Online identification method of electric locomotive idling based on fuzzy entropy and kernel extreme learning machine

Abstract

The invention discloses an electric locomotive idling online recognition method based on fuzzy entropy and kernel extreme learning machine, which includes a multi-scale fuzzy entropy feature extraction module, an optimal kernel extreme learning machine model module and an idling online recognition module. Compared with the prior art, the positive effect of the present invention is: the present invention identifies idling from the perspective of signal feature extraction and machine learning classification, and can significantly represent the idling state / The characteristics of the sticking state are classified by a classifier to realize idling online recognition. Compared with the traditional idling recognition method using input with errors, the effect of online recognition is more excellent, more accurate and faster; the method of the present invention is aimed at the complex The adaptability of the operating conditions and operating environment is stronger than the traditional method, and it has better recognition accuracy than the traditional idling recognition method, and effectively solves the problem of difficult threshold setting in the online recognition in the traditional method.

Description

technical field [0001] The invention relates to rail transit train traction control technology, in particular to an online identification method for idling electric locomotives. Background technique [0002] Transportation plays an increasingly important role in economic construction, and railway transportation is the backbone of the transportation field. In railway transportation, the adhesion between the moving wheels and the wheel rails of electric locomotives is the ultimate driving force to drive the locomotives. Therefore, only by ensuring that the effective adhesion between the wheel rails is not destroyed, that is, to avoid idling of the locomotive wheels, can the traction be effectively used. The output power of the motor maximizes the power utilization of the traction motor. However, due to the operating conditions of the locomotive and the complex environment, and the adhesion between the wheel and rail is largely dependent on the condition of the wheel-rail cont...

AI Technical Summary

Problems solved by technology

[0021] From the foregoing, it can be seen that the current main locomotive idling identification methods are all signal analysis methods directly based on input data, and there is no effective extraction of high-precision features representing the idling state in the signal for idling identification; at the same time, the existing methods are accurate in online identification. There are deficiencies in performance, rapidity, linearity or anti-interference

Images