Motor train unit wheel parameter abrasion trend analysis method

Abstract

The invention provides a motor train unit wheel parameter abrasion trend analysis method, which comprises the steps of integrating historical data, removing points with large measurement errors, and distinguishing turning repair points; calculating the goodness of fit of fitting methods of different orders according to the actual parameters, and selecting the fitting mode with the highest goodnessof fit from the goodness of fit; displaying the wheel parameter abrasion trend, the turning repair data and the LY detection data in a centralized mode, and using different colors for classificationand identification. According to the method, the degradation trend of the wheel can be displayed more accurately and visually through a series of wheel LY detection data, and each time node of wheel turning repair can be clearly identified based on a time scale display form. By means of the design, the problems that in the prior art, analysis on the abrasion trend of the wheels of the motor trainunit lacks multi-source historical data integration capacity, historical data value is not fully mined, and data abnormal point processing capacity is insufficient are solved; the wheel abrasion analysis precision of the motor train unit is effectively improved, and the safety and comfort of the motor train unit are guaranteed.

Description

technical field [0001] The invention belongs to the technical field of wheel wear analysis, and in particular relates to a method for analyzing the wear trend of wheel parameters of an EMU. Background technique [0002] In order to track and monitor the status information of a certain component in the EMU, and avoid its failure, failure, etc. affecting the safety and reliability of the system or causing huge economic losses, the parameter trend analysis method based on historical data fitting can be used to study the component performance during service, and to achieve a certain degree of early warning at the end of life. Although the characteristic parameter fitting based on historical detection data has been gradually applied to the trend analysis and early warning of the railway industry, the existing algorithm cannot be fully applied to the wheel parameter trend analysis, and its main technical defects are reflected in the following aspects: (1) Lack of multi-source his...

AI Technical Summary

Problems solved by technology

(2) The value of historical data has not been fully exploited: In recent years, with the upgrading and transformation of operation, testing and maintenance equipment, as well as the development of informatization and digitalization, many key information and data of mechanical parts have been stored electronically, and the accumulated Historical data can support the analysis of key performance of components during service

(3) Insufficient data processing ability for abnormal points: LY measurement data of wheel parameters usually have large errors, and the errors have a certain stage. Direct regression analysis on historical data will not only lose a certain amount of fitting accuracy, but also cause errors. Excessively large detection data is more likely to cause calculation results that do not match the actual situation, such as the wheel diameter increases with the running mileage

In addition, due to the lack of vehicle repair data, two repair cycles are classified into one fitting interval, which not only ignores the important influence of repair on the evolution of wheel parameters, but also ignores the relative relationship between the actual detection data of LY and the fitted degradation trend. The error may gradually increase

Images