{"title":"工业锅炉故障建模、识别与检测(2015.07)","authors":"Navaseelan Paul David, B. Swaminathan","doi":"10.1109/TIAR.2015.7358557","DOIUrl":null,"url":null,"abstract":"The Boiler plays vital role in electric power plants, fertilizer industries, petrochemical and in other industries. In such industries the boiler actuates turbines, compressors for generating electric power, pneumatic power respectively. The overall operation and efficiency of any plant is depending on the quality of steam produced in terms of its flow rate, pressure and temperature and also reliability. Any kind of fault or failure of the boiler may reduce the quality of production and tend to shut down the operation of entire plant. Hence early detection of faults will enhance availability of steam and reduce plant shutdown. In order to diagnose the faults, complete operation of all the loops like feed water circuit, air fuel circuit, steam circuit and cooling water circuit are studied and possible failures at the input side, inside the boiler and output side of the boiler are studied. One such 130 tons per hour capacity water tube boiler in Petrochemical industry is studied. The required data for complete transient part of its operation is collected for identification of the boiler model. In this paper simple models using first principle balance equations were developed for the subsystems of the boiler like furnace, boiler tubes and drums and heat exchangers. The mathematical models are also obtained based on measured data during real time operation of the boiler. Then the parameters are identified by choosing proper model structure like non-linear ARX and Hammerstein-Wiener and it is validated with real time plant data. The model based fault detection using Kalman filter algorithm is presented in this paper among the different methods being practiced. In this method, Kalman filter estimates all the process variables at the input side, output side and inside the boiler. Residual is generated as the difference between measured and estimated values of these variables. If the residual generated surpasses threshold value indicates fault in the boiler.","PeriodicalId":281784,"journal":{"name":"2015 IEEE Technological Innovation in ICT for Agriculture and Rural Development (TIAR)","volume":"155 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2015-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":"{\"title\":\"Modeling, identification and detection of faults in industrial boiler(July2015)\",\"authors\":\"Navaseelan Paul David, B. Swaminathan\",\"doi\":\"10.1109/TIAR.2015.7358557\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The Boiler plays vital role in electric power plants, fertilizer industries, petrochemical and in other industries. In such industries the boiler actuates turbines, compressors for generating electric power, pneumatic power respectively. The overall operation and efficiency of any plant is depending on the quality of steam produced in terms of its flow rate, pressure and temperature and also reliability. Any kind of fault or failure of the boiler may reduce the quality of production and tend to shut down the operation of entire plant. Hence early detection of faults will enhance availability of steam and reduce plant shutdown. In order to diagnose the faults, complete operation of all the loops like feed water circuit, air fuel circuit, steam circuit and cooling water circuit are studied and possible failures at the input side, inside the boiler and output side of the boiler are studied. One such 130 tons per hour capacity water tube boiler in Petrochemical industry is studied. The required data for complete transient part of its operation is collected for identification of the boiler model. In this paper simple models using first principle balance equations were developed for the subsystems of the boiler like furnace, boiler tubes and drums and heat exchangers. The mathematical models are also obtained based on measured data during real time operation of the boiler. Then the parameters are identified by choosing proper model structure like non-linear ARX and Hammerstein-Wiener and it is validated with real time plant data. The model based fault detection using Kalman filter algorithm is presented in this paper among the different methods being practiced. In this method, Kalman filter estimates all the process variables at the input side, output side and inside the boiler. Residual is generated as the difference between measured and estimated values of these variables. If the residual generated surpasses threshold value indicates fault in the boiler.\",\"PeriodicalId\":281784,\"journal\":{\"name\":\"2015 IEEE Technological Innovation in ICT for Agriculture and Rural Development (TIAR)\",\"volume\":\"155 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2015-07-10\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"1\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2015 IEEE Technological Innovation in ICT for Agriculture and Rural Development (TIAR)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/TIAR.2015.7358557\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2015 IEEE Technological Innovation in ICT for Agriculture and Rural Development (TIAR)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/TIAR.2015.7358557","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Modeling, identification and detection of faults in industrial boiler(July2015)
The Boiler plays vital role in electric power plants, fertilizer industries, petrochemical and in other industries. In such industries the boiler actuates turbines, compressors for generating electric power, pneumatic power respectively. The overall operation and efficiency of any plant is depending on the quality of steam produced in terms of its flow rate, pressure and temperature and also reliability. Any kind of fault or failure of the boiler may reduce the quality of production and tend to shut down the operation of entire plant. Hence early detection of faults will enhance availability of steam and reduce plant shutdown. In order to diagnose the faults, complete operation of all the loops like feed water circuit, air fuel circuit, steam circuit and cooling water circuit are studied and possible failures at the input side, inside the boiler and output side of the boiler are studied. One such 130 tons per hour capacity water tube boiler in Petrochemical industry is studied. The required data for complete transient part of its operation is collected for identification of the boiler model. In this paper simple models using first principle balance equations were developed for the subsystems of the boiler like furnace, boiler tubes and drums and heat exchangers. The mathematical models are also obtained based on measured data during real time operation of the boiler. Then the parameters are identified by choosing proper model structure like non-linear ARX and Hammerstein-Wiener and it is validated with real time plant data. The model based fault detection using Kalman filter algorithm is presented in this paper among the different methods being practiced. In this method, Kalman filter estimates all the process variables at the input side, output side and inside the boiler. Residual is generated as the difference between measured and estimated values of these variables. If the residual generated surpasses threshold value indicates fault in the boiler.