Method for predicting residual life of axle box

The invention discloses a method for predicting the residual life of an axle box, which comprises the steps of first, collecting an original vibration signal of a test vehicle axle box with the running speed of 70km/h at an ambient temperature of 20 DEG C; second, building an artificial intelligence model based on the least squares support vector machine; third, preprocessing the original vibration signal to obtain a correlation coefficient table; and fourth, inputting the correlation coefficient table into the artificial intelligence model to obtain the residual life of the axle box. According to the method, the artificial intelligence model based on the least squares support vector machine is utilized, the correlation coefficient table is enabled to serve as the input of the artificial intelligence model, and the residual life of the axle box is obtained. The method is simple, and the obtained residual life is high in precision. The method can be applied to the field of residual lifeprediction for functiona