Study On Theoretic Method And Its Experiment Of Subhealth State Assessment Of Hydraulic Cylinder

With the development of modern hydraulic system for the direction of high-speed,high-reliability and high-power, the function of electromechanical equipment hydraulicsystem is more and more powerful, the structure of equipment and the mechanism ofequipment fault is also more and more complex, the traditional maintenance method andconcept has been gradually incapable of ensuring the normal operation of equipment. Somany differently new maintenance policies have been developed. Condition basedmaintenance (CBM) is one of them. In the long-term use of hydraulic equipment, therewill be a state of performance degradation of equipment, which is called the sub-healthstate between a healthy state and complete failure state. Through assessment of sub-healthstate based on equipment, the equipment can be timely maintenanced and repairedaccording to its current state, that is a part of condition based maintenance too.Internal leakage in hydraulic cylinder is mainly caused by the seal wear failure. It isnot easy to be diagnosed. So this article is to assess the health states of the hydrauliccylinder through studying the internal leakage of hydraulic cylinder.First of all, this paper analyzes the failure mechanism of the hydraulic cylinderleakage. Through the leakage fault mathematical model analyze the effect of leakage onthe system pressure. Using the hydraulic cylinder internal leakage fault simulation modelin the hydraulic servo system to study the impact of leakage within the static and dynamiccharacteristics of the servo system.Secondly, using laboratory experiments that simulated the internal leakage of thehydraulic cylinder failure by adjusting the opening degree of the throttle valve, collectinga pressure signal of the rodless side chamber of the hydraulic cylinder, using the waveletpacket decomposition of pressure signal in time-frequency domain analysis, extract theinternal leakage fault characteristics, and using principal component analysis to reduce thedimension of the pressure signals in time domain indicators. Extracting core characteristicparameters.Finally, design the BP neural network status classification model, do health statusassessment from the angle of hydraulic cylinder leakage by using the fault characteristic parameters. And the test sample validation of the model is feasible. The sub-health stateassessment system of hydraulic cylinder is developed by using MATLAB GUI.The mainfunctions of the system is composed of analysis and feature extraction of the signal andstate assessment. the operation interface of the system is simple, and it is successfullyvisual assessment of the health status of the hydraulic cylinder.