Research On Method Of Engineering Machinery Hydraulic Cylinder's Fault Feature Extraction Based On Wavelet Analysis

Phydraulic system plays a central role in the field of construction machinery. It will cause enormous economic losses in case of its failure. Hydraulic cylinder is the actuator of the hydraulic system and occupies an important position, its failure will directly affect the normal work and life of the equipment. Leakage is the common failure of the hydraulic cylinder and the serious leakage can not only cause environmental pollution, but also lead to the pressure decrease of hydraulic cylinder chamber, resulting in the failure of hydraulic cylinder. This paper analyzes the Hydraulic cylinder oil inlet's pressure signal and the piston rod's displacement signal to extract the failure feature vectors of the hydraulic cylinder leakage failure, as a basis for condition monitoring and failure diagnosis of hydraulic cylinders.The object of this thesis is the leakage failure of typical components of the hydraulic cylinder in a complete hydraulic system. As a prerequisite, the wavelet transform as the theoretical basis, wavelet analysis the advantages of non-stationary signals applied to the hydraulic cylinder leakage fault feature extraction. With the parameter measurement method as a prerequisite, and the wavelet transformation as theoretical basis, we apply the advantages of the wavelet analysis on non-stationary signals to extract the hydraulic cylinder leakage failure feature. Leakage includes external leakage and internal leakage. External leakage is obvious, so this paper is focused on internal leakage.The main achievements include:(1)This paper describes the failure mechanism and causes of hydraulic system leakage. By analyzing its mathematical model to explain the rate of increase or decrease of the cylinder pressure will be affected by leakage, laying a theoretical foundation for the data processing in the subsequent chapters.(2) Analysis of the progress of the fault diagnosis method based on signal processing, with emphasis on the wavelet transform theory,this paper proposes the failure diagnosis method applying for the hydraulic system. It focuses on the wavelet transform time-frequency analysis and multi-resolution analysis capabilities, which is the basis of this study. (3)Using wavelet packet decomposition to analyze the pressure signal and the displacement signal, to extract the rate of pressure increase of the pressure signal (PY1), the pressure signal (62.5-125Hz) wavelet packet energy value (PY2), the pressure signal wavelet packet energy entropy (PY3), the pressure signal wavelet packet energy variance (PY4), and the displacement signal of the piston rod stretching speed (S1).(4)In order to analyze the sensitivity of each fault vector to the leakage,we normalized the obtained data so that analyze the variance of the eigenvalue distribution in the case of leakage. The sequence of the fault feature sensitivity is: PY4> PY1PY2> S1> PY3.The result will guide the practical application of the theoretical analysis results.(5)With Visual Basic and Matlab mixed programming ideas,we build a "Construction Machinery Hydraulic Cylinder's Fault Diagnosis System"(CMHCFDS)to achieve visualization diagnosis of construction machinery hydraulic cylinder internal leakage fault.