The thermal state of the power battery pack is a fundamental determinant of the safety, reliability, and longevity of electric vehicles (EVs). Among various cooling strategies, liquid cooling has become predominant for high-power EV battery packs due to its high heat transfer coefficient, compact structure, and efficient cooling capacity. However, the effectiveness of this system is highly sensitive to its operational and structural parameters. This study investigates the influence of key factors—coolant inlet temperature, inlet flow velocity, and cold plate thickness—on the thermal performance of a liquid-cooled lithium-ion battery module. Employing a combined approach of experimental validation and computational simulation, we analyze their impact on the maximum temperature and temperature uniformity within the EV battery pack. The findings provide a theoretical foundation for optimizing thermal management systems to enhance safety and prolong battery life.
Theoretical Foundation and Single Cell Modeling
The heat generation within a lithium-ion cell during operation is a complex electrochemical process. For discharge cycles, the heat generation rate can be effectively modeled using the Bernardi equation, which is crucial for accurate thermal simulation of the EV battery pack components.
The governing equation for volumetric heat generation is:
$$ q = \frac{I}{V_b} \left( IR – T_b \frac{dE_{OC}}{dT_b} \right) $$
where \( I \) is the operational current, \( R \) is the total internal resistance of the cell, \( V_b \) and \( T_b \) are the volume and instantaneous temperature of the cell, respectively, and \( dE_{OC}/dT_b \) represents the temperature coefficient of the open-circuit voltage. The internal resistance \( R \) is not constant but varies with the state of charge (SOC), as characterized by Hybrid Pulse Power Characterization (HPPC) tests.
A three-dimensional geometric model of a prismatic lithium-ion cell was constructed and discretized using a combination of hexahedral and layered meshes, particularly refined near the cooling surfaces. The material properties for the cell and other components are assigned as listed in the following table, which are essential parameters for simulating the EV battery pack thermal behavior.
| Component | Parameter | Value | Unit |
|---|---|---|---|
| Battery Cell | Nominal Voltage / Capacity | 3.7 / 210 | V / Ah |
| Density | 1646 | kg/m³ | |
| Specific Heat Capacity | 1118 | J/(kg·°C) | |
| Thermal Conductivity (X, Y, Z) | 20, 20, 2 | W/(m·K) | |
| Coolant | Dynamic Viscosity | 0.00338 | Pa·s |
| Density | 1072 | kg/m³ | |
| Specific Heat Capacity | 3201 | J/(kg·°C) | |
| Thermal Conductivity | 0.379 | W/(m·K) | |
| Cold Plate (Aluminum) | Specific Heat Capacity | 904 | J/(kg·°C) |
| Thermal Conductivity | 237 | W/(m·K) | |
| Density | 2603 | kg/m³ |
Model Validation via Single Cell Analysis
The electro-thermal coupling model for a single cell was first validated under various charge and discharge rates (0.5C, 1C, 1.5C charging; 1C, 2C, 3C discharging) at an ambient temperature of 25°C. The simulations revealed that the highest temperature consistently occurs at the core of the cell, with heat dissipating outward. The terminal tabs act as effective heat fins, aiding in heat rejection.
The critical observation is the direct correlation between the C-rate and thermal stress. For instance, during charging, the maximum cell temperature rose from 30.3°C at 0.5C to 46.1°C at 1.5C, while the temperature difference (ΔT) within the cell increased from 0.9°C to 3.9°C. Furthermore, for the same 1C rate, the discharge process generated more heat than charging, leading to a higher maximum temperature (45.7°C vs. 36.8°C) and a larger ΔT (4.0°C vs. 2.2°C). This underscores the more stringent cooling requirements during high-power discharge, a common scenario for an EV battery pack.
Experimental validation was conducted using a dedicated test bench with a DC load, a climate chamber, and surface-mounted temperature sensors. The single cell was discharged at a constant current from full charge to a cut-off voltage. A comparison between the simulated and experimentally measured maximum surface temperatures during a 2C discharge showed excellent agreement, with a maximum deviation of less than 7%. This confirms the accuracy and reliability of the single-cell model, forming a solid basis for constructing the full EV battery pack module simulation.
Investigation of Thermal Management Factors for the Battery Module
The study proceeded to analyze a battery module consisting of 5 cells connected in series. Cooling plates with dual-inlet and dual-outlet square flow channels were placed between the cells, forming a representative unit of a larger EV battery pack. The 3D model of this module was created, and simulations were performed to dissect the influence of three key factors.
1. Effect of Coolant Inlet Flow Velocity
Coolant flow velocity is a primary operational parameter controlling the convective heat transfer rate. Simulations were run with inlet velocities ranging from 0.01 m/s to 0.4 m/s, keeping the coolant and ambient temperature at 25°C and a discharge rate of 2C.
