyzed,and the learning rate and iterative times of the model are studied.Through the DDS fault diagnosis platform to verify the feasibility of the diagnosis scheme;Aiming at the problem of low efficiency of model diagnosis,a fault diagnosis model of shearer rocker arm transmission system based on PSO-BPDBN is put forward and tested.Finally,combining the methods of feature extraction and pattern recognition,the ground test experiment of shearer rocker arm is carried out,the feature vector is extracted from the collected signal data,and the corresponding feature extraction sample set is constructed.The high-speed and low-speed regions of the transmission system are tested to explore the efficiency of fault diagnosis under different diagnostic schemes and to verify the validity of the model proposed in this paper.