Research On Remote Monitoring And Fault Diagnosis Technology Of High Torque Wheel Hub Drive System

Due to the characteristics of strong transportation capacity and good economical applicability of mining dump trucks,the mining dump trucks are the main transportation force in mine transportation,and the proportion of their use scale is increasing year by year.The high-torque wheel hub is a key component of the electric wheel mining dump truck.Whether it fails or not determines the safe operation of the mining dump truck.Therefore,it is very necessary to monitor and diagnose the high-torque wheel hub.However,at present,there is still not perfect of the monitoring and diagnosis system to ensure the high-torque wheel hub’s safety.Most of the user’s diagnosis of high-torque wheel hubs is carried out by the naked eye and experience,which is seriously affected by people’s subjective consciousness,and cannot achieve real-time monitoring and diagnosis,which has a very serious safety hazard.Aiming at the imperfect monitoring and fault diagnosis system for high-torque wheel hub,a web-based high-torque wheel hub monitoring and fault diagnosis system is developed to realize real-time monitoring and fault diagnosis of high-torque wheel hub.The main contents are as follows:(1)The signal processing method of the high-torque wheel hub drive system is studied and the characteristic frequency analysis is carried out.According to the structure and composition of the high-torque hub drive system,the operation principle,the transmission ratio of the planetary gear and the structure of the bearing,the meshing frequency of each gear and the failure frequency of each component of the bearing are calculated.In order to completing the analysis of the relevant knowledge of the high-torque wheel hub drive system,the six signal processing methods including fourier transform,wavelet transform,cepstrum,envelope spectrum,empirical mode decomposition and variational mode decomposition are used to analyze eigenfrequency analysis of vibration signal of the high-torque wheel hub drive system,compared to the calculated meshing frequency of the gear and the fault frequency of the bearing.(2)Multi-scale convolutional neural network(MS-CNN)fault diagnosis method based on convolutional neural network is proposed.Aiming at the situation that the fault recognition accuracy of the general convolutional neural network is not high under the interference of strong noise,multi-scale convolutional neural network is proposed.The algorithm is an end-to-end fault diagnosis algorithm,which can not only directly extract features from complex signal types without any operation,but also extract features of fault signals with convolution kernels in different parameters,which can ensure the robustness and representativeness of the extracted features under non-stationary conditions,and the effectiveness of the algorithm is verified by relevant data.(3)The overall structure of the remote monitoring and fault diagnosis system for the high-torque wheel hub is designed.According to the relevant requirements and the functional requirements of the system,the functions of the system are determined,the technical selection of the front and back ends and the solution for real-time monitoring are completed,and the main contents of the monitoring and diagnosis of the core functions of the system are established.(4)Web-based high-torque wheel hub remote monitoring and fault diagnosis system software is developed.System related functions is analyzed,system data structure is clarified,and relevant database table structure for the information is designed that the system needs to store.The system software has achieved the established functions,and has undergone a stress test and nuclear compatibility test.The test results show that it meets the design requirements.