[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"1bf1j0q":3,"1crdqpn":67},{"common":4,"navPages":22,"productGroups":30,"seo":63,"strategies":66},{"_id":5,"_type":5,"cta_label":6,"cta_label_impact":7,"email":8,"socials":9},"settings_common","let’s talk","read impact studies",null,[10,14,18],{"_key":11,"label":12,"type":12,"url":13},"668f9a890fd5","linkedin","https:\u002F\u002Fwww.linkedin.com\u002Fcompany\u002Fsphere-energy\u002F",{"_key":15,"label":16,"type":16,"url":17},"5a142892073b","medium","https:\u002F\u002Fmedium.com\u002F@lutz.lukas88",{"_key":19,"label":20,"type":20,"url":21},"f97a157c8bf7","youtube","https:\u002F\u002Fwww.youtube.com\u002F@SphereEnergy",[23,27],{"_id":24,"slug":25,"title":26},"8b8dc55d-3fef-48a3-8152-48bcedea3bf6","studies","impact studies",{"_id":28,"slug":29,"title":29},"15dbbb59-c8d8-46d2-bd7c-722ffe763ccf","about",[31,47],{"_id":32,"products":33,"title":46},"865c3cc9-1bd1-4eb7-9b45-997456c72b35",[34,38,42],{"_id":35,"slug":36,"title":37},"c02a84c3-1cd4-4ccf-881d-ec8ebd92a8b5","battery-life-prediction","Battery life prediction",{"_id":39,"slug":40,"title":41},"ff6446f7-e857-4ac5-9361-ef70a82e4e92","soh-estimation","SOH Estimation",{"_id":43,"slug":44,"title":45},"c1e2a74c-c4e9-4b06-80cb-19c7a9a12867","parametrisation","Parametrisation","Battery AI",{"_id":48,"products":49,"title":62},"f62d95db-3849-497c-aced-e4224dc82287",[50,54,58],{"_id":51,"slug":52,"title":53},"26365138-af07-43a8-b1f5-8b8539034357","requirements-processing","Requirements processing",{"_id":55,"slug":56,"title":57},"8a92b84d-7114-43dd-97fe-7965c6bfcaac","cad-intelligence","CAD Intelligence",{"_id":59,"slug":60,"title":61},"d5321be1-6a2e-4414-8196-88e25567ce30","crash-simulation","Crash Simulation","Engineering AI",{"description":8,"og_description":8,"og_image":64,"og_title":8,"title":65},"https:\u002F\u002Fcdn.sanity.io\u002Fimages\u002Fowk14ohd\u002Fproduction\u002Fa750d9adc1a676399bfcb13d2eee377f5583c9de-1200x630.jpg","Sphere Intelligence",[],{"_id":68,"_type":69,"content":70,"customerProfile":324,"groups":325,"image":333,"plainTitle":337,"relatedStudies":338,"seo":8,"slug":432,"subtitle":433,"title":434},"eb74791e-a917-4c32-ba1b-65005c773e3d","study",[71,82,91,99,107,124,132,140,148,156,165,172,180,188,196,204,212,220,227,235,243,251,263,274,285,296,304],{"_key":72,"_type":73,"children":74,"markDefs":80,"style":81},"fbeedf0a074f","block",[75],{"_key":76,"_type":77,"marks":78,"text":79},"8cf912910f2e","span",[],"The situation",[],"h2",{"_key":83,"_type":73,"children":84,"markDefs":89,"style":90},"e19f1e2d7cb9",[85],{"_key":86,"_type":77,"marks":87,"text":88},"2cc98c637e2c",[],"Validating battery cells through physical testing is extremely time- and resource-intensive. A standard end-of-life (EoL) validation process for a single cell can take months, or even years. The client needed to make better use of historical data and reduce the number of physical Design-of-Experiments (DOE) test series required.",[],"normal",{"_key":92,"_type":73,"children":93,"markDefs":98,"style":90},"607221f00238",[94],{"_key":95,"_type":77,"marks":96,"text":97},"eda2c1959119",[],"",[],{"_key":100,"_type":73,"children":101,"markDefs":106,"style":81},"69807491f680",[102],{"_key":103,"_type":77,"marks":104,"text":105},"37ab6f9a2254",[],"The challenge",[],{"_key":108,"_type":109,"items":110},"c396838dae5f","cardsSmall",[111,116,120],{"_key":112,"_type":113,"description":114,"title":115},"d3d7a2c82a24","card","from only limited early-cycle input