Jalal Jahanpanah

The accelerating transition toward electric mobility demands batteries capable of fast, reliable, and safe charging. However, current lithium-ion systems encounter major challenges under high-rate charging: uneven temperature distribution, lithium plating, and accelerated aging, which are further intensified under extreme ambient conditions such as cold starts or high-temperature operation. These issues restrict charging speed and compromise both battery life and safety. At present, thermal management systems (BTMS) are typically designed with limited adaptability and weak integration with electrochemical dynamics, preventing real-time mitigation of degradation risks. There is thus a clear need for intelligent, integrated thermal management frameworks that ensure safe fast charging and sustained performance across a wide range of use cases.

This PhD develops an Integrated Active Battery Thermal Management System that combines electro-thermal modelling, advanced thermal control strategies, and experimental validation to enable safe and efficient fast charging of lithium-ion batteries. The work begins by constructing a physics-based electro-thermal framework grounded in the P2D electrochemical model and reduced-order thermal representations. This framework is used to characterize internal heat generation and transport phenomena during high C-rate charging. To capture the complex interactions between temperature gradients and plating dynamics, the modelling is extended across multiple scales—from single cell to module and pack levels—using computational fluid dynamics (CFD) simulations for realistic heat propagation assessment.

Building on these insights, the study designs and optimizes thermally modulated charging protocols (TMCP) that dynamically adjust current and thermal conditions to minimize plating and degradation risks. The advanced BTMS concept further integrates hybrid cooling architectures, including air-liquid and immersion systems, to achieve efficient temperature regulation under transient conditions. Experimental work focuses on cell-level modelling and validation using li ion battery cells tested with battery cyclers, electrochemical impedance spectroscopy (EIS) modules, and a thermal chamber to replicate realistic cold-climate operating conditions. These experiments provide essential data for calibrating and validating the electro-thermal and degradation models. In addition, the approach explores the potential of hybrid data-driven methods, such as physics-informed neural networks (PINNs), to enhance parameter identification and improve real-time predictive capability. Altogether, the research aims to establish a robust, physics-informed framework that unifies simulation, experimentation, and control for next-generation battery thermal management.

The proposed advanced BTMS is expected to provide more stable temperature control, reduced degradation during fast charging, and enhanced operational safety through proactive thermal and electrochemical management. By coupling physical modelling with adaptive control, the system can improve the efficiency, durability, and safety of EV batteries while reducing the reliance on extensive physical testing. Moreover, the research will produce a validated digital-twin foundation that supports future integration of advanced diagnostic and control strategies. These outcomes will contribute to safer and more climate-resilient electrification, particularly for heavy-duty and off-road applications relevant to Finland’s Intelligent Work Machines domain.

Existing BTMS solutions are typically limited to thermal regulation and do not actively consider electrochemical phenomena such as lithium plating or degradation during charging. Recent developments in digital battery management focus either on cell-level modelling or system-level control, but their interaction remains loosely coupled. Projects like THOR, DigiBatt, and BASE have advanced digitalization and test acceleration but have not addressed fully integrated electro-thermal-control frameworks. This PhD distinguishes itself by pursuing a holistic, predictive approach that connects modelling, simulation, and experimental testing in a unified system. The resulting framework aims to offer stronger adaptability, physical interpretability, and validation capacity, positioning it beyond the capabilities of conventional BTMS designs.