A track fault identification method and system based on HSROA
Through the signal processing method and deep belief network combining HSROA and EEMD, the problems of low efficiency and insufficient accuracy of traditional track fault detection are solved, and intelligent and accurate identification of track faults is achieved.
CN120277617BActive Publication Date: 2025-09-12EAST CHINA JIAOTONG UNIVERSITY
0 Cites 0 Cited by
Patent Information
- Application Number
- CN202510749047.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-06
- Publication Date
- 2025-09-12
- Estimated Expiration
- 2045-06-06
AI Technical Summary
Technical Problem
Traditional rail fault detection methods are inefficient and easily affected by environmental noise, making it difficult to accurately identify complex fault types.
Method used
The signal processing method combining HSROA and EEMD is adopted. Through the hyperband signal refinement optimization algorithm, EEMD decomposition, bandpass filtering and envelope power spectrum analysis, a fusion feature matrix is constructed, and the deep belief network is used for fault identification.
Benefits of technology
It effectively removes environmental noise, accurately screens out characteristic information of track faults, improves the ability to identify complex types of defects, and realizes intelligent and precise track fault identification.
✦ Generated by Eureka AI based on patent content.
Abstract
The present invention discloses a track fault identification method and system based on HSROA, comprising: calculating the frequency band entropy value of at least one IMF component signal, selecting a sensitive IMF component signal from the at least one IMF component signal according to the frequency band entropy value; performing a second filtering on the sensitive IMF component signal according to a bandpass filter to obtain a target signal, and performing envelope power spectrum analysis on the target signal to obtain a fault characteristic frequency; fusing the fault characteristic frequency with time domain features, frequency domain features, and time-frequency domain features in a track vibration signal to obtain a fused feature vector, and constructing a fused feature matrix according to the fused feature vector; inputting the fused feature matrix into a preset deep belief network for iterative training to obtain a track fault identification model, and inputting the acquired real-time track vibration signal into the track fault identification model, which outputs a track fault identification result.
Need to check novelty before this filing date? Find Prior Art