Fault Diagnosis System Of Shearer's Key Parts Based On Data Driving

As an important equipment for coal mining,the shearer has a huge size and complex structure.The safety and stability of its equipment have a great impact on the efficiency of coal mining.Due to factors such as the harsh working environment of the shearer and the long-term exposure to electromagnetic interference,the shearer's failures often occur.Once the shearer fails,it will inevitably affect production efficiency and even cause serious casualties.Therefore,real-time fault diagnosis and analysis of the shearer is of great significance.The coal mine has installed software with monitoring and diagnosis functions to ensure the safe production of the shearer,but the software uses a relatively simple diagnosis method,usually only able to diagnose obvious faults and the results are partial,failing to accurately determine the specific fault location and diagnosis effect Poor,low accuracy,lagging in intelligence.In addition,the maintenance personnel of the shearer still use the traditional diagnosis method,which takes a long time and the accuracy of diagnosis is not high,usually requires the assistance of professional equipment,and has higher requirements for the diagnosis personnel.With the rapid development of big data and artificial intelligence,neural network technology has made a lot of research results in the field of fault diagnosis,but it is still less applied in coal mining.In response to the above problems,this paper takes the key parts of the shearer as the research object,studies the parameter diagnosis method based on traditional diagnosis methods and the deep residual network fault diagnosis method based on neural network technology,and analyzes the above two methods in the key to the shearer The feasibility of component fault diagnosis application.On the basis of the research of the diagnosis method,a fault diagnosis system of the coal mining machine is built,and the reliability and stability of the system are verified.Combining the common faults of the shearer and the traditional diagnosis method,a parameter diagnosis method for the online real-time diagnosis of the shearer is proposed.According to the requirements of the shearer operation manual and actual production,the alarm threshold is set for the shearer fault,and the fault diagnosis is realized through the parameter diagnosis process.The coal mining machine data collected at the coal mine production site is selected as the data source,and two computers are used to simulate mine data transmission and field fault diagnosis for verification experiments.The experimental result is that the comprehensive accuracy rate of the parameter diagnosis method is 98.9%,which shows that the method is effective for coal mining.The machine has a high fault recognition rate and is practical.In addition,the data reading methods with different time intervals are set for comparative experiments,and the results show that the time interval between the two pieces of data is 5 seconds when the parameter is diagnosed.Based on the deep residual network theory,a fault diagnosis method for the shearer's deep residual network(Res Net)is proposed.First,a deep residual network fault diagnosis model is constructed,and then the structure of the model is optimized through the dropout strategy,BN layer and appropriate activation function.This method is mainly to diagnose the gears and bearings in the rocker gearbox of the coal shearer.The diagnosis results are more detailed and the fault type and location are accurately determined.Using the vibration data of the rocker gearbox to verify the fault diagnosis model of the deep residual network,the experimental result is that the accuracy of the fault diagnosis can reach 99.6%,indicating that the model has a high fault recognition rate.Using confusion matrix evaluation method to analyze the classification performance of the model,the result is that the average classification accuracy of the model can be 99.6%,which proves that the model has high classification accuracy and classification stability.The fault diagnosis method of the deep residual network and the intelligent fusion of the parameter diagnosis method are used to jointly complete the faults of the internal gears and bearings of the rocker arm gearbox of the coal shearer,which is not only real-time,but also enriches the diagnosis results.Using the Visual Studio 2010 development platform and SQL database to design and build the shearer fault system,the shearer's monitoring,fault diagnosis and data query functions are realized.The system realizes the application of parameter diagnosis method with logic code,and realizes the application of its fault diagnosis method by calling and executing the method of deep residual network fault diagnosis model.Finally,the system test verifies that the system runs stably and the diagnosis result is reliable.