Research On Fault Diagnosis Method Of Plunger Pump Of Switch Machine

The plunger pump is an important power component of the hydraulic system of the switch machine.The running state of the plunger pump affects the normal switching function of the switch machine.Due to switch machine plunger pump generally work in complex business environment,cause easily fails,the light will cause the vibration of the switch machine piston pump and leakage phenomenon,affect work efficiency or will have severe consequences for the railway safety,therefore counter rotating machine plunger pump fault diagnosis research has important practical significance.Various parameter signals of plunger pump of switch machine contain abundant information that can reflect the running state of plunger pump.Through analyzing these signals,accurate fault diagnosis of plunger pump of switch machine can be carried out.So this article to build switch machine plunger pump fault diagnosis system as the goal,through the rotating machine plunger pump fault diagnosis method of research,put forward based on vibration signal of the switch machine plunger pump fault diagnosis method based on multi-sensor information fusion and the switch machine plunger pump fault diagnosis methods,finally realizes the accurate diagnosis of rotating machine piston pump.On the basis of analyzing the structure and working principle of plunger pump of point machine,the failure form and reason of plunger pump of point machine are analyzed in detail.The fault diagnosis test platform for axial piston pump of point machine was built and the installation position and test method of each sensor were introduced respectively.The test data of piston pump of point machine under various working conditions were obtained,which provided data support for the study of fault diagnosis method.In order to reduce the influence of strong background noise on diagnosis accuracy,a feature extraction method combining Improved Complete Ensemble Empirical Mode Decomposition with adaptive Noise（ICEEMDAN）and optimized Improved Refined Composite Multiscale Dispersion Entropy was proposed to solve the problems of complex working mechanism and difficult feature extraction of vibration signal of plunger pump of switch machine.This method can carry out deep scale division of original signals,improve the identification accuracy of different signal components,effectively extract signal fault features,and effectively identify complex faults.The fault classification model of plunger pump of switch machine is studied,and The Arithmetic Optimization Algorithm is applied to the parameter selection of kernel extreme learning machine based on The principle analysis of kernel extreme learning machine.The model is applied to the fault diagnosis of plunger pump in point machine,and compared with the traditional ELM、PSO-KELM、 HHO-KELM algorithms,and six working conditions,such as normal state,plunger ball head wear,valve plate wear,piston ball head wear and valve plate wear,were identified and classified effectively.The experimental results show that AOA-KELM algorithm has the advantages of high diagnosis accuracy and short training time.The multi-sensor information fusion technology applied to the switch machine plunger pump fault diagnosis methods,make full use of the switch machine piston pump parameter information,more data layer acquisition switch machine plunger pump shell,oil pressure signal and vibration signal of acoustic pressure signal,motor current signal and speed signal,feature extraction and normalization processing respectively;AOA-KELM model was used for local subnet diagnosis in feature layer.The D-S evidence theory is used to fuse the diagnosis results of subnets at the feature layer,which greatly improves the accuracy of fault diagnosis.