Abstract
With the rapid growth of energy storage demands, individual battery capacities are increasing, making large-capacity batteries the mainstream in electrochemical energy storage systems. However, research on battery pack cooling systems primarily focuses on small-capacity batteries. This paper investigates the cooling performance of an immersion liquid cooling system for 280Ah large-capacity battery packs. The effects of battery spacing, coolant inlet/outlet configurations, inlet flow rates, and coolant types on the cooling performance are discussed. Additionally, the study analyzes the influence of coolant thermophysical properties on the cooling effect. Results show that an appropriate increase in battery spacing positively impacts cooling, with the optimal spacing being 5mm, reducing the maximum temperature difference (ΔT max) and maximum temperature (T max) by 14.3% and 15.0%, respectively, compared to 0mm spacing. The inlet position significantly influences ΔT max and T max compared to the outlet position, and increasing the inlet flow rate decreases ΔT max and T max. Among various coolants, deionized water exhibits the best cooling performance, whereas silicone oil performs the worst. The study ranks the importance of coolant thermophysical properties as density, specific heat capacity, thermal conductivity, and dynamic viscosity. These findings provide guidance for the design of large-capacity battery pack immersion liquid cooling systems.
1. Introduction
Global energy crises have intensified, prompting a shift from fossil fuels to renewable energy sources, which has become a common goal for many countries. However, the volatility, intermittency, and unpredictability of renewable sources like wind and solar power have led to challenges in peak shaving, power transmission, and consumption, hindering their sustainable development . Energy storage technologies, particularly electrochemical storage, have gained popularity due to their rapid response, energy time-shifting capabilities, and flexibility in deployment, playing a crucial role in ensuring power system stability and facilitating the transition to a low-carbon energy system .
Lithium-ion batteries (LIBs) are widely used in energy storage systems due to their high energy density, low environmental impact, and good cycling performance. However, LIBs generate substantial heat during charging and discharging, and maintaining optimal operating temperatures is crucial for performance and safety. Ideally, LIBs should operate between 15°C and 35°C, with a maximum temperature difference of no more than 5°C between individual cells within a pack. Therefore, efficient battery thermal management systems (BTMS) are essential for regulating temperature rise and ensuring uniform temperatures across the battery pack.
BTMS can be categorized based on the heat transfer medium: air cooling, phase change material (PCM) cooling, and liquid cooling. Air