The Analysis To The Chassis Of Heavy-duty Vehicle Based On Vibration Signal

The automobile engine,components and system assembly are all installed on the chassis.The chassis carries the weight of the whole vehicle and realizes the acceptance and transmission of the power of the engine.When a chassis failure occurs,the vehicle will not work properly or even pose a threat to life safety.The fault diagnosis of chassis can effectively guarantee the completion of various vehicles.In this paper,the fault diagnosis and state monitoring of vehicle chassis are studied by means of empirical mode decomposition,set empirical mode decomposition,inherent time scale decomposition and multidimensional intrinsic time scale decomposition.This paper selects the bearing bearing fault data from the Bearing Data Center of the Western Reserve University as the research object.The vibration signals of the bearing under normal condition,inner ring failure,outer ring failure and rolling element failure,and the gear in normal state and crack failure The vibration signals under faults of broken teeth and broken teeth are analyzed and denoised by means of empirical mode decomposition,set empirical mode decomposition,and intrinsic time-scale decomposition;and the fault features are extracted by combining frequency domain analysis,energy entropy,and Teager energy spectrum analysis methods.Based on eigenvectors,three kinds of kernel function support vector machines are built,support vector machines are trained and failure mode recognition of rotating machinery is completed.For heavy vehicles,there are complex mechanical movements in the horizontal and vertical directions,and there is a coupling relationship in the vibration signals.The situation,explored the multidimensional empirical mode decomposition and proposed a multi-dimensional intrinsic time-scale decomposition,using this algorithm to decompose the multi-channel bearing vibration signal and accurately extract the fault feature frequency.In this paper,we find that the feature information in non-stationary and nonlinear signals can be extracted accurately according to the method of vibration signal analysis used in this paper,and the pattern recognition of the working state can be completed by this feature.