Effective thermal management is paramount for the safety, longevity, and performance of lithium-ion energy storage systems. The performance of an energy storage cell is highly sensitive to its operating temperature. Maintaining the temperature within an optimal range, typically 25–45 °C, is critical. Inadequate heat dissipation can lead to localized hot spots, accelerated aging, and in severe cases, thermal runaway, posing significant safety risks. Therefore, developing efficient cooling strategies is a cornerstone of reliable energy storage battery module design. Among various cooling methods, liquid cooling has gained prominence due to its superior heat transfer coefficient and compactness compared to air cooling. This work focuses on the thermal design and optimization of a liquid-cooled module comprising 52 individual energy storage cells. We establish and validate a computational fluid dynamics (CFD) model to analyze the thermal behavior. The investigation systematically evaluates the impact of coolant inlet/outlet diameter and, more significantly, explores novel liquid-cooling plate configurations—specifically side-plate cooling and a combined bottom-side-plate structure—against a traditional bottom-cooling baseline. The primary objectives are to enhance heat dissipation efficiency, improve temperature uniformity across the energy storage cell pack, and identify optimal operational strategies that minimize system energy consumption.
1. Model Establishment
1.1 Physical Model
The core object of this study is a commercial-scale energy storage battery module. The module enclosure has dimensions of 1160 mm (L) × 810 mm (W) × 245 mm (H) and houses 52 prismatic Lithium Iron Phosphate (LiFePO₄) energy storage cells connected in series, each with a nominal capacity of 280 Ah. To facilitate the numerical simulation while preserving essential thermal characteristics, the internal complex structure of the energy storage cell is homogenized. The cell is treated as a uniform heat generation body with anisotropic thermal conductivity. Cells are electrically connected via aluminum busbars and separated by foam pads to minimize thermal radiation exchange between adjacent cells.
Three distinct liquid-cooling structures are proposed and compared, as conceptually illustrated below:
- Bottom-Plate Cooling (Baseline): A traditional design where a liquid cold plate is attached to the bottom surface of the energy storage cell assembly.
- Side-Plate Cooling: A novel configuration where liquid cold plates are positioned on the two larger side surfaces of the module.
- Bottom-Side-Plate Cooling: A hybrid configuration incorporating both bottom and side cold plates for enhanced cooling.
The cold plates feature internal serpentine or parallel flow channels. The bottom plate design employs wider channels, while the side plates use narrower, deeper channels to fit within the spatial constraints. Both designs are symmetric, featuring one inlet, one outlet, and four distributing/collecting branch channels. The relevant material properties for the energy storage cell, busbars, cold plates (aluminum), and coolant are summarized in Table 1.
| Component | Density (kg/m³) | Specific Heat Capacity (J/(kg·K)) | Thermal Conductivity (W/(m·K)) | Viscosity (mPa·s) |
|---|---|---|---|---|
| Energy Storage Cell (Homogenized) | 2024 | 964 | λ_x = 3.56, λ_y = 9.04, λ_z = 11.0 | – |
| Electrode / Busbar (Aluminum) | 2700 / 2719 | 900 / 871 | 234 | – |
| Cold Plate (Aluminum) | 2719 | 871 | 234 | – |
| Coolant (50% Water-Glycol) | 1073.35 | 3281 | 0.389 | 3.94 |
1.2 Mathematical Model
The following assumptions are made to simplify the model while retaining sufficient accuracy for engineering analysis:
- Thermophysical properties of all solid materials and the coolant are constant.
- The coolant is incompressible and Newtonian.
- Heat generation within the energy storage cell is uniform during discharge.
- Heat transfer between the module exterior and the ambient air is modeled using a constant convective heat transfer coefficient of 5.0 W/(m²·K).
