Xu Weng , Xiaobin Xu , Xiping Wang , Li Yang , Zhe Zhou
{"title":"可靠报警生成的混合随机-动态框架:基于gmm的概率模式与时间证据融合的集成","authors":"Xu Weng , Xiaobin Xu , Xiping Wang , Li Yang , Zhe Zhou","doi":"10.1016/j.cherd.2025.08.043","DOIUrl":null,"url":null,"abstract":"<div><div>Industrial alarm systems typically utilize critical process variables for anomaly detection. This paper introduces a probabilistic evidence fusion framework that integrates long-term probabilistic trends with short-term temporal dynamics to enhance robust alarm generation. For modeling long-term probabilistic trends, Gaussian mixture models (GMMs) are employed to estimate baseline probability density functions (PDFs) for normal and abnormal states based on historical data. Online samples are then converted into probabilistic evidence via normalized likelihood comparisons against these baseline PDFs. To address short-term characteristics, a temporal evidence combination mechanism is developed, which dynamically fuses current and historical probabilistic evidence using the evidence reasoning (ER) rule. Alarms are subsequently generated based on the fused results. This dual-scale modeling approach effectively captures both persistent stochastic patterns and transient behaviors, thereby ensuring reliable detection across various operational regimes. Numerical simulations and case studies of motor fault detection validate superior performance of the proposed framework in terms of detection accuracy compared to conventional methods, particularly in handling gradual fault progression and transient disturbances.</div></div>","PeriodicalId":10019,"journal":{"name":"Chemical Engineering Research & Design","volume":"222 ","pages":"Pages 73-85"},"PeriodicalIF":3.9000,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"A hybrid stochastic-dynamic framework for reliable alarm generation: Integrating GMM-based probability patterns with temporal evidence fusion\",\"authors\":\"Xu Weng , Xiaobin Xu , Xiping Wang , Li Yang , Zhe Zhou\",\"doi\":\"10.1016/j.cherd.2025.08.043\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><div>Industrial alarm systems typically utilize critical process variables for anomaly detection. This paper introduces a probabilistic evidence fusion framework that integrates long-term probabilistic trends with short-term temporal dynamics to enhance robust alarm generation. For modeling long-term probabilistic trends, Gaussian mixture models (GMMs) are employed to estimate baseline probability density functions (PDFs) for normal and abnormal states based on historical data. Online samples are then converted into probabilistic evidence via normalized likelihood comparisons against these baseline PDFs. To address short-term characteristics, a temporal evidence combination mechanism is developed, which dynamically fuses current and historical probabilistic evidence using the evidence reasoning (ER) rule. Alarms are subsequently generated based on the fused results. This dual-scale modeling approach effectively captures both persistent stochastic patterns and transient behaviors, thereby ensuring reliable detection across various operational regimes. Numerical simulations and case studies of motor fault detection validate superior performance of the proposed framework in terms of detection accuracy compared to conventional methods, particularly in handling gradual fault progression and transient disturbances.</div></div>\",\"PeriodicalId\":10019,\"journal\":{\"name\":\"Chemical Engineering Research & Design\",\"volume\":\"222 \",\"pages\":\"Pages 73-85\"},\"PeriodicalIF\":3.9000,\"publicationDate\":\"2025-09-02\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Chemical Engineering Research & Design\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S0263876225004630\",\"RegionNum\":3,\"RegionCategory\":\"工程技术\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"ENGINEERING, CHEMICAL\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Chemical Engineering Research & Design","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0263876225004630","RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENGINEERING, CHEMICAL","Score":null,"Total":0}
A hybrid stochastic-dynamic framework for reliable alarm generation: Integrating GMM-based probability patterns with temporal evidence fusion
Industrial alarm systems typically utilize critical process variables for anomaly detection. This paper introduces a probabilistic evidence fusion framework that integrates long-term probabilistic trends with short-term temporal dynamics to enhance robust alarm generation. For modeling long-term probabilistic trends, Gaussian mixture models (GMMs) are employed to estimate baseline probability density functions (PDFs) for normal and abnormal states based on historical data. Online samples are then converted into probabilistic evidence via normalized likelihood comparisons against these baseline PDFs. To address short-term characteristics, a temporal evidence combination mechanism is developed, which dynamically fuses current and historical probabilistic evidence using the evidence reasoning (ER) rule. Alarms are subsequently generated based on the fused results. This dual-scale modeling approach effectively captures both persistent stochastic patterns and transient behaviors, thereby ensuring reliable detection across various operational regimes. Numerical simulations and case studies of motor fault detection validate superior performance of the proposed framework in terms of detection accuracy compared to conventional methods, particularly in handling gradual fault progression and transient disturbances.
期刊介绍:
ChERD aims to be the principal international journal for publication of high quality, original papers in chemical engineering.
Papers showing how research results can be used in chemical engineering design, and accounts of experimental or theoretical research work bringing new perspectives to established principles, highlighting unsolved problems or indicating directions for future research, are particularly welcome. Contributions that deal with new developments in plant or processes and that can be given quantitative expression are encouraged. The journal is especially interested in papers that extend the boundaries of traditional chemical engineering.