The Search Of Intelligence Quantitative Detection Of Railway Vehicle Wheel Tread Irregularities

There are tWo kinds of ways to measure the railway vehicle wheel treadirreguIarities. One is the static measurement that uses the mechanical ruler,electro-testing aPparatus and device equipped with CCD. Another is the dgnamicmeasurement that mainly depends on the technology of ultrasonic and CCD. In thisthesis, We use an original method called the Collision Vibration EnergyTransmission Time-Difference Analysis System for the automatic detection ofrailway vehicle wheel tread irregularities, which provides fOr both the qualitativeanalysis and the quantitative detection. In practical application, the method worksmuch better than existing methods in the world.On the basis of the system function, this thesis presents the design schema ofthe system model. In the design of the system constituting structure, this system hasmore anti-vibrating, anti-snow, anti-rain, anti-wind, anti-sand and durablecharactistics than other systems due to we adopt the sensor of electric eddy current.The introduction of the positioning sensor makes our software easier. In the designof the system model, the original method ca1led the Collision Vibration EnergyTransmission Time-Difference Analysis System fOr the automatic detection ofrailway vehicle wheel tread irregularities can position the tread irregularities wheelprecisely.Then, we begin the work of the software design. On the basis of comparing tWomethods of data samp1ing, we adopt the first samPling last processing schema. Afterthat job we begin the work of data processing module design. This module includesthe disceming of the vibrating wave, the positioning of the tread irregularities wheeland the computing of the tread irregularities dePth. Because the main program waswritten by VB and the data processing module was written by C, so we packaged thedata processing module into DLL to make the main program can use it just like afunction.m aWe tested the system in the lab. The results prove the success of the design. In the following spot testing, the percentage of the wheel treads irregularities detection precision is high. Although the tread irregularities depth computing result error is a little big due to the sampling frequency is too low. We believe that the tread irregularities depth computing result will improve with enhancing the frequency of the sampling.