A temperature anomaly detection method for high-speed trains based on spatio-temporal fusion decision-making

Abstract

The invention discloses a method for detecting abnormal temperature of a high-speed train based on a time-space fusion decision. Through the fusion of the time-dimension axle temperature anomaly detection model and the space-dimension axle temperature anomaly detection model, combined with the D-S evidence theory, the temperature anomaly of high-speed train vehicle components is detected based on spatio-temporal fusion. The invention uses the D-S evidence theory to combine the temperature anomaly detection models in the two dimensions of time and space, which can not only reduce the false alarm rate, but also ensure a large amount of early warning, and realizes the "time and space dimension model". Take advantage of strengths and avoid weaknesses."

Description

technical field [0001] The invention relates to a temperature abnormality detection method of a high-speed train, which belongs to the field of temperature abnormality prediction during high-speed train operation, and is a method for detecting temperature abnormality during high-speed train operation based on deep learning. Background technique [0002] The high-speed train temperature prediction method is to transmit the original signals collected by several sensors to the computer during the running of the high-speed train, and perform a series of signal processing on the original signals to establish a mapping relationship with real-time temperature prediction, and finally through intelligent The temperature prediction algorithm is used to predict the temperature of vehicle components for a period of time after the current time, and use it as the basis for abnormal state discrimination. [0003] The current vehicle-mounted intelligent operation and maintenance system for ...

AI Technical Summary

Problems solved by technology

[0003] The current vehicle-mounted intelligent operation and maintenance system for high-speed trains is mainly a monitoring of temperature, and the state identification of vehicle components is realized by formulating corresponding logic rules and temperature thresholds. This method can avoid major safety accidents, but once such an alarm occurs, it means Significant damage to components, accidents in vehicle operation, and early abnormal detection of component temperature, fault location and early warning cannot be achieved

[0004] In recent years, the use of temperature data has provided a wealth of ideas and methods for the detection of temperature anomalies in high-speed trains, but there are still defects that it is difficult to realize the early warning of abnormal temperature rises in the on-board intelligent operation and maintenance system.

Images