Data-driven research promises to accelerate the development of novel electrochemical energy storage technologies. Atomistic- to materials-scale efforts include discovery and synthesis through autonomous “self-driving” laboratories [1,2]. Here, we present several continuum-scale efforts where a complete electrochemical couple is tested at length scales ranging from coin cell (laboratory) to complete system (field) and linked with physics-based AI algorithms to compress development time. We highlight successes and challenges transferring learning across chemistries, designs [3], and operating conditions [4], as well as parameter inference, uncertainty quantification [5,6], multi-modal and explainable AI/ML.The Battery Data Genome (BDG) [7] laid out a vision for a collection of disparate open-source and proprietary data hubs to unify around a standardized/tagged data format and interchangeable algorithms, linking multiple research groups and industry to bridge “valleys of death” in technology development. BDG also described challenge problems around performance, state of health, lifetime, safety, and acceleration of materials & manufacturing development. This talk further highlights two programs supported by U.S. Department of Energy (DOE) which seek to accelerate battery technology transition from lab to full-scale commercialization.The first program, funded by DOE’s Vehicle Technologies Office, is new data hub (batterydata.energy.gov) hosting pre-competitive battery R&D data focused on materials/electrode performance and battery design. To promote transfer learning, metadata define electrode & electrolyte recipes/chemistry, cell design, and test condition. Data can be either public [3,8] or private, the latter protected by two-factor authentication. Visualization and analysis algorithms enable researchers to share insights faster than previously possible. Compared to datahubs storing commercial data, batterydata.energy.gov is unique in that it stores pre-commercial R&D data. It thus serves as a repository for challenge problems related to battery materials research, design, and manufacturing.The second program, Rapid Operational Validation Initiative (ROVI), is funded by DOE’s Office of Electricity. The ROVI objective is to predict 15-year investment-grade performance with just one year of R&D data plus limited field data for long-duration energy storage technologies. The program is linking empirical and physics-based models with AI diagnostics/prognostics algorithms for monitoring/forecasting health/life from cell-level lab testing to system-scale real-world operation [9]. Amici et al., “A Roadmap for Transforming Research to Invent the Batteries of the Future Designed within the European Large Scale Research Initiative BATTERY 2030+,” Adv. Energy Mat., 2022, 12, 2102785.Dave, J. Mitchell, S. Burke, H. Lin, J. Whitacre, V. Viswanathan, “Autonomous optimization of nonaqueous Li-ion battery electrolytes via robotic experimentation and machine learning coupling,” Nature Communications 13, 5454 (2022).J. Weddle, S. Kim, B.R. Chen, Z. Yi, P. Gasper, A.M. Colclasure, K. Smith, K.L. Gering, T.R. Tanim, E.J. Dufek, “Battery state-of-health diagnostics during fast cycling using physics-informed deep-learning,” J. Power Sources, 585 (2023) 233582.Gasper, N. Collath, H.C. Hesse, A. Jossen, K. Smith, “Machine-learning assisted identification of accurate battery lifetime models with uncertainty,” J. Electrochem. Society 169 (2022) 080518.Hassanaly, P.J. Weddle, R.N. King, S. De, A. Doostan, C.R. Randall, E. Dufek, K. Smith, “PINN surrogate of Li-ion battery models for parametric inference. Part I: Implementation and multi-fidelity hierarchies for the single-particle model,” submitted.Hassanaly, P.J. Weddle, R.N. King, S. De, A. Doostan, C.R. Randall, E. Dufek, K. Smith, “PINN surrogate of Li-ion battery models for parametric inference. Part II: Regularization and application of the pseudo-2D model,” submitted.Ward, S. Babinec, E.J. Dufek, D.A. Howey, V. Viswanathan, M. Aykol, D.A.C. Beck, B. Blaiszik, B.-R. Chen, G. Crabtree, S. Clark, V. de Angelis, P. Dechent, M. Dubarry, E.E. Eggleton, D.P. Finegan, I. Foster, C. Gopal, P. Herring, V.W. Hu, N.H. Paulson, Y. Preger, D.U. Sauer, K. Smith, S. Snyder, S. Sripad, T.R. Tanim, L. Tao, “Principles of the Battery Data Genome,” Joule 6 (2022) 2253.Smith, A. Saxon, M. Keyser, B. Lundstrom, Z. Cao, A. Roc, “Life Prediction Model for Grid-Connected Li-ion Battery Energy Storage System,” American Control Conference, Seattle, WA, May 30-June 1, 2017.D. Guittet, P. Gasper, M. Shirk, M. Mitchell, M. Gilleran, E. Bonnema, K. Smith, P. Mishra, M. Mann, “Levelized cost of charging of extreme fast charging with stationary LMO/LTO batteries,” J. Energy Storage, 82 (2024) 110568.
Read full abstract