The governing equations for fluid flow and heat transfer are applied. For the coolant domain within the cold plates, the conservation equations are:
Mass Conservation (Continuity):
$$ \frac{\partial \rho_{liq}}{\partial t} + \nabla \cdot (\rho_{liq} \vec{v}) = 0 $$
Momentum Conservation (Navier-Stokes):
$$ \frac{\partial (\rho_{liq} \vec{v})}{\partial t} + \nabla \cdot (\rho_{liq} \vec{v} \vec{v}) = -\nabla p + \nabla \cdot \vec{\tau} $$
Energy Conservation:
$$ \rho_{liq} c_{p,liq} \frac{\partial T_{liq}}{\partial t} + \nabla \cdot (\rho_{liq} c_{p,liq} \vec{v} T_{liq}) = \nabla \cdot (\lambda_{liq} \nabla T_{liq}) $$
where $\rho_{liq}$ is density, $\vec{v}$ is the velocity vector, $p$ is pressure, $\vec{\tau}$ is the stress tensor, $c_{p,liq}$ is specific heat, $T_{liq}$ is temperature, and $\lambda_{liq}$ is thermal conductivity.
For the solid domains (energy storage cells, busbars, cold plate), only the energy equation is solved:
$$ \rho_s c_{p,s} \frac{\partial T_s}{\partial t} = \nabla \cdot (\lambda_s \nabla T_s) + \dot{q}_{gen} $$
where $\dot{q}_{gen}$ is the volumetric heat generation rate within the energy storage cell.
Key performance metrics are defined. The average temperature $T_{avg}$ and temperature standard deviation $T_{\delta}$ across all energy storage cell surfaces quantify thermal uniformity:
$$ T_{avg} = \frac{\int_{A_{wall}} T \, dA}{\int_{A_{wall}} dA} $$
$$ T_{\delta} = \sqrt{ \frac{\int_{A_{wall}} (T – T_{avg})^2 \, dA}{\int_{A_{wall}} dA} } $$
System energy consumption is estimated by summing the pump power ($P_w$) and the chiller power ($P_c$). The pump power is calculated from the pressure drop $\Delta p$ and volumetric flow rate $V$:
$$ P_w = V \Delta p $$
The chiller power is related to the cooling load $P_{cool}$ and the Coefficient of Performance (COP, denoted as $\eta_{COP}$):
$$ P_c = \frac{P_{cool}}{\eta_{COP}} = \frac{\rho_{liq} V c_{p,liq} (T_{out} – T_{in})}{\eta_{COP}} $$
where a typical $\eta_{COP}$ value of 5 is used for the chiller unit.
1.3 Numerical Model and Boundary Conditions
The commercial CFD software ANSYS Fluent is employed to solve the coupled conjugate heat transfer problem. The pressure-based solver with the SIMPLE algorithm for pressure-velocity coupling is selected. The realizable k-ε turbulence model with standard wall functions is adopted for the coolant flow. A constant heat generation rate, derived from the cell’s internal resistance and efficiency at different discharge rates (C-rates), is applied to each energy storage cell. The heat generation values are listed in Table 2.
| Discharge Rate (C) | Cell Efficiency | Total Heat Generation (kW) | Volumetric Heat Source (W/m³) |
|---|---|---|---|
| 0.5 | 94% | 0.699 | 5363.7 |
| 0.75 | 93% | 1.223 | 6257.7 |
| 1.0 | 92% | 1.864 | 14303.2 |
Boundary conditions are set as follows: The initial temperature for the entire module is 25°C. A velocity inlet boundary condition is specified at the cold plate inlet(s), with coolant temperature fixed. A pressure-outlet condition is set at the outlet(s). The exterior surfaces of the module enclosure are subject to a convective condition with a heat transfer coefficient of 5.0 W/(m²·K) to an ambient temperature of 32°C.
A comprehensive grid independence study was conducted for all three cooling structures. The number of mesh elements was increased until the variation in the predicted maximum energy storage cell temperature was less than 0.1°C. The final mesh sizes selected were approximately 1.22 million, 4.40 million, and 5.14 million elements for the bottom-plate, side-plate, and bottom-side-plate models, respectively, ensuring both accuracy and computational efficiency.
