A method for predicting the axle temperature by adopting layered multivariate stepwise regression analysis

Abstract

The invention discloses a method for predicting the shaft temperature by adopting layered multivariate stepwise regression analysis. According to the hierarchical multivariate stepwise regression analysis method, a shaft temperature prediction equation is established by combining real-time data, collected during operation of a high-speed train, of relevant factors of shaft temperature change, andmeanwhile three factors, namely the environment temperature, the operation speed and the air spring load, which greatly influence the train shaft temperature, are screened out. A prediction result andan actual measurement result are compared and analyzed, the result shows that a prediction model established through layered multivariate stepwise regression analysis has high precision, the change trend of the axle temperature can be accurately reflected, and certain help is provided for analysis and prediction of the axle temperature of a train, timely discovery of axle faults and guarantee ofsafe operation of the train.

Description

technical field [0001] The invention relates to the technical field of axles, in particular to a method for predicting axle temperature by adopting layered multivariate stepwise regression analysis. Background technique [0002] The axle is an important part of a high-speed train and a key part to ensure the safe operation of the train. The temperature of the axle is the most direct response to the operating condition of the axle. Once the axle is burned or cut, it is easy to cause a major train accident. From the fused sensor installed on the train in the early 1980s to monitor the axle temperature, to the installation of the wireless axle temperature alarm on the train now, the axle temperature detection system is constantly developing and improving. However, the axle temperature detection system still has its defects. The reasons for these system failures are as follows: First, the logical structure design of the shaft temperature alarm is not perfect, so that the shaft ...

AI Technical Summary

Problems solved by technology

However, the shaft temperature detection system still has its defects

The reasons for these system failures are as follows: First, the logical structure design of the shaft temperature alarm is not perfect, so that the shaft temperature information cannot be collected correctly and the error information cannot be filtered; secondly, the sensor is susceptible to interference from external factors (climate, etc.) , mistakenly make the shaft temperature alarm alarm when the shaft temperature is not abnormal or the shaft temperature is abnormal but has not yet alarmed; finally, the software part of the shaft temperature detection system also has areas that need to be improved

Images