Machine Learning‐Based Lifetime Prediction of Lithium‐Ion Cells

Abstract

Precise lifetime predictions for lithium‐ion cells are crucial for efficient battery development and thus enable profitable electric vehicles and a sustainable transformation towards zero‐emission mobility. However, limitations remain due to the complex degradation of lithium‐ion cells, strongly influenced by cell design as well as operating and storage conditions. To overcome them, a machine learning framework is developed based on symbolic regression via genetic programming. This evolutionary algorithm is capable of inferring physically interpretable models from cell aging data without requiring domain knowledge. This novel approach is compared against established approaches in case studies, which represent common tasks of lifetime prediction based on cycle and calendar aging data of 104 automotive lithium‐ion pouch‐cells. On average, predictive accuracy for extrapolations over storage time and energy throughput is increased by 38% and 13%, respectively. For predictions over other stress factors, error reductions of up to 77% are achieved. Furthermore, the evolutionary generated aging models meet requirements regarding applicability, generalizability, and interpretability. This highlights the potential of evolutionary algorithms to enhance cell aging predictions as well as insights.

Computational Simulation for Breakdown and Energy Storage Performances with Optimization in Polymer Dielectrics

“…The machine learning approach, as the core tool of the datadriven materials research approach, the 4 th research paradigm of materials science, has emerged rapidly in recent years. [216][217][218][219] Combining the machine learning approach with highthroughput computation enables more efficient and intelligent screening and development of high-energy storage dielectrics, as well as a deeper understanding of the intrinsic structure-activity relationships between microstructure and macro-properties of polymer dielectrics. The machine learning approaches and highthroughput computations can provide each other with the necessary data.…”

Building Smarter Aqueous Batteries

“…[121,122] On one hand, machine learning/deep learning methods (combined with simulation [152] or advanced characterizations [125,153,154] ) can assist the investigation of electrode structure evolution, such as ion plating/dendrite growth [152,155] and crack formation. [156] On the other hand, these AI methods can also promote battery states and performance prediction, including capacity, [157,158] lifetime, [124,159] and cycling protocols. [160] Moreover, they can also accelerate the modeling of materials/batteries [117,122] and the decoding of degradation modes.…”

Critical Contribution of Imbalanced Charge Loss to Performance Deterioration of Si-Based Lithium-Ion Cells during Calendar Aging

“…The needed long-term RPT-OCV testing has been the major barrier to increasing the turnaround rate of the technology exploration/optimization. There is also a recent effort in deploying artificial intelligence (AI) and machine learning (ML) algorithms to assist in data modeling and predicting, which nonetheless rely on the availability of extensive data sets that implicitly cover all possible underlying reaction mechanisms to train the mathematical model. − Capturing the true performance loss mechanisms constitutes the foundation for reliable prediction for AI/ML models.…”