{"title":"最优控制系统中的传感器故障检测","authors":"M. Saif, F. E. Villaseca","doi":"10.23919/ACC.1986.4789278","DOIUrl":null,"url":null,"abstract":"A new scheme for solving the sensor failure detection, isolation and accommodation (FDIA) problem in optimally controlled linear systems has been developed. The proposed technique uses Luenberger observers with a detection logic based on the minimum value of the cost functional at any time instant. The advantages of this new technique are: (1) it can detect and isolate as many failed sensors as there are sensors, even when the failures are multiple and simultaneous; (2) FDIA can be accomplished at any time, since it does not depend on the convergence of the observer estimates to the actual system states; (3) it is not applications dependent, since it requires neither a priori information concerning likely failure modes nor the knowledge of the nominal path resulting from a specific input command. The method assumes the system is observable from each available output.","PeriodicalId":266163,"journal":{"name":"1986 American Control Conference","volume":"48 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1986-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"6","resultStr":"{\"title\":\"Sensor Failure Detection in Optimally Controlled Systems\",\"authors\":\"M. Saif, F. E. Villaseca\",\"doi\":\"10.23919/ACC.1986.4789278\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"A new scheme for solving the sensor failure detection, isolation and accommodation (FDIA) problem in optimally controlled linear systems has been developed. The proposed technique uses Luenberger observers with a detection logic based on the minimum value of the cost functional at any time instant. The advantages of this new technique are: (1) it can detect and isolate as many failed sensors as there are sensors, even when the failures are multiple and simultaneous; (2) FDIA can be accomplished at any time, since it does not depend on the convergence of the observer estimates to the actual system states; (3) it is not applications dependent, since it requires neither a priori information concerning likely failure modes nor the knowledge of the nominal path resulting from a specific input command. The method assumes the system is observable from each available output.\",\"PeriodicalId\":266163,\"journal\":{\"name\":\"1986 American Control Conference\",\"volume\":\"48 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"1986-06-18\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"6\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"1986 American Control Conference\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.23919/ACC.1986.4789278\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"1986 American Control Conference","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.23919/ACC.1986.4789278","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Sensor Failure Detection in Optimally Controlled Systems
A new scheme for solving the sensor failure detection, isolation and accommodation (FDIA) problem in optimally controlled linear systems has been developed. The proposed technique uses Luenberger observers with a detection logic based on the minimum value of the cost functional at any time instant. The advantages of this new technique are: (1) it can detect and isolate as many failed sensors as there are sensors, even when the failures are multiple and simultaneous; (2) FDIA can be accomplished at any time, since it does not depend on the convergence of the observer estimates to the actual system states; (3) it is not applications dependent, since it requires neither a priori information concerning likely failure modes nor the knowledge of the nominal path resulting from a specific input command. The method assumes the system is observable from each available output.