Haiyang Pan , Zhangping Wu , Jian Cheng , Jinde Zheng , Shuchao Deng , Jinyu Tong , Long Zhang
{"title":"Overlapping resonance compensation: A composite fault detection methodology","authors":"Haiyang Pan , Zhangping Wu , Jian Cheng , Jinde Zheng , Shuchao Deng , Jinyu Tong , Long Zhang","doi":"10.1016/j.ymssp.2025.113430","DOIUrl":null,"url":null,"abstract":"<div><div>The frequency bands corresponding to various state characteristics of composite faults typically exhibit superposition, mixing, and nonlinear characteristics. Conventional frequency band segmentation methods frequently disrupt resonant frequency bands, thereby hindering accurate extraction and separation of composite fault features. To overcome this limitation, this paper proposes a novel neighborhood extension Ramanujan decomposition (NERD) method guided by the neighborhood extension factor (NEF). Firstly, it implements frequency band cross-segmentation through NEF, representing a soft-segmentation approach that permits overlapping boundaries between adjacent frequency bands. This method can capture local features more accurately while maintaining the integrity of the frequency band and achieve compensation for overlapping resonant frequency bands. Secondly, the NEF employs the signal’s energy distribution as a reference framework, quantifying the membership degree of overlapping regions through both macroscopic rate of change and microscopic stability metrics. This dual perspective approach facilitates better transmission of coupled composite fault information across frequency band boundaries while minimizing the loss of effective signal components. Furthermore, the method defines the generalized Ramanujan periodic aggregation index (GRPA), which visualizes fault information within filtered components, thereby enabling precise extraction of composite fault features. Comprehensive validation using both simulated bearing fault signals and experimental datasets confirms the efficacy and superiority of the NERD method.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"240 ","pages":"Article 113430"},"PeriodicalIF":8.9000,"publicationDate":"2025-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Mechanical Systems and Signal Processing","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0888327025011318","RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, MECHANICAL","Score":null,"Total":0}
引用次数: 0
Abstract
The frequency bands corresponding to various state characteristics of composite faults typically exhibit superposition, mixing, and nonlinear characteristics. Conventional frequency band segmentation methods frequently disrupt resonant frequency bands, thereby hindering accurate extraction and separation of composite fault features. To overcome this limitation, this paper proposes a novel neighborhood extension Ramanujan decomposition (NERD) method guided by the neighborhood extension factor (NEF). Firstly, it implements frequency band cross-segmentation through NEF, representing a soft-segmentation approach that permits overlapping boundaries between adjacent frequency bands. This method can capture local features more accurately while maintaining the integrity of the frequency band and achieve compensation for overlapping resonant frequency bands. Secondly, the NEF employs the signal’s energy distribution as a reference framework, quantifying the membership degree of overlapping regions through both macroscopic rate of change and microscopic stability metrics. This dual perspective approach facilitates better transmission of coupled composite fault information across frequency band boundaries while minimizing the loss of effective signal components. Furthermore, the method defines the generalized Ramanujan periodic aggregation index (GRPA), which visualizes fault information within filtered components, thereby enabling precise extraction of composite fault features. Comprehensive validation using both simulated bearing fault signals and experimental datasets confirms the efficacy and superiority of the NERD method.
期刊介绍:
Journal Name: Mechanical Systems and Signal Processing (MSSP)
Interdisciplinary Focus:
Mechanical, Aerospace, and Civil Engineering
Purpose:Reporting scientific advancements of the highest quality
Arising from new techniques in sensing, instrumentation, signal processing, modelling, and control of dynamic systems