data (the first 25–75 cycles)","Predict a cell's aging behavior through end-of-life",{"_key":117,"_type":113,"description":118,"title":119},"dab392824451","temperature, C-rates, and different SOC (State of Charge) windows","Account for multiple influencing factors —",{"_key":121,"_type":113,"description":122,"title":123},"84d428d215a2","to new cell types (transfer  learning) for which little to no data exists","Test the model's transferability",{"_key":125,"_type":73,"children":126,"markDefs":131,"style":81},"d5240e8821b2",[127],{"_key":128,"_type":77,"marks":129,"text":130},"649d16eb10d6",[],"How [vetta] changes the process",[],{"_key":133,"_type":73,"children":134,"markDefs":139,"style":90},"daaf07f4a7f5",[135],{"_key":136,"_type":77,"marks":137,"text":138},"1dc556053a92",[],"vetta combines historical tests, simulations and new validation data into a single battery intelligence model. Weeks of testing become years of insight.",[],{"_key":141,"_type":142,"image":143},"3359ef61ba46","imageAlt",{"_type":144,"asset":145},"image",{"_ref":146,"_type":147},"image-549ec7488b9dd5246bcd3102b4a32efef87cd25c-1362x766-png","reference",{"_key":149,"_type":73,"children":150,"markDefs":155,"style":81},"99ece390a500",[151],{"_key":152,"_type":77,"marks":153,"text":154},"ee3f293d3ffa",[],"The approach",[],{"_key":157,"_type":73,"children":158,"markDefs":163,"style":164},"ffce41e19c1f",[159],{"_key":160,"_type":77,"marks":161,"text":162},"87f636119b62",[],"A Foundation Model for Battery Aging — Tuned to the Client's Cells",[],"h3",{"_key":166,"_type":73,"children":167,"markDefs":171,"style":90},"db073d08293b",[168],{"_key":160,"_type":77,"marks":169,"text":170},[],"Sphere used VETTA, its time-series foundation model pre-trained on 10+ million synthetic datasets and thousands of real cell tests. The approach comprised:",[],{"_key":173,"_type":73,"children":174,"markDefs":179,"style":164},"f3ff02014d98",[175],{"_key":176,"_type":77,"marks":177,"text":178},"9a8c63369e99",[],"Model Fine-Tuning",[],{"_key":181,"_type":73,"children":182,"markDefs":187,"style":90},"0d8942e83d9e",[183],{"_key":184,"_type":77,"marks":185,"text":186},"730a3402aa5b",[],"The pre-trained model was refined using the client's specific cell data.",[],{"_key":189,"_type":73,"children":190,"markDefs":195,"style":164},"16a78b869cfd",[191],{"_key":192,"_type":77,"marks":193,"text":194},"5c0b0ca3eb44",[],"Hybrid Simulation",[],{"_key":197,"_type":73,"children":198,"markDefs":203,"style":90},"fa73220288aa",[199],{"_key":200,"_type":77,"marks":201,"text":202},"6fbc955f8e08",[],"A combination of data-driven AI and physics-based modeling (P2D models) to not only predict but physically explain varying degradation mechanisms, such as SEI growth, lithium plating, etc.",[],{"_key":205,"_type":73,"children":206,"markDefs":211,"style":164},"b9a5c2ccea4c",[207],{"_key":208,"_type":77,"marks":209,"text":210},"2501fe1167fa",[],"Defined Use Cases",[],{"_key":213,"_type":73,"children":214,"markDefs":219,"style":90},"789a5ee41cf8",[215],{"_key":216,"_type":77,"marks":217,"text":218},"c4e08f751daa",[],"Three specific use cases were defined: DOE optimization for 2 cell types, and transfer learning to a new cell for further validation.",[],{"_key":221,"_type":73,"children":222,"markDefs":226,"style":164},"34e64a73dc92",[223],{"_key":216,"_type":77,"marks":224,"text":225},[],"What changes for the engineer",[],{"_key":228,"_type":73,"children":229,"markDefs":234,"style":90},"243715a13712",[230],{"_key":231,"_type":77,"marks":232,"text":233},"d5067fe9373b",[],"Instead