Analysis Of Failure Mechanism Of Electro-hydraulic Servo Valve And Research On Intelligent Failure Prediction Method

Electro-hydraulic servo valve is the most important part of electro-hydraulic servo control system.It is a hydraulic control valve that outputs the corresponding pressure and flow when the sensor receives a radio signal.However,because of its high precision,high integration,the fault type of electro-hydraulic servo valve is complex and changeable,and most of the fault types of electro-hydraulic servo valve can not be measured directly,which makes it difficult to predict the fault of electro-hydraulic servo valve.Therefore,it is very important to develop an effective intelligent prediction method of electro-hydraulic servo valve fault,and it will be very important to make accurate and rapid intelligent fault prediction of electro-hydraulic servo valve in operation.Compliance with the "Intelligent Manufacturing Engineering Implementation Guide(2016-2020)" clearly mentions that one of the key elements of the new intelligent manufacturing model is "remote operation and maintenance services",Its high-precision and highly integrated characteristics make the fault pattern of the electro-hydraulic servo valve more complicated.The fault types of most electro-hydraulic servo valves cannot be directly measured,which makes it difficult to predict the fault of electro-hydraulic servo valves.This paper mainly studies the fault prediction of moog-g761 double nozzle flapper electro-hydraulic servo valve,establishes failure prediction targets and evaluation indicators for common electro-hydraulic servo valve failures,and explores failure prediction strategies for electro-hydraulic servo valves.The research results to be obtained can be Lay a theoretical foundation for the promotion of electro-hydraulic servo valve failure prediction.First,establish the mathematical model of the electro-hydraulic servo valve to analyze the mechanism of servo valve failure.Analyze the structural composition,working mechanism and dynamic and static characteristics of the servo valve,and the mathematical model of the electro-hydraulic servo valve is established,and the characteristics of hydraulic oil temperature and pollution particles are analyzed,summarize the fault classification of the servo valve,at the same time,the failure principle of servo valve is revealed.Sensitivity analysis is carried out on the influence of structural parameters of electro-hydraulic servo valves with different fault types on flow rate,pressure and internal leakage,pressure,and internal leakage is carried out,and the reasons for the occurrence of different failures and the degree of influence on the performance of the servo valve are obtained.Then,it focuses on the analysis of the influence of the blockage of the orifice and the wear of the spool valve on the performance of the servo valve.Secondly,based on the above research and analysis,a network model combining Convolutional Neural Network(CNN)and Bi-directional Long-Short Term Memory neural network(Bi LSTM)is proposed.CNN is used to extract the high-dimensional spatial features of the sensor to realize the feature extraction of the early weak fault signals of the electro-hydraulic servo valve;at the same time,the Bi LSTM neural network is selected to process the feature sequence output by the CNN part to extract the electrical characteristics in the time dimension.Performance degradation characteristics of hydraulic servo valve,and realize fault classification and prediction of electro-hydraulic servo valve.Finally,the AMESIM software was used to simulate the servo valve core wear and orifice plate clogging fault data set.At the same time,the experimental data was used to accelerate the performance degradation of the servo valve.The fault prediction model is verified by comparing the simulation data with the experimental data.Prediction accuracy.At the same time,the four models of LSTM,CNN,Bi LSTM,and CNN+LSTM are compared for the accuracy of electro-hydraulic servo valve fault prediction and diagnosis.The CNN+Bi LSTM fault prediction model has faster training time,higher prediction accuracy,and better Adaptability.In order to solve the problems of complicated and changeable fault types,weak early faults,and difficult time series in the fault prediction of electro-hydraulic servo valves,an electro-hydraulic servo combining CNN and Bi LSTM was constructed.This model replaces the artificial feature selection and extraction,and solves the problem of time series of fault prediction.