Global energy systems are undergoing a profound transformation, shifting decisively toward sustainable and low-carbon pathways. The transportation sector, a major contributor to global greenhouse gas emissions, is at the heart of this transition, with electric vehicles (EVs) emerging as the definitive solution. At the core of every electric vehicle battery system lies a fundamental challenge: thermal management. The lithium-ion batteries that power these vehicles are highly sensitive to temperature. They demand an optimal operating window—typically between 25°C and 40°C—to maximize performance, ensure safety, and prolong lifespan. Operating outside this window accelerates degradation and, in extreme cases, leads to catastrophic thermal runaway. However, the very systems designed to protect the battery, such as liquid cooling and air conditioning, consume significant energy, directly diminishing the vehicle’s driving range. In my research, I have addressed this critical engineering conflict by focusing on the co-optimization of temperature control and energy consumption for electric vehicle battery thermal management systems (BTMS). My work establishes a comprehensive framework that integrates theoretical modeling, experimental validation, statistical parameter screening, and advanced intelligent control strategies to achieve optimal battery thermal behavior with minimal energy penalty.
The significance of this research is underscored by the operational realities of modern EVs. Under high ambient temperatures, aggressive driving, or fast charging, the heat generated by an electric vehicle battery pack can be substantial, raising its temperature far beyond the safety threshold. Conversely, in cold climates, battery performance and charge acceptance plummet, requiring active heating to restore functionality. Historically, thermal management systems have relied on simple rule-based threshold control, which often leads to significant temperature overshoots, excessive energy use, and increased component wear, thereby limiting the vehicle’s effective range and safety margins. Therefore, a paradigm shift is required—moving away from reactive, energy-intensive thermal management toward predictively optimized, energy-aware control. In this context, my dissertation delves into the synergistic control of temperature and energy consumption, proposing a hierarchical, multi-mode control architecture tailored to different environmental conditions.
In the subsequent sections, I will systematically guide you through my research journey. I will begin by establishing a rigorous theoretical foundation, detailing the electro-thermal coupled modeling of the battery and the vehicle dynamics. Following this, I will describe the experimental platforms and procedures used to calibrate and validate my simulation models. A key aspect of my work involves the application of design of experiments (DoE) and analysis of variance (ANOVA) to isolate the most critical control factors affecting system performance. Finally, I will present the design and evaluation of my advanced control strategies, contrasting the outcomes achieved with conventional methods. My goal is to demonstrate a holistic pathway toward designing BTMS that not only ensure battery safety but also significantly extend the driving range and energy efficiency of future electric vehicles.
Theoretical Modeling of the Electric Vehicle Battery Thermal Management System
To develop effective control strategies, it is imperative to first establish a precise mathematical representation of the system’s underlying physics. My modeling efforts were structured into two primary domains: the electro-thermal coupled behavior of the electric vehicle battery, and the integrated vehicle powertrain dynamics that dictate the battery’s operating load. Understanding the heat generation and dissipation mechanisms within the cell is paramount, as it directly defines the boundary conditions for the thermal management system.
In my analysis of the electric vehicle battery’s thermal behavior, I systematically decomposed the total heat generation into its constituent physical components. These components include reaction heat, polarization heat, Joule heat, and side-reaction heat. The total heat generation \( Q \) can be formally expressed as the sum of these contributors. The reaction heat \( Q_r \) originates from the entropic changes associated with lithium-ion intercalation and de-intercalation. The polarization heat \( Q_p \) arises from the electrochemical polarization effects at the electrodes, the Joule heat \( Q_j \) is the irreversible heat generated due to the internal electrical resistance, and the side-reaction heat \( Q_s \) typically contributes negligibly under normal operating conditions and is often omitted for model simplification.
| Mechanism | Symbol | Formula | Description |
|---|---|---|---|
| Reaction Heat | \( Q_r \) | \( \frac{n m Q_e I}{M F} \) | Heat from Li+ insertion/extraction |
| Polarization Heat | \( Q_p \) | \( I^2 R_p \) | Heat due to polarization resistance |
| Joule Heat | \( Q_j \) | \( I^2 R_e \) | Heat from internal electrical resistance |
| Side-reaction Heat | \( Q_s \) | \( \approx 0 \) | Negligible in normal operation |
To estimate the heat generation rate, I adopted the widely recognized Bernardi equation. This model, which is derived from the general energy balance of the battery system, provides a robust and computationally efficient method for calculating the instantaneous heat generation rate \( \dot{Q}_{gen} \). This model is particularly effective for system-level thermal control because it elegantly captures the combined effects of irreversible Joule heating and reversible entropic heating.
$$ \dot{Q}_{gen} = I \left( I R_e – T_b \frac{dU_{ocv}}{dT_b} \right) $$
In this equation, \( I \) represents the current (A), \( R_e \) is the equivalent internal resistance of the battery (Ω), \( T_b \) is the battery temperature (°C), and \( dU_{ocv}/dT_b \) is the entropy coefficient representing the change in open-circuit voltage with respect to temperature. The internal resistance \( R_e \) is not constant; it is a nonlinear function of temperature, SOC, and current, which I determined experimentally through HPPC testing. Similarly, the entropy coefficient is typically a function of the state of charge, especially in lithium iron phosphate cells.
The transfer of heat away from the electric vehicle battery surface occurs through three fundamental modes: conduction, convection, and radiation. In a liquid-cooled battery pack, conduction is the primary mode of heat transfer from the cell core to its surface. The convective heat transfer at the surface, primarily driven by the coolant flow, dictates the cooling rate. The heat transfer from the battery to the surrounding environment follows Fourier’s law and Newton’s law of cooling. I incorporated these mechanisms into a lumped-parameter thermal model to balance computational efficiency with acceptable accuracy for control design.
$$ Q_f = -\lambda \frac{\partial t}{\partial x