b'Intensified CharacterizationMachine learning-based analytics to identify and quantify failure modes and Analysis of Energy Storageand effects enable intelligent approach to battery energy storage System to Support Integratedsystemsdevelopment.Energy Systems B attery energy storage systems play a pivotal role in the advancements of the portable electronics and Internet of Things, the electrification of transportation and mobility, and the energy efficiency and security in the supply chain of the power grid. To enable more reliable functionalities of lithium-ion battery energy storage systems to support this intensified electrification in the energy supply chain, basic understanding of how to improve the durability, performance, reliability, and safety of the lithium-ion battery energy storage systems TOTAL APPROVED AMOUNT:is vital. Development of a consistent platform to evaluate, diagnose, and assure $1,894,656 over 3 years the quality of the battery and its function through the entire life cyclefrom PROJECT NUMBER:material preparation, component and battery design, fabrication, and control and 19P45-013 management of the lithium-ion battery energy storage systems in operation is needed to yield better functionalities.PRINCIPAL INVESTIGATOR: Boryann Liaw This project developed a quantitative failure mode and effect analysis platform that can enhance battery diagnostic and prognostic capability with advanced CO-INVESTIGATORS: data analytic and regression techniques. The approach combined quantitative Bor-Rong Chen, INL failure analyses and separated algorithmic and mechanistic model regressions Gorakh Pawar, INL and predictions, respectively, using advanced artificial intelligence and machine John Koudelka, INL learning-based analytics to identify failure mechanisms, quantify the effects, verify Meng Li, INL the analyses, and validate the predictions. This novel approach achieved a high Qiang Wang, INL precision and accurate analytic platform for future high-throughput data handling Ross Kunz, INL and high-performance computing-based failure mode and effect analysis processes.Leslie Kerby, Idaho State University78'