Multiscale Thermal Management of Electric Vehicles, Battery and Charging Systems: Experimental Validation and Computational Modeling Spearheading Innovation in Electric Vehicles and Battery Systems
ATOMS Laboratory implements a novel experimental-computational multiscale approach for thermal-electrical-chemical modeling and battery development activities integrating experiments and computational simulations across multiple physical domains and length scales from the electrode level to the vehicle level
Motivation
- Lithium-ion battery (LIB) packs are intricate, involving various components and processes
- Exhibit phenomena across different scales, from atomic interactions to macroscopic behavior
- Interconnected physical processes such as electrochemistry and thermal phenomena play crucial roles
- Thermal management, a key to improve performance, safety and longevity, mitigating degradation and ensuring safe operation
Research Activities
- Hierarchical multiscale modeling for battery thermal management optimization
- Characterizing innovative electrochemical-thermal Li-ion systems
- Estimating thermophysical properties and heat generation rates
- Predicting thermal performance across LIB systems
- Isolating heat generation rates distribution through Electrochemical Impedance Spectroscopy (EIS)
Motivation
- Need for intelligent solutions for vehicle cooling and heating challenges despite battery thermal management system (BTMS) advances
- Long computational times and limited parameter access issues in traditional BTMS optimization methods
- ATOMS lab pioneers advanced thermal simulation frameworks for BTMS optimization
- Innovative approach integrating deep convolutional neural network for optimal design predictions
Research Activities
- Hierarchical thermal modeling considering spatio-temporal battery heat generation effects
- Integrating novel deep convolutional encoder-decoder hierarchical (DeepEDH) neural network for surrogate-model based design optimization
- Investigating thermally optimal liquid-cooled cold plates for battery thermal management systems
- Predicting effectively cold plates’ temperature, pressure, and velocity fields through DeepEDH neural network surrogate model
Motivation
- Thermal runaway (TR) and its propagation between cells pose major safety risks for EV and grid battery energy storage systems (BESS).
- Many existing TR models require input parameters that are difficult to obtain or not readily available, limiting their practical use.
- Practical predictive tools are needed to inform cell and system design with respect to TR propagation.
Research Activities
- Develop physics-based models to predict TR propagation in large-format battery modules and BESS.
- Quantify spatial and temporal TR dynamics and assess the influence of cell and system configuration.
- Combine transient heat transfer modelling with targeted TR experiments for validation and physical insight.
Motivation
- Widespread adoption of LIBs require overcoming critical technological constraints impacting battery aging and safety
- Battery aging results in irreversible capacity & performance losses
- Increase EV battery performance reliability to mitigate battery aging
Research Activities
- Developing physics-based calendar and cycle aging LIB models to simultaneously predict state of health, loss of lithium inventory and loss of active material for negative/positive electrodes
- Predicting battery health while simultaneously predicting aging root causes
- Providing solution-based approach for modelling to capture thermo-electric gradients within pouch cells and effects on accelerating battery aging
- Leveraging industry partnerships for manufacturing novel pouch cells and ATOMS thermal management systems (TMS) laboratory for model validation
Motivation
- High-power EV and grid BESS require accurate thermal management to ensure safety, performance, and lifetime.
- Large-format Li-ion cells exhibit spatially non-uniform heat generation, particularly under fast charging and high loads.
- These spatial thermal effects remain inadequately characterized, limiting practical modelling and next-generation thermal management design.
Research Activities
- Developing physics-based electro-thermal models for large-format lithium-ion cells.
- Quantifying heat generation and spatial temperature distributions across operating conditions and state of charge.
- Validating models using infrared thermography and applying them to thermal management optimization and digital-twin development.
Motivation
- Real-time state of charge, state of health, and remaining useful life estimation for safe and reliable battery operation.
- Battery aging and changing operating conditions cause model drift and reduce the accuracy of fixed parameter models.
- Lack of fast, high-fidelity models for real-time thermal prediction and feedback control under dynamic loads and charging.
- Need for closed-loop integration of live measurements with a digital model for degradation prediction, early anomaly detection, and lifetime extension.
Research Activities
- Building real-time data pipelines from experimental systems to a live database with historical data logging.
- Developing live dashboards for online monitoring of experimental sensor measurements.
- Developing data-driven and physics-based models compatible with real-time estimation and online updating.
- Predicting battery temperature and performance to support operational decisions using a digital twin framework.