2. Simulation Analysis and Results
2.1 Effect of Inlet/Outlet Diameter
The diameter of the coolant inlet/outlet ports influences the flow velocity and distribution for a given volumetric flow rate, thereby affecting cooling performance. Simulations were conducted for the baseline bottom-plate design at a 1C discharge rate, with a constant coolant flow rate and varying port diameters. The results for the maximum energy storage cell temperature are summarized in Table 3 and plotted graphically.
| Diameter (mm) | 5 | 7.5 | 10 | 12.5 | 15 |
|---|---|---|---|---|---|
| Max Cell Temp. (°C) | 44.27 | 43.62 | 42.64 | 43.07 | 44.23 |
The data reveals a non-monotonic relationship. As the diameter increases from 5 mm to 10 mm, the maximum temperature of the energy storage cell decreases from 44.27°C to 42.64°C. This improvement is attributed to a more uniform distribution of coolant entering the flow channels at a lower velocity, enhancing thermal contact. However, further increasing the diameter to 15 mm raises the maximum temperature back to 44.23°C. Excessively large diameters may lead to unfavorable flow patterns, such as recirculation zones or reduced turbulence, which impair heat transfer. Consequently, an optimal diameter of 10 mm is selected for all subsequent simulations to balance flow distribution and pressure loss.
2.2 Comparative Analysis of Cooling Structures
The thermal performance of the three cooling structures—bottom-plate, side-plate, and bottom-side-plate—was evaluated under various discharge rates (0.5C, 0.75C, 1C). The coolant inlet temperature was 25°C, and the inlet velocity was 0.9 m/s. Key results for maximum temperature ($T_{max}$), minimum temperature ($T_{min}$), temperature difference ($\Delta T = T_{max} – T_{min}$), and temperature standard deviation ($T_{\delta}$) are consolidated in Table 4.
| Cooling Structure | Discharge Rate (C) | Performance Metrics | |||
|---|---|---|---|---|---|
| $T_{max}$ (°C) | $T_{min}$ (°C) | $\Delta T$ (°C) | $T_{\delta}$ (°C) | ||
| Bottom-Plate | 1.0 | 42.5 | 25.7 | 16.8 | 4.65 |
| 0.75 | 36.9 | 25.5 | 11.4 | 3.12 | |
| 0.5 | 34.1 | 25.4 | 8.7 | 2.38 | |
| Side-Plate | 1.0 | 40.8 | 25.5 | 15.3 | 3.38 |
| 0.75 | 35.7 | 25.0 | 10.7 | 2.69 | |
| 0.5 | 33.2 | 25.1 | 8.1 | 2.04 | |
| Bottom-Side-Plate | 1.0 | 36.5 | 25.2 | 11.3 | 2.86 |
| 0.75 | 32.8 | 25.1 | 7.7 | 1.97 | |
| 0.5 | 31.0 | 25.2 | 5.8 | 1.45 | |
The analysis leads to several critical findings:
- Superior Cooling Efficacy: The side-plate structure consistently outperforms the traditional bottom-plate design. At the demanding 1C rate, it reduces the maximum energy storage cell temperature by 1.7°C. The combined bottom-side-plate structure delivers the most dramatic improvement, lowering the maximum temperature by 6.0°C compared to the baseline. This translates to a significantly safer operating window for the energy storage cell pack.
- Enhanced Temperature Uniformity: Temperature uniformity is crucial for preventing individual cell overstress and balancing degradation. The temperature standard deviation $T_{\delta}$ is a key indicator. The side-plate design reduces $T_{\delta}$ by approximately 27% (from 4.65°C to 3.38°C at 1C), while the bottom-side-plate design achieves a reduction of about 39% (to 2.86°C). This demonstrates that side cooling effectively mitigates the vertical thermal gradient inherent in bottom-cooled designs, leading to a more uniform temperature field across all energy storage cells.
- Thermal Pattern Shift: The temperature contours reveal distinct patterns. The bottom-cooled module shows a clear bottom-to-top temperature gradient. The side-cooled module exhibits a pattern where the central, interior cells are warmer, but the overall gradient is reduced. The hybrid design combines both effects, resulting in the most uniform temperature distribution.
2.3 System Energy Consumption and Optimal Operation Strategy
Beyond cooling performance, the energy consumed by the thermal management system itself (pump and chiller) is a critical factor for overall system efficiency. An operational optimization was conducted for the two superior designs—side-plate and bottom-side-plate—with the constraint that the maximum energy storage cell temperature must not exceed 45°C. The coolant inlet temperature ($T_{in}$) and inlet velocity ($v_{in}$) were varied to find the combination that minimizes total system power consumption ($P_{total} = P_w + P_c$).
