Quang Le, M. Hossain, Tristan M. Evans, Yarui Peng, H. Mantooth
{"title":"Thermal Runaway Mitigation through Electrothermal Constraints Mapping for MCPM Layout Optimization","authors":"Quang Le, M. Hossain, Tristan M. Evans, Yarui Peng, H. Mantooth","doi":"10.1109/DMC55175.2022.9906468","DOIUrl":null,"url":null,"abstract":"Along with the developments in power electronic packaging technology, many studies on design automation for MCPMs layout further push the design limits for their power density and compactness. Among these studies, PowerSynth has shown the complete design flow for MCPMs, which offers a multiobjective layout optimization algorithm and reduced-order models for electrical parasitic extraction and thermal evaluation. While these models are accurate, there is no connection between the electrical parasitic and device temperature during the layout optimization process. Hence, the multi-objective optimization algorithm optimizes these objectives separately without insights into their impacts on the reliability and performance of the wide bandgap (WBG) device. This limitation can lead to a layout solution with undesirable performance compared to the WBG device’s safe operation area (SOA). Therefore, this work incorporates the WBG physics-based device knowledge into the power loss calculation for a more accurate electro-thermal prediction in PowerSynth. A better decision can then be made on the most suitable thermal management system.","PeriodicalId":245908,"journal":{"name":"2022 IEEE Design Methodologies Conference (DMC)","volume":"357 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 IEEE Design Methodologies Conference (DMC)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/DMC55175.2022.9906468","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
Abstract
Along with the developments in power electronic packaging technology, many studies on design automation for MCPMs layout further push the design limits for their power density and compactness. Among these studies, PowerSynth has shown the complete design flow for MCPMs, which offers a multiobjective layout optimization algorithm and reduced-order models for electrical parasitic extraction and thermal evaluation. While these models are accurate, there is no connection between the electrical parasitic and device temperature during the layout optimization process. Hence, the multi-objective optimization algorithm optimizes these objectives separately without insights into their impacts on the reliability and performance of the wide bandgap (WBG) device. This limitation can lead to a layout solution with undesirable performance compared to the WBG device’s safe operation area (SOA). Therefore, this work incorporates the WBG physics-based device knowledge into the power loss calculation for a more accurate electro-thermal prediction in PowerSynth. A better decision can then be made on the most suitable thermal management system.