Research On Spatial Feature Recognition Of Rail Surface Based On Response Of Vehicle System

All long,the two adjacent rails are connecting by fish plate joints.With the constant improvement of technological level,it will be an inevitable development tendency of Continuous Welded Rails(CWR)to replace the old railway.The safety of the railway plays an important part in transportation system.Railway surface detection can reduce the traffic accidents greatly.It’s a challenge to detect rail-steel’s surface defects in a vision system,because they appear randomly,have different sizes,distribute discretely.To more specifically identify the spatial feature of rail surface,the priority research object of this text will be the welded joint of the rail surface feature.The key technology of seamless line is rail welding,the joint quality is directly related to the running stability,passenger comfort and driving safety.So it can guide the safety of train operation,road monitoring and fault diagnosis by understanding the vibration state of these parts.Firstly,this paper uses harmonic model to describe the rail welding joint irregularity,and measure the actual operation line of welded joint data structure;uses the theoretical model and the experimental data to establish the joint irregularity samples;superposes the weld joint irregularity and the 6 level random irregularity samples,giving track random irregularity samples which have rail joint,providing incentives for the research on the vibration data of track irregularity under the action of other parts of the axle box.Then according to the relevant dynamics theories to establish the vehicle track vertical coupling dynamic model,using the explicit integral method combined with MATLAB software to solve the equation,obtaining the vibration data of the car body,bogie and wheel joint before and after the rail welding,and verifying the dynamic performance of the model;similarly,using the Universal Mechanism simulation software to establish the rail track dynamics model,input the vertical and horizontal irregularity samples of left and right rail track which are the same as the MATLAB model,and output axle vibration data;through the correctness of the vibration acceleration model under different speeds;finally using adjustable second order blind identification algorithm based on weight,isolates the characteristics of vibration signal before and after the malfunction from the vibration output of the coupling system,then finds the spatial position of the rail joint added in the track irregularity,and is coincide with the malfunction of the rail joint,which verifies the accuracy of the algorithm and model.