{"title":"基于双元的钻床刀具状态预测模型","authors":"Sunidhi Dayam, K. A. Desai","doi":"10.1115/msec2022-85449","DOIUrl":null,"url":null,"abstract":"\n Digital Twin technology can be effectively employed for prognosis and predictive maintenance tasks by establishing interconnections between manufacturing equipment and its virtual counterpart. This paper presents the Tool State Prognosis (TSP) model based on Digital Twin philosophy to perceive the state of a twist drill during the drilling operation. The TSP model estimates the state of a twist drill viz. initial, intermediate, or worn during the operation rather than obtaining the precise wear value. The Digital Twin collects input information as time-series data by establishing an appropriate connection protocol with a drilling machine using vibration and acoustic emission sensors. The Root Mean Square (RMS) approach and Quadratic Support Vector Machine (QSVM) are employed for feature extraction and recognizing the twist drill status with Remaining Useful Life (RUL) prediction from the time-series data. The model also includes integrating a Human Machine Interface (HMI) unit for displaying tool status and RUL information to assist operators in tool replacement decisions. The developed model can be integrated as an edge-level solution with manual and CNC drilling machines without significant hardware changes for perceiving the status of a twist drill. The prediction abilities of the digital twin are corroborated by performing a set of drilling experiments for various cutting tool-workpiece combinations. The confusion matrices demonstrated the effectiveness and generalization abilities of the developed model by comparing predicted and actual classes for each combination. The developed Digital Twin model can quickly respond to tool status and failure with enhanced man-machine interactions and improved prognosis abilities for the drilling machines.","PeriodicalId":23676,"journal":{"name":"Volume 2: Manufacturing Processes; Manufacturing Systems; Nano/Micro/Meso Manufacturing; Quality and Reliability","volume":"7 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2022-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Digital Twin-Based Tool State Prognosis Model for Drilling Machines\",\"authors\":\"Sunidhi Dayam, K. A. Desai\",\"doi\":\"10.1115/msec2022-85449\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"\\n Digital Twin technology can be effectively employed for prognosis and predictive maintenance tasks by establishing interconnections between manufacturing equipment and its virtual counterpart. This paper presents the Tool State Prognosis (TSP) model based on Digital Twin philosophy to perceive the state of a twist drill during the drilling operation. The TSP model estimates the state of a twist drill viz. initial, intermediate, or worn during the operation rather than obtaining the precise wear value. The Digital Twin collects input information as time-series data by establishing an appropriate connection protocol with a drilling machine using vibration and acoustic emission sensors. The Root Mean Square (RMS) approach and Quadratic Support Vector Machine (QSVM) are employed for feature extraction and recognizing the twist drill status with Remaining Useful Life (RUL) prediction from the time-series data. The model also includes integrating a Human Machine Interface (HMI) unit for displaying tool status and RUL information to assist operators in tool replacement decisions. The developed model can be integrated as an edge-level solution with manual and CNC drilling machines without significant hardware changes for perceiving the status of a twist drill. The prediction abilities of the digital twin are corroborated by performing a set of drilling experiments for various cutting tool-workpiece combinations. The confusion matrices demonstrated the effectiveness and generalization abilities of the developed model by comparing predicted and actual classes for each combination. The developed Digital Twin model can quickly respond to tool status and failure with enhanced man-machine interactions and improved prognosis abilities for the drilling machines.\",\"PeriodicalId\":23676,\"journal\":{\"name\":\"Volume 2: Manufacturing Processes; Manufacturing Systems; Nano/Micro/Meso Manufacturing; Quality and Reliability\",\"volume\":\"7 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2022-06-27\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Volume 2: Manufacturing Processes; Manufacturing Systems; Nano/Micro/Meso Manufacturing; Quality and Reliability\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1115/msec2022-85449\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Volume 2: Manufacturing Processes; Manufacturing Systems; Nano/Micro/Meso Manufacturing; Quality and Reliability","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1115/msec2022-85449","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Digital Twin-Based Tool State Prognosis Model for Drilling Machines
Digital Twin technology can be effectively employed for prognosis and predictive maintenance tasks by establishing interconnections between manufacturing equipment and its virtual counterpart. This paper presents the Tool State Prognosis (TSP) model based on Digital Twin philosophy to perceive the state of a twist drill during the drilling operation. The TSP model estimates the state of a twist drill viz. initial, intermediate, or worn during the operation rather than obtaining the precise wear value. The Digital Twin collects input information as time-series data by establishing an appropriate connection protocol with a drilling machine using vibration and acoustic emission sensors. The Root Mean Square (RMS) approach and Quadratic Support Vector Machine (QSVM) are employed for feature extraction and recognizing the twist drill status with Remaining Useful Life (RUL) prediction from the time-series data. The model also includes integrating a Human Machine Interface (HMI) unit for displaying tool status and RUL information to assist operators in tool replacement decisions. The developed model can be integrated as an edge-level solution with manual and CNC drilling machines without significant hardware changes for perceiving the status of a twist drill. The prediction abilities of the digital twin are corroborated by performing a set of drilling experiments for various cutting tool-workpiece combinations. The confusion matrices demonstrated the effectiveness and generalization abilities of the developed model by comparing predicted and actual classes for each combination. The developed Digital Twin model can quickly respond to tool status and failure with enhanced man-machine interactions and improved prognosis abilities for the drilling machines.