Physics-Enhanced Deep Learning for Proactive Thermal Runaway Forecasting in Li-Ion Batteries

• URL: http://arxiv.org/abs/2604.20175v1
• Date: Wed, 22 Apr 2026 04:34:39 GMT
• Title: Physics-Enhanced Deep Learning for Proactive Thermal Runaway Forecasting in Li-Ion Batteries
• Abstract summary: This study proposes a Physics-Informed Long Short-Term Memory (PI-LSTM) framework that integrates governing heat transfer equations directly into the deep learning architecture.<n>The proposed PI-LSTM achieves an 81.9% reduction in root mean square error (RMSE) and an 81.3% reduction in mean absolute error (MAE) compared to the standard LSTM baseline.<n>Results confirm that physics-informed deep learning offers a viable pathway toward interpretable, accurate, and real-time thermal management in next-generation battery systems.
• Score: 14.616396048555615
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Accurate prediction of thermal runaway in lithium-ion batteries is essential for ensuring the safety, efficiency, and reliability of modern energy storage systems. Conventional data-driven approaches, such as Long Short-Term Memory (LSTM) networks, can capture complex temporal dependencies but often violate thermodynamic principles, resulting in physically inconsistent predictions. Conversely, physics-based thermal models provide interpretability but are computationally expensive and difficult to parameterize for real-time applications. To bridge this gap, this study proposes a Physics-Informed Long Short-Term Memory (PI-LSTM) framework that integrates governing heat transfer equations directly into the deep learning architecture through a physics-based regularization term in the loss function. The model leverages multi-feature input sequences, including state of charge, voltage, current, mechanical stress, and surface temperature, to forecast battery temperature evolution while enforcing thermal diffusion constraints. Extensive experiments conducted on thirteen lithium-ion battery datasets demonstrate that the proposed PI-LSTM achieves an 81.9% reduction in root mean square error (RMSE) and an 81.3% reduction in mean absolute error (MAE) compared to the standard LSTM baseline, while also outperforming CNN-LSTM and multilayer perceptron (MLP) models by wide margins. The inclusion of physical constraints enhances the model's generalization across diverse operating conditions and eliminates non-physical temperature oscillations. These results confirm that physics-informed deep learning offers a viable pathway toward interpretable, accurate, and real-time thermal management in next-generation battery systems.