The results demonstrate a clear but nonlinear relationship. Increasing the flow velocity significantly enhances the cooling performance initially, but the marginal benefit diminishes at higher velocities. The data is summarized in the table below:
| Coolant Inlet Velocity (m/s) | Maximum Temperature (°C) | Minimum Temperature (°C) | Temperature Difference ΔT (°C) |
|---|---|---|---|
| 0.01 | 47.8 | 43.73 | 4.07 |
| 0.03 | 46.4 | 42.52 | 3.88 |
| 0.05 | 45.8 | 42.37 | 3.43 |
| 0.1 | 45.0 | 41.86 | 3.14 |
| 0.2 | 43.9 | 41.54 | 2.36 |
| 0.3 | 43.5 | 41.35 | 2.15 |
| 0.4 | 43.2 | 41.12 | 2.08 |
The trend shows a rapid decrease in both maximum temperature and ΔT as velocity increases from 0.01 to 0.2 m/s. Beyond 0.2 m/s, the reduction becomes gradual. For example, increasing velocity from 0.3 to 0.4 m/s only reduced ΔT by 0.07°C. This indicates that while higher flow improves the thermal performance of the EV battery pack, it comes with diminishing returns and increased parasitic power consumption from the coolant pump. An optimal velocity must balance cooling efficacy with system efficiency.
2. Effect of Coolant Inlet Temperature
The coolant inlet temperature directly sets the lower bound for the EV battery pack temperature. Simulations were conducted with inlet temperatures ranging from 20°C to 35°C, with a fixed flow velocity of 0.4 m/s and a 2C discharge rate.
As expected, lower coolant temperatures result in significantly better cooling. When cooled by 35°C coolant, the module’s maximum temperature exceeded 49°C. In contrast, with 20°C coolant, the maximum temperature was contained below 41°C—a difference of over 8°C. The temperature rise curves show distinct behaviors: with coolant below ambient temperature (25°C), the module temperature drops initially before rising as heat generation accumulates. With coolant at ambient temperature, the temperature rises monotonically. With coolant above ambient, the system initially heats the cells until their temperature surpasses the coolant temperature, after which cooling begins. This study confirms that actively controlling the coolant temperature is vital for maintaining the EV battery pack within its optimal operating window, especially under high load.
3. Effect of Cold Plate Thickness
The cold plate thickness influences the cross-sectional area of the flow channel (assuming constant channel height and wall thickness). Intuitively, a larger flow area might improve cooling. To investigate this for the EV battery pack, simulations were run with cold plate thicknesses varying from 6 mm to 10 mm, while maintaining a constant coolant volumetric flow rate (i.e., velocity decreases as area increases).
The results were counter-intuitive. Increasing the cold plate thickness from 6 mm to 10 mm actually led to an *increase* in the module’s maximum temperature from just over 50°C to over 53°C. Although the coolant temperature rise was smaller for the thicker plate (0.6°C vs. 1.1°C), indicating better heat absorption, the overall cooling performance degraded. This is because, under a constant flow rate condition, increasing the channel area reduces the flow velocity, which in turn reduces the convective heat transfer coefficient. The relationship can be conceptualized by the effect on the Nusselt number in laminar flow. This demonstrates that merely increasing the cold plate thickness (and thus flow area) without proportionally increasing the pump’s flow rate is ineffective and can be detrimental to the thermal management of the EV battery pack. It also leads to increased system mass and volume, highlighting the need for holistic design optimization.
Conclusion
This study systematically investigated the impact of key parameters on the thermal performance of a liquid-cooled EV battery pack module through validated simulations. The single-cell model was first established and verified with experimental data, showing a maximum error of less than 7%. Based on this, a module-level analysis was conducted, leading to the following conclusions:
- Coolant Flow Velocity: There exists a nonlinear relationship between coolant velocity and cooling performance. Increasing the inlet flow velocity initially causes a rapid decline in both the maximum temperature and temperature difference (ΔT) within the EV battery pack. However, beyond a certain point (approximately 0.2 m/s in this study), the rate of improvement diminishes significantly. This indicates an optimal range that balances cooling performance with pump energy consumption.
- Coolant Inlet Temperature: The inlet temperature of the coolant is a dominant factor. Lower coolant temperatures consistently yield lower maximum temperatures and better temperature uniformity in the EV battery pack. Actively controlling this parameter is essential for managing battery temperature under high-stress conditions like high-rate discharge.
- Cold Plate Thickness / Flow Area: Under a constant coolant volumetric flow rate, simply increasing the cold plate thickness (and thus the flow channel cross-sectional area) does not improve cooling and can even worsen it due to the consequent reduction in flow velocity and convective heat transfer. This parameter must be optimized in conjunction with the flow rate and pump capability, considering the trade-offs with system weight and volume.
In summary, effective thermal management of an EV battery pack requires a integrated approach that carefully optimizes operational parameters like coolant temperature and flow velocity alongside structural design parameters. The insights from this study provide a valuable foundation for designing more efficient, safe, and durable liquid cooling systems for electric vehicles.