of waiting months for a full DOE test campaign, the engineer feeds the first few weeks of cycling data into VETTA and receives an end-of-life prediction — together with the physical degradation drivers behind it — within days. This supports earlier go\u002Fno-go decisions and a smaller, more targeted physical test matrix.",[],{"_key":236,"_type":73,"children":237,"markDefs":242,"style":81},"a88f1fd558f7",[238],{"_key":239,"_type":77,"marks":240,"text":241},"356b67cfc956",[],"The results",[],{"_key":244,"_type":73,"children":245,"markDefs":250,"style":90},"36362be1e379",[246],{"_key":247,"_type":77,"marks":248,"text":249},"45dd8d14bc2c",[],"The model moved directly from proof of concept to a validated prediction tool, integrated into the client's cell validation workflow across multiple cell types and projects.",[],{"_key":252,"_type":73,"children":253,"markDefs":262,"style":90},"72123a02741d",[254,258],{"_key":247,"_type":77,"marks":255,"text":257},[256],"strong","Prediction accuracy of ~ \u003C 3% ",{"_key":259,"_type":77,"marks":260,"text":261},"cf99c972867d",[],"MAPE achieved for DOE optimization",[],{"_key":264,"_type":73,"children":265,"markDefs":273,"style":90},"c6a65af6996a",[266,269],{"_key":259,"_type":77,"marks":267,"text":268},[256],"Zero-shot transfer",{"_key":270,"_type":77,"marks":271,"text":272},"8ed43b35f1cc",[]," to new cell types achieved under 3% error, improving to under 2% MAE with just 150 cycles of data",[],{"_key":275,"_type":73,"children":276,"markDefs":284,"style":90},"fba5aa3a5d38",[277,280],{"_key":270,"_type":77,"marks":278,"text":279},[256],"Degradation mechanisms confirmed",{"_key":281,"_type":77,"marks":282,"text":283},"fc42fccfd90e",[]," — predictions aligned with known physical drivers such as SEI growth and lithium plating",[],{"_key":286,"_type":73,"children":287,"markDefs":295,"style":90},"8c198c5d0eca",[288,291],{"_key":281,"_type":77,"marks":289,"text":290},[256],"Early test termination",{"_key":292,"_type":77,"marks":293,"text":294},"1032c8fd4f1a",[]," enabled by flagging tests likely to miss requirements before they run to completion",[],{"_key":297,"_type":73,"children":298,"markDefs":303,"style":81},"0693b1cb185e",[299],{"_key":300,"_type":77,"marks":301,"text":302},"35abce072095",[],"Measurable impact",[],{"_key":305,"_type":306,"items":307},"38347bd442e8","cardsLarge",[308,312,316,320],{"_key":309,"_type":113,"description":310,"title":311},"bbcbe185f60f","Reduction in the physical DoE Matrix","20-70%",{"_key":313,"_type":113,"description":314,"title":315},"797135f2e6ab","MAPE prediction error","3%",{"_key":317,"_type":113,"description":318,"title":319},"430ac5dcb534","Transfer-learning MAE on a new cell type, with 150 cycles of input","\u003C 2%",{"_key":321,"_type":113,"description":322,"title":323},"4cb3993e4a7f","vs. months for EoL Simulation results","Days","Industry\nAutomotive Manufacturing\n\nApplications\nElectric Vehicles, battery cell validation\n\nOperations\nDOE optimization, transfer learning, multi-project\n\nDeployment\nVETTA time-series foundation model, fine-tuned, workflow-integrated",[326,329],{"_id":327,"slug":328,"title":328},"ee19de53-305b-43c5-9bce-ab5858e7041e","battery",{"_id":330,"slug":331,"title":332},"e906f1cd-3b86-4740-b3c3-8ab8e2f3d639","cell-validation","cell validation",{"_type":142,"image":334},{"_type":144,"asset":335},{"_ref":336,"_type":147},"image-9d1208754b757a57a06f67d3c2408175f2f6a99c-1308x735-png","From