Intelligent train traction fault big data abnormality detection and recognition method

Abstract

The invention discloses an intelligent train traction fault big data abnormality detection and recognition method. The method includes: collecting working signals of traction equipment of an intelligent train in real time; sampling the working signals to acquire a sampling dataset; using an isolated forest abnormal value quick detector to quickly detect the sampling dataset, and acquiring remarkable abnormal point in the working signals; building a Hampel sliding window identifier filter according to the remarkable abnormal point, and processing the sampling dataset to acquire a sampling dataset after being processed; acquiring a training sample set, and training a classifier based on a slack supporting vector machine by taking processed sampling data of a training sample as input and corresponding fault type as output; using a traction fault diagnosis classifier acquired by training to diagnose the sampling dataset after being processed and acquired in the step 3, and detecting faulttype of tiny and suspicious fault in the working signal S. By the method, fault detection efficiency and diagnosis accuracy of the train traction equipment are improved.

Description

technical field [0001] The invention relates to the field of mechanical fault diagnosis, in particular to an intelligent train traction fault big data anomaly detection and identification method. Background technique [0002] With the development of modernization, the degree of automation of modern machinery and equipment is getting higher and higher, the functions are more and more comprehensive, and the corresponding mechanical complexity and precision are getting higher and higher. The traction equipment of the train is the most important guarantee for the normal operation of the train. The failure of the train traction equipment will lead to the inability of the whole vehicle to run, which will bring serious safety hazards and economic losses. However, some traditional train traction equipment diagnosis methods and manual judgment methods are difficult to meet the fault diagnosis needs of modern intelligent trains, and the diagnostic accuracy and judgment accuracy of the...

AI Technical Summary

Problems solved by technology

The traction equipment of the train is the most important guarantee for the normal operation of the train. The failure of the train traction equipment will lead to the inoperability of the whole vehicle, which will bring serious safety hazards and economic losses.

However, some traditional train traction equipment diagnosis methods and manual judgment methods are difficult to meet the fault diagnosis needs of modern intelligent trains, and the diagnosis accuracy and judgment accuracy of these methods are difficult to meet the requirements

[0003] In order to make up for the shortcomings of traditional train manual diagnosis methods, the method of using sensors to collect working data signals of train traction equipment, and analyzing and diagnosing vibration data is gradually popularized. Some new machine learning intelligent diagnosis methods are applied to mechanical fault diagnosis. However, the accuracy of some intelligent fault diagnosis methods proposed before is difficult to be guaranteed.

Images