Deep model predicts organic battery performance over many charge cycles
HorizON: A Multi-Horizon Deep Operator Network for Property Prediction of Organic Battery Stacks
Machine Learning
Summary
Batteries that use metals like lithium and cobalt have environmental and supply problems. The authors study batteries made from organic materials, which are more tunable but less stable. They created a deep learning model that predicts how these batteries perform not just once, but over many cycles of charging and discharging. This helps understand and optimize battery systems by considering the whole battery setup, not just one part. Their model can also help design better organic batteries and improve battery manufacturing.
What this means in practice
- •For battery design engineers: Predict how organic battery stacks perform across full discharge cycles to improve design decisions for new battery systems.
- •For battery manufacturing teams: Optimize production processes by forecasting long-term battery properties based on complete stack representations and use conditions.
Authors
Subhash V. S. Ganti, Lukas Woelfel, Arla Diengdoh, Tzvetan Mihaylov, Thomas Nevolianis, Jan Meyer, Christopher Kuenneth
Abstract
Transition-metal-based batteries rely on critical elements such as lithium, cobalt, and nickel. Their extraction creates environmental and supply-chain challenges. Organic electrode materials (OEMs) reduce this reliance and provide broad molecular tunability. However, OEM-based batteries still show limited long-term cycling stability. Their design also requires the joint optimization of coupled metrics such as voltage, specific capacity, capacity retention, and energy density. These metrics depend on the full battery stack and on the interactions among its components rather than on the active material alone. In our previous work, we developed a data-fusion multi-task framework for average discharge voltage and first-cycle specific capacity. Both targets are single-valued. Here, we extend that framework to multi-horizon prediction by predicting a property at every cycle of a discharge trajectory rather than at a single cycle. Our multi-fidelity and Multi-horizon Deep Operator Network learns DFT redox potentials and cycle-dependent specific-capacity trajectories within one model. We train it on complete battery-stack representations and usage conditions, so it learns battery-stack interactions rather than cathode-only effects. This framework supports long-term property prediction, inverse design, and process optimization for organic battery systems.