Accurate diagnosis of the health degradation of retired batteries is crucial for ensuring their safe and reliable reuse. While machine learning offers promising solutions, training models to overcome the high heterogeneity of retired batteries requires massive degradation data, leading to high testing costs and substantial energy waste. Here, we reveal the potential to recover large-scale and high-quality second-life battery datasets from field data to assist in diagnosing the health of retired batteries. By seamlessly fusing deep learning and domain knowledge, we enabled the accurate recovery of capacity labels for operating data without regular fully charging or discharging calibrations. To validate the proposed method, we develop a large-scale degradation test on 96 realistic retired batteries, performing over 50,000 charge/discharge cycles to simulate different stationary energy storage scenarios with 24 charge-discharge intervals. With only 3 capacity measurements available over the second-life, the proposed method accurately recovers the capacity labels with a root mean square error below 30 mAh. Furthermore, the health diagnostic model trained on regenerated dataset is comparable to the model trained on real data, with an almost negligible error of less than 5 mAh. More importantly, we expect to save at least 98% of test time, electricity and energy consumption to generate datasets cost-effectively. This study highlights the potential of field data to bridge the critical data gap in diagnosing the health degradation of retired batteries.
{"title":"Regenerate large-scale retired second-life battery datasets via recovered capacity labels-based deep learning","authors":"Yuchen Xu, Tianxiang Zeng, Weiwen Peng, Jinpeng Tian, Xiaojian Yi, Qizhi Xu, Shun-Peng Zhu","doi":"10.1016/j.ensm.2026.104947","DOIUrl":"https://doi.org/10.1016/j.ensm.2026.104947","url":null,"abstract":"Accurate diagnosis of the health degradation of retired batteries is crucial for ensuring their safe and reliable reuse. While machine learning offers promising solutions, training models to overcome the high heterogeneity of retired batteries requires massive degradation data, leading to high testing costs and substantial energy waste. Here, we reveal the potential to recover large-scale and high-quality second-life battery datasets from field data to assist in diagnosing the health of retired batteries. By seamlessly fusing deep learning and domain knowledge, we enabled the accurate recovery of capacity labels for operating data without regular fully charging or discharging calibrations. To validate the proposed method, we develop a large-scale degradation test on 96 realistic retired batteries, performing over 50,000 charge/discharge cycles to simulate different stationary energy storage scenarios with 24 charge-discharge intervals. With only 3 capacity measurements available over the second-life, the proposed method accurately recovers the capacity labels with a root mean square error below 30 mAh. Furthermore, the health diagnostic model trained on regenerated dataset is comparable to the model trained on real data, with an almost negligible error of less than 5 mAh. More importantly, we expect to save at least 98% of test time, electricity and energy consumption to generate datasets cost-effectively. This study highlights the potential of field data to bridge the critical data gap in diagnosing the health degradation of retired batteries.","PeriodicalId":306,"journal":{"name":"Energy Storage Materials","volume":"31 1","pages":""},"PeriodicalIF":20.4,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146089610","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-30DOI: 10.1016/j.ensm.2026.104934
Changhyeon Lee, Kiyeon Sim, Jinhyeon Jo, KwangSup Eom
{"title":"Construction of Heterostructured Multi-Grain Solid Electrolyte Interphase with Trace Alloying for Fast Li ion Transfer and Dendrite-Free Lithium Metal Batteries","authors":"Changhyeon Lee, Kiyeon Sim, Jinhyeon Jo, KwangSup Eom","doi":"10.1016/j.ensm.2026.104934","DOIUrl":"https://doi.org/10.1016/j.ensm.2026.104934","url":null,"abstract":"","PeriodicalId":306,"journal":{"name":"Energy Storage Materials","volume":"30 1","pages":""},"PeriodicalIF":20.4,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146072525","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-28DOI: 10.1016/j.ensm.2026.104933
Qirui Wang, Jie Gao, Lei Mao, Yan Lyu
{"title":"Magnetic field sensing of inhomogeneous degradation in Lithium-ion batteries with spatio-temporal evolution","authors":"Qirui Wang, Jie Gao, Lei Mao, Yan Lyu","doi":"10.1016/j.ensm.2026.104933","DOIUrl":"https://doi.org/10.1016/j.ensm.2026.104933","url":null,"abstract":"","PeriodicalId":306,"journal":{"name":"Energy Storage Materials","volume":"31 1","pages":""},"PeriodicalIF":20.4,"publicationDate":"2026-01-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146072528","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-28DOI: 10.1016/j.ensm.2026.104938
Changhoon Kim, Jae-Seung Kim, Juhyoun Park, Jun Pyo Son, Jaehan Park, Seokjae Hong, Youngsu Lee, Kyu-Young Park, Hyungsub Kim, Dong-Hwa Seo, Yoon Seok Jung
{"title":"Vacancy-Induced Li+ Conduction of Li3–xAl1–xTixF6 Fluoride Solid Electrolyte for 5 V All-Solid-State Batteries","authors":"Changhoon Kim, Jae-Seung Kim, Juhyoun Park, Jun Pyo Son, Jaehan Park, Seokjae Hong, Youngsu Lee, Kyu-Young Park, Hyungsub Kim, Dong-Hwa Seo, Yoon Seok Jung","doi":"10.1016/j.ensm.2026.104938","DOIUrl":"https://doi.org/10.1016/j.ensm.2026.104938","url":null,"abstract":"","PeriodicalId":306,"journal":{"name":"Energy Storage Materials","volume":"450 1","pages":""},"PeriodicalIF":20.4,"publicationDate":"2026-01-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146072530","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：