Research On Fault Diagnosis And Operation-maintenance Of Key Components For Gearbox

In recent years,rail transit has become the lifeblood of China’s major cities transport system,and sustained high-speed development.Under the increasing operating pressure,the ineffectiveness of the current maintenance technology method is difficult to meet the needs of a large number of rail vehicles,and become a prominent contradiction of train safety and security.Gearbox as an indispensable part of the vehicle powertrain system,its operating status directly affects the safety performance of the vehicle.Therefore,real-time monitoring means timely and accurate access to the operation of the gear box,and accurately analyze the type of failure,so targeted and predictive to develop maintenance plans and management decisions,to ensure the safe and reliable operation of rail vehicles to prevent accidents Is of great importance.In this paper,the gearbox fault diagnosis and train maintenance program optimization systematically studied,the main contents are as follows:(1)In order to effectively characterize the difficulty of complex system fault,this paper adopts multi-dimensional feature extraction method.Firstly,the signals are decomposed by wavelet packet decomposition,empirical mode decomposition and local mean decomposition.Then,direct time domain characteristics,frequency domain statistical characteristics and time-frequency energy characteristics are extracted.The characterization of gears by different fault types is selected.And then use this indicator set to extract the characteristics of better response failure,through the principal component analysis method for feature fusion and dimensionality reduction,with the main element as the final characteristic index.(2)Based on the high degree of ambiguity between the different fault states of train gearboxes,a state identification method based on interval fuzzy C-means clustering is proposed and compared with the traditional fuzzy C-means clustering method.,And their performance differences in classification effect and iterative efficiency algorithm are analyzed and compared.(3)Based on the research results of gearbox fault diagnosis,this paper further studies the concrete scheme of integration of operation and maintenance plan.Based on the shortcomings of the current operation and maintenance and the expectation of the operator,a fuzzy multi-objective maintenance model with maximum availability,minimum cost and reliability is constructed.Based on the existing technical conditions,on the basis of on-line monitoring,a dynamic adjustment monitoring period method based on the threshold of failure rate is proposed.Finally,two kinds of gear data are used to verify the case,and the best maintenance time and maintenance strategy based on the fault are given.A series of process of active maintenance of the train gearbox is realized.