Embedded Rolling Bearing Intelligent Measurement And Control System Based On LABVIEW And MATLAB

With the change of the times,China's urban rail transit industry is also developing rapidly.At the same time,the safe operation of rail transit has attracted more attention among people.Rolling bearing is one of the most basic and widely used components of rotating machinery and industrial equipment.How to monitor and diagnose its running state accurately and effectively is significant to ensure the safe and steady operation of the machines and avoid major emergencies.The fault of rolling bearing is usually random,through regular manual inspection,it is difficult to avoid the occurrence of fault.Therefore,it is very important to monitor its operation status online and diagnose its fault types in time.Rolling bearing is usually in relatively bad working conditions and environment,which is easy to be interfered by noise.At the same time,the vibration signal is not stable enough.Therefore,it is necessary to adopt appropriate signal feature extraction methods and efficient fault diagnosis technologies to ensure its safe and stable operation.Based on virtual instrument technologies and interactive algorithms,this paper designs and develops an embedded intelligent monitoring and fault diagnosis system of rolling bearing based on LABVIEW and MATLAB.The system uses the c RIO9075 chassis provided by NI company and its voltage module NI9215 as hardware,and completes the acquisition of vibration signals,data storage,graphical display of signals,and the sending and receiving of commands by using LABVIEW to achieve software programming and hardware driving calls.Through the MATLAB Script in the upper computer LABVIEW.It can effectively extract the fault features from the signals,and then use a variety of intelligent algorithms to diagnose the possible and common fault types as well as normal states.In order to further improve the diagnosis performance and accuracy of rolling bearing,this paper proposes a comprehensive and intelligent fault diagnosis technique based on the common fault diagnosis algorithms.Firstly,the complementary ensemble empirical mode decomposition(CEEMD)and energy moment normalization are used to extract the energy feature vectors which reflect different fault states in vibration signals;secondly,the extracted fault feature vectors are input into the back propagation neural network(BPNN)and the probabilistic neural network(PNN)model optimized by the improved particle swarm optimization(PSO)for fault classification.And the classification results are returned back to the monitoring interface to achieve accurate fault diagnosis.The simulation results show that the interactive interface of the system is easy to operate and it displays clearly.It can also collect the bearing vibration signals in real time and effectively.Compared with the standard neural network,the diagnosis rate of the model is higher,and the diagnosis performance is very good.