Related papers

• Climate Physics Dynamic Matching [46.03496190842378]
  We present Climate Physics Dynamic Matching (ClimPhyDM), a variational simulation-free dynamics informed framework for weather forecasting.<n>On the ERA5 benchmark at hourly (42-hour) and monthly (5-month) resolutions, ClimPhyDM outperforms ClimODE, and GB-DM, keeping the lower error at extended horizon.
  arXiv  Detail & Related papers  (2026-08-27T10:07:40Z)
• Virtual Temperature Sensors in Power Transformers Using Neural Ordinary Differential Equations [42.50242605797505]
  This paper develops a physics-aware Neural Ordinary Differential Equation framework for forecasting transformer thermal behavior.<n>A key contribution is the integration of simplified heat-transfer equations directly into the Neural ODE formulation.<n>The framework provides a standardized, physics-aware, and robust forecasting approach for heterogeneous transformer units.
  arXiv  Detail & Related papers  (2026-08-13T13:59:32Z)
• PiDDM: Physics-Informed Differentiable Degradation Modeling for Lithium-Ion Battery State-of-Health Prediction [2.9577202646353284]
  We develop a physics-informed differentiable degradation modeling framework for battery state of health prediction.<n> framework incorporates empirical Arrhenius degradation kinetics associated with solid electrolyte interphase growth and loss of lithium inventory into the training objective.
  arXiv  Detail & Related papers  (2026-07-31T07:19:26Z)
• AI-Driven Surrogate Models for Predicting Electrode-Scale Discharge Behavior in Lithium-Ion Batteries [47.27974914502511]
  We introduce a deep learning surrogate pipeline based on the Swin3D Transformer to predict discharge dynamics directly from data.<n>The proposed method reduces the computational overhead by orders of magnitude, providing a scalable and efficient framework for high- throughput battery design and optimization.
  arXiv  Detail & Related papers  (2026-07-22T09:10:48Z)
• Physics-Informed Neural Network with Transfer Learning for State Estimation in Lithium-Ion Batteries using the Single Particle Model with Electrolyte [0.5371337604556311]
  Physics-informed neural networks (PINNs) have emerged as a powerful tool for solving nonlinear partial differential equations.<n> PINNs offer improved scalability, particularly for reduced-order models like the single particle model with electrolyte (SPMe)
  arXiv  Detail & Related papers  (2026-06-26T16:11:00Z)
• Fast Bayesian equipment condition monitoring via simulation based inference: applications to heat exchanger health [39.61652266573024]
  We propose an AI-driven framework powered by amortized neural posterior estimation to diagnose complex failure modes in heat exchangers.<n>We benchmark this framework against an MCMC baseline across various synthetic fouling and leakage scenarios.<n>The amortized nature of the neural network enables near-instantaneous inference, establishing SBI as a highly scalable, real-time alternative.
  arXiv  Detail & Related papers  (2026-04-22T16:21:25Z)
• Physics-Informed Machine Learning for Pouch Cell Temperature Estimation [5.435523531334183]
  This paper presents a physics-informed machine learning (PIML) framework for the efficient estimation of steady-state temperature profiles.<n>The PIML approach integrates the governing heat transfer equations directly into the neural network's loss function, enabling high-fidelity predictions.<n>Results demonstrate that the PIML model converges more rapidly and achieves markedly higher accuracy, with a 49.1% reduction in mean squared error over the data-driven model.
  arXiv  Detail & Related papers  (2026-04-16T02:58:10Z)
• Real-time distortion prediction in metallic additive manufacturing via a physics-informed neural operator approach [3.607834195988809]
  This paper proposes a physics-informed smart Neuralhorizon Operator (PINO) to predict z and y-direction for the future 15 s.<n>The performance of PINO model highlights its potential for real-time long-time distortion field prediction in controlling defects.
  arXiv  Detail & Related papers  (2025-11-17T09:37:04Z)
• Temperature-Aware Recurrent Neural Operator for Temperature-Dependent Anisotropic Plasticity in HCP Materials [1.2374932078540024]
  We introduce the Temperature-Aware Recurrent Neural Operator (TRNO), a time-resolution-independent neural architecture.<n>The TRNO achieves high predictive accuracy and generalizes effectively across diverse loading cases, temperatures, and time resolutions.<n>It also outperforms conventional GRU and LSTM models in training efficiency and predictive performance.
  arXiv  Detail & Related papers  (2025-08-26T08:40:42Z)
• Modeling Membrane Degradation in PEM Electrolyzers with Physics-Informed Neural Networks [45.32169712547367]
  Proton exchange membrane (PEM) electrolyzers are pivotal for sustainable hydrogen production.<n>Their long-term performance is hindered by membrane degradation, which poses reliability and safety challenges.<n>Traditional physics-based models have been developed, offering interpretability but requiring numerous parameters that are often difficult to measure and calibrate.<n>This study presents the first application of Physics-Informed Neural Networks (PINNs) to model membrane degradation in PEM electrolyzers.
  arXiv  Detail & Related papers  (2025-06-19T15:46:49Z)
• NeuralOM: Neural Ocean Model for Subseasonal-to-Seasonal Simulation [41.41450298461784]
  We propose NeuralOM, a general neural operator framework for simulating complex, slow-changing dynamics.<n>We validate NeuralOM on the challenging task of global Subseasonal-to-Seasonal (S2S) ocean simulation.<n>NeuralOM achieves a 13.3% lower RMSE compared to the best-performing baseline.
  arXiv  Detail & Related papers  (2025-05-27T10:54:40Z)
• Spatiotemporal Field Generation Based on Hybrid Mamba-Transformer with Physics-informed Fine-tuning [46.67399627400437]
  This research confronts the challenge of substantial physical equation discrepancies in the generation of physical fields through trained models.<n>A physical field generation model, named HMT-PF, is developed based on the hybrid Mamba-Transformer architecture.
  arXiv  Detail & Related papers  (2025-05-16T14:40:56Z)
• Physics-Informed Regularization for Domain-Agnostic Dynamical System Modeling [41.82469276824927]
  We present a framework that achieves high-precision modeling for a wide range of dynamical systems. It helps preserve energies for conservative systems while serving as a strong inductive bias for non-conservative, reversible systems. By integrating the TRS loss within neural ordinary differential equation models, the proposed model TREAT demonstrates superior performance on diverse physical systems.
  arXiv  Detail & Related papers  (2024-10-08T21:04:01Z)
• A Neural PDE Solver with Temporal Stencil Modeling [44.97241931708181]
  Recent Machine Learning (ML) models have shown new promises in capturing important dynamics in high-resolution signals. This study shows that significant information is often lost in the low-resolution down-sampled features. We propose a new approach, which combines the strengths of advanced time-series sequence modeling and state-of-the-art neural PDE solvers.
  arXiv  Detail & Related papers  (2023-02-16T06:13:01Z)
• Thermal Neural Networks: Lumped-Parameter Thermal Modeling With State-Space Machine Learning [0.0]
  Thermal models for electric power systems are required to be both, real-time capable and of high estimation accuracy. In this work, the thermal neural network (TNN) is introduced, which unifies both, consolidated knowledge in the form of heat-transfer-based lumped- parameter models. A TNN has physically interpretable states through its state-space representation, is end-to-end trainable, and requires no material, geometry, nor expert knowledge for its design.
  arXiv  Detail & Related papers  (2021-03-30T13:15:48Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.