A Prediction Method Based on Stepwise Regression Analysis for Train Axle Temperature

The axle temperature of the high speed train is the most direct reflection of the train operating conditions while it is also affected by many factors. The factors which significantly affect the axle temperature are screened out by using the stepwise regression analysis and the prediction equation of the axle temperature is established, so as to compare the predicted data and the measured data. The validity of the coefficients in the equation is verified through R-squared, F test and T test. The experiment shows that R-squared is between 0.81 and 0.93, indicating a high degree of fitting prediction equations, and F test results show that the overall equation is significantly better. The results of T test indicate that the velocity, the carrying capability and the ambient temperature have significant influence on the change of axle temperature. But the traction and the power of traction have less effect. The result shows that this method can reflects the variation trend of axle temperature, which can provide support to the operation and maintenance of axle.