The required minimum inlet velocity to meet the 45°C limit for different inlet temperatures was first determined, as shown in Table 5.
| Case | Cooling Structure | Inlet Temp. $T_{in}$ (°C) | Min. Inlet Velocity $v_{in}$ (m/s) |
|---|---|---|---|
| Case 1 | Side-Plate | 25 | ≥ 0.5 |
| Case 2 | 27 | ≥ 0.6 | |
| Case 3 | 29 | ≥ 0.8 | |
| Case 4 | Bottom-Side-Plate | 25 | ≥ 0.3 |
| Case 5 | 27 | ≥ 0.3 | |
| Case 6 | 29 | ≥ 0.3 | |
| Case 7 | 31 | ≥ 0.4 |
For each viable case in Table 5, the total system energy consumption was calculated. The results are presented in Table 6.
| Case | Cooling Structure | $T_{in}$ (°C) | $v_{in}$ (m/s) | Total Power $P_{total}$ (W) |
|---|---|---|---|---|
| Case 1 | Side-Plate | 25 | 0.5 | 338.3 |
| Case 2 | 27 | 0.6 | 538.7 | |
| Case 3 | 29 | 0.8 | 1228.4 | |
| Case 4 | Bottom-Side-Plate | 25 | 0.3 | 135.0 |
| Case 5 | 27 | 0.3 | 204.1 | |
| Case 6 | 29 | 0.3 | 191.9 | |
| Case 7 | 31 | 0.4 | 389.9 |
The analysis reveals a clear hierarchy in efficiency:
- For the side-plate structure, the most energy-efficient operating point is Case 1 ($T_{in}=25°C$, $v_{in}=0.5$ m/s), consuming 338.3 W.
- For the bottom-side-plate structure, the optimum is Case 4 ($T_{in}=25°C$, $v_{in}=0.3$ m/s), consuming only 135.0 W.
This result is profound: the bottom-side-plate cooling system can achieve the required thermal control with less than 40% of the energy consumed by the optimized side-plate-only system. This efficiency gain stems from the hybrid design’s exceptional cooling capability, which allows it to operate at a significantly lower coolant flow rate while still maintaining safe energy storage cell temperatures, thereby drastically reducing pump power. Furthermore, it can tolerate a slightly higher coolant inlet temperature in some cases (e.g., Case 6) while maintaining low flow, which reduces the chiller load. The optimal strategy for the hybrid system is therefore to utilize a moderately low coolant temperature (25°C) and a low flow rate (0.3 m/s).
3. Conclusions
This study presented a comprehensive thermal design and optimization analysis for a liquid-cooled energy storage battery module. Through detailed CFD modeling and simulation, the following key conclusions were drawn:
- The diameter of the coolant inlet/outlet ports has a significant impact on cooling performance. An optimal diameter exists (10 mm in this study) that ensures good flow distribution without inducing detrimental flow patterns. Both undersized and oversized ports can lead to elevated maximum temperatures in the energy storage cell pack.
- Innovative cooling structures incorporating side plates offer substantial advantages over traditional bottom-cooling. The side-plate design improves temperature uniformity by reducing the standard deviation by approximately 27% and lowers the peak temperature. The combined bottom-side-plate structure represents a breakthrough, reducing the peak energy storage cell temperature by up to 6.0°C and improving temperature uniformity by 39% at a 1C discharge rate compared to the baseline. This dramatically enhances the safety and potential lifespan of the module.
- System-level energy optimization is crucial. The superior cooling performance of the bottom-side-plate structure translates directly into operational energy savings. Its optimal operating point requires only 135.0 W of total system power, which is 2.5 times lower than the best achievable with the side-plate-only design (338.3 W). The recommended strategy for the hybrid system is to employ a coolant inlet temperature of 25°C and a relatively low flow velocity of 0.3 m/s.
In summary, the integration of side cooling plates, particularly in a hybrid configuration with a bottom plate, is a highly effective strategy for thermal management of energy storage battery modules. It simultaneously addresses the critical challenges of maximum temperature suppression, temperature uniformity enhancement, and minimization of system energy consumption. This work provides a validated modeling framework and clear design guidelines for developing next-generation, high-performance, and energy-efficient thermal management systems for large-scale energy storage applications.