Research On Dynamic Detection Method And Technology Of Out-of-round Railway Wheel

With the continuous increase of the operating speed,load capacity and transportation density of railways and urban rail transit,the interaction between wheel and rail is becoming more frequent,and the force is also increasing,which is easy to cause wheel failure.Out-of-round wheel is a common fault of railway and urban rail transit wheels,which is define as the rolling circle of wheel tread deviates from the ideal circle.Out-ofround treads will bring additional periodic impact to the wheelsets and rails,which is one of the main causes of cracks in the rails and wheelsets.Serious out-of-round wheel will cause major accidents such as train derailment,causing heavy casualties and property losses.As a result,research on the method,technology and device for on-line dynamic detection of out-of-round wheel can not only reduce the maintenance cost of rolling stock,but also avoid the occurrence of train accidents,which has become the research focus in the field of railway and urban rail transit.Although there are many dynamic detection methods for out-of-round wheel,these methods generally have some shortcomings,such as the detection parameters are not comprehensive and the out-of-round features is difficult to extract and quantify.In this thesis,the research on the high-precision dynamic detection technology for out-of-round wheel was carried out based on the comprehensive analysis of domestic and foreign related research,a dynamic model of out-of-round wheel was established,the dynamic quantitative detection method for wheel flat and out-ofroundness was proposed,the related measuring instruments were developed and the effectiveness of the system was verified through a series of experiments.The main work of this thesis include:(1)The dynamic model of out-of-round wheel is established,which can simulate and analyze the dynamic response of out-of-round wheel to rail.First,the dynamic model of out-of-round wheel is established based on the finite element method and the multibody dynamics method.Then,different wheel flats and wheel polygons are used to simulate and analyze the influence of different speeds,load capacities,flat lengths,flat positions,and wheel polygon orders on rail deformation.Finally,the dynamic response of the out-of-round wheel to the rail is studied,which provides a theoretical basis for the construction of out-of-round wheel detection system.(2)The wheel flat detection system based on the parallelogram mechanism is studied,and the simulation models of the wheel,rail and parallelogram mechanism are established.By optimally selecting the design parameters of the mechanism,the adaptive speed of the parallelogram mechanism for dynamic measurement of wheel flat is improved.Based on the rigid-flexible coupled multibody dynamics theory,a three-dimensional simulation model of wheel,rail and parallelogram is established.By optimally selecting the damping coefficients of the hydraulic damper,the measuring positions,and the downward displacements;the velocity threshold of the measuring mechanism under the impact vibration of the wheel is improved.After parameter optimal selection,the speed impact threshold of the measuring mechanism was increased from 14 km/h to 40 km/h,which meets the speed requirements of low-speed train entering garage.The correctness of the model simulation through field experiments is verified by installing the parallelogram mechanism with optimal parameters on the railway site.(3)A statistical detection method of out-of-roundness based on laser displacement sensor is proposed,which can realize high-precision and dynamic quantitative detection of out-of-roundness of trains.Three laser displacement sensors are used to measure the diameter of the wheel,and the out-of-roundness is calculated by counting the repeated passing of the vehicle in a period of time.In order to verify the effectiveness of the system,a series of experiments were carried out.The experimental results show that the wheel diameter repeatability standard deviation is ±0.09 mm,and the contrast deviation of outof-roundness is 0.37 mm.(4)A high-precision dynamic detection method for out-of-round wheel based on laser collimation method is proposed,and the detection system is developed,which has the advantages of non-contact detection and adapting to high-speed train operation.The out-of-round wheel detection system has been developed based on the optical design,mechanical structure and software design.In order to realize the quantitative detection of wheel flat,a quantitative calculation model for wheel flats is established and two different recognition algorithms based on time-frequency multi-parameter fusion and PSO-SVM algorithm and based on the Empirical Model Decomposition and PSO-SVM algorithm are adopt respectively.The effectiveness of the system is verified by laboratory experiments,and the recognition accuracy of flat signal is 90%.The correctness of the model simulation results is verified through field experiments.Through the above research,various detection methods for out-of-round wheel treads have been obtained,and the corresponding detection system has been developed,which provides a technical way to ensure the safty of train operation.