{"title":"工业过程的健康监测。挑战和解决方案","authors":"Shen Yin","doi":"10.1109/ICET.2016.7813282","DOIUrl":null,"url":null,"abstract":"To ensure the health and reliability of increasingly complicated industrial processes, the study and application of health monitoring systems is necessary. Considering the difficulty of mathematical modeling and the availability of massive process data, the so-called data-driven methods gain advantage over model-based ones. Therefore, it is promising and significant to design efficient data-driven health monitoring schemes for particular industrial conditions. In this talk, three circumstances, i.e. stationary operating conditions, dynamic processes and large-scale processes involving changes, are investigated respectively. Correspondingly, the modifications of the standard multivariate statistical approaches, the advanced schemes based on the identification of key components and a novel data-driven adaptive scheme are proposed, which all provide enhanced health monitoring performance.","PeriodicalId":285090,"journal":{"name":"2016 International Conference on Emerging Technologies (ICET)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Health monitoring of industrial processes — Challenges and solutions\",\"authors\":\"Shen Yin\",\"doi\":\"10.1109/ICET.2016.7813282\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"To ensure the health and reliability of increasingly complicated industrial processes, the study and application of health monitoring systems is necessary. Considering the difficulty of mathematical modeling and the availability of massive process data, the so-called data-driven methods gain advantage over model-based ones. Therefore, it is promising and significant to design efficient data-driven health monitoring schemes for particular industrial conditions. In this talk, three circumstances, i.e. stationary operating conditions, dynamic processes and large-scale processes involving changes, are investigated respectively. Correspondingly, the modifications of the standard multivariate statistical approaches, the advanced schemes based on the identification of key components and a novel data-driven adaptive scheme are proposed, which all provide enhanced health monitoring performance.\",\"PeriodicalId\":285090,\"journal\":{\"name\":\"2016 International Conference on Emerging Technologies (ICET)\",\"volume\":\"24 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2016-10-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2016 International Conference on Emerging Technologies (ICET)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICET.2016.7813282\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2016 International Conference on Emerging Technologies (ICET)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICET.2016.7813282","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Health monitoring of industrial processes — Challenges and solutions
To ensure the health and reliability of increasingly complicated industrial processes, the study and application of health monitoring systems is necessary. Considering the difficulty of mathematical modeling and the availability of massive process data, the so-called data-driven methods gain advantage over model-based ones. Therefore, it is promising and significant to design efficient data-driven health monitoring schemes for particular industrial conditions. In this talk, three circumstances, i.e. stationary operating conditions, dynamic processes and large-scale processes involving changes, are investigated respectively. Correspondingly, the modifications of the standard multivariate statistical approaches, the advanced schemes based on the identification of key components and a novel data-driven adaptive scheme are proposed, which all provide enhanced health monitoring performance.