Months to Days:\n\nAI-Predicted Battery Cell End-of-Life",[339,377,409],{"_id":340,"card_subtitle":341,"groups":342,"image":348,"plainTitle":352,"slug":353,"title":354},"d54a0a66-0baf-4723-a45f-1b8a4f92c56a","Global battery manufacturer ",[343,347],{"_id":344,"slug":345,"title":346},"268e1d85-c92c-4117-b9fb-accfa6b0839d","data-governance","data governance",{"_id":327,"slug":328,"title":328},{"_type":142,"image":349},{"_type":144,"asset":350},{"_ref":351,"_type":147},"image-f6748766b6f73f6d134f457c12c352ab5a19b961-686x406-jpg","From 18 Months to Weeks:\n\nAI-Powered Battery Development for\n\na Global Cell Manufacturer","from-18-months-to-weeks-ai-powered-battery-development-for-a-global-cell-manufacturer",[355,363,370],{"_key":356,"_type":73,"children":357,"markDefs":362,"style":90},"684ec6bbd161",[358],{"_key":359,"_type":77,"marks":360,"text":361},"713df5652b04",[],"From 18 Months to Weeks:",[],{"_key":364,"_type":73,"children":365,"markDefs":369,"style":90},"36a3e4fc646a",[366],{"_key":359,"_type":77,"marks":367,"text":368},[],"AI-Powered Battery Development for",[],{"_key":371,"_type":73,"children":372,"markDefs":376,"style":90},"ebe119251d4e",[373],{"_key":359,"_type":77,"marks":374,"text":375},[],"a Global Cell Manufacturer",[],{"_id":378,"card_subtitle":379,"groups":380,"image":386,"plainTitle":390,"slug":391,"title":392},"509b1daf-911d-41e4-a5b7-8b9c305d2bcf","Leading Global Manufacturer",[381,385],{"_id":382,"slug":383,"title":384},"1318f5c3-ccea-4039-94a3-fdd021b4cb3f","generative-engineering","generative engineering",{"_id":327,"slug":328,"title":328},{"_type":142,"image":387},{"_type":144,"asset":388},{"_ref":389,"_type":147},"image-7b51797c365977a45066e5ca1e11f33c1201f8aa-872x490-png","From Months to Minutes:\n\nAn AI-Driven Battery Pack Design Assistant","from-months-to-minutes-an-ai-driven-battery-pack-design-assistant",[393,401],{"_key":394,"_type":73,"children":395,"markDefs":400,"style":90},"f483a97471b8",[396],{"_key":397,"_type":77,"marks":398,"text":399},"0a50a59a985d",[],"From Months to Minutes:",[],{"_key":402,"_type":73,"children":403,"markDefs":408,"style":90},"7c8774d02ee0",[404],{"_key":405,"_type":77,"marks":406,"text":407},"69da32a62284",[],"An AI-Driven Battery Pack Design Assistant",[],{"_id":410,"card_subtitle":411,"groups":412,"image":418,"plainTitle":422,"slug":423,"title":424},"a989e167-9d23-4283-984c-92d3404cefa3","Independent power producer",[413,414],{"_id":327,"slug":328,"title":328},{"_id":415,"slug":416,"title":417},"a1267711-5343-4aac-83ce-2b604e2ea049","energy-storage","energy storage",{"_type":142,"image":419},{"_type":144,"asset":420},{"_ref":421,"_type":147},"image-9913de0ee6feb255b1e868f9bbcef677be5dca86-686x406-jpg","AI-Powered SOC and SOH Monitoring for Utility-Scale BESS Fleets","ai-powered-soc-and-soh-monitoring-for-a-utility-scale-bess-fleet",[425],{"_key":426,"_type":73,"children":427,"markDefs":431,"style":90},"9347ed56f8d2",[428],{"_key":429,"_type":77,"marks":430,"text":422},"ab8f366af0d2",[],[],"from-months-to-days-ai-predicted-battery-cell-end-of-life","Leading Global Automotive OEM",[435,443],{"_key":436,"_type":73,"children":437,"markDefs":442,"style":90},"f9cd36076189",[438],{"_key":439,"_type":77,"marks":440,"text":441},"a6603a34267e",[],"From Months to Days:",[],{"_key":444,"_type":73,"children":445,"markDefs":449,"style":90},"0b7fb07e1d26",[446],{"_key":439,"_type":77,"marks":447,"text":448},[],"AI-Predicted Battery Cell End-of-Life",[]]