Research On Intelligent Diagnosis Method Of Gearbox Fault

As the mechanical equipment becomes more and more complex and intelligent,the economic benefits it can produce are also increased.As the core component of the mechanical equipment,the gearbox determines the operating state of the mechanical equipment and plays a key role in the development of the machinery industry.Therefore,the research of gearbox fault diagnosis has important research value.At the same time,the vigorous development of artificial intelligence technology provides a new method for gearbox fault diagnosis.This thesis studies the fault diagnosis technology of gearbox,introduces artificial intelligence technology into gearbox fault diagnosis,and conducts research on gearbox fault diagnosis methods.The main research contents are as follows:(1)The comparative analysis of traditional fault diagnosis technology and intelligent fault diagnosis technology confirms the research significance of intelligent diagnosis method.The vibration mechanism of the gear is analyzed and the corresponding mechanical model is established.The vibration signal model of different states is simulated.The corresponding modulation information and the distribution characteristics of the side frequency band are analyzed.(2)A gearbox vibration signal acquisition system with FPGA as the core is designed to complete data sampling,conversion,storage and transmission.The gear meshing frequency and shaft rotation frequency during the operation of the gearbox are analyzed and experimentally verified.The results show that the acquisition system can sample and transmit 500 Hz vibration signals without distortion at a sampling frequency of 5 kHz,which can meet the requirements of the vibration test bench used in this article.The undistorted sampling and transmission of the gear meshing frequency and its sidebands verifies the feasibility of the signal acquisition system.(3)Through the wavelet packet decomposition of the vibration signal of the gearbox under different working conditions,the energy corresponding to different frequency bands in the wavelet packet energy spectrum is analyzed.Research has shown that the energy features contain a wealth of information about operating conditions,which can be used as sample data for intelligent diagnosis of gearbox faults.(4)Research on the intelligent diagnosis method of gearbox.The GA-BP fault diagnosis model is established on the basis of artificial neural network,and the S VM fault diagnosis model is established on the basis of support vector machine.Aiming at the problem that the optimal parameters of support vector machines are difficult to determine,particle swarm algorithm and genetic algorithm are introduced to establish PSO-S VM fault diagnosis model and GA-S VM fault diagnosis model respectively.Comprehensive evaluation of the diagnostic performance of the four models,the results show that the GA-SVM fault diagnosis model has a high fault identification accuracy rate,with an average accuracy rate of 97.68%,and the shortest running time,with an average running time of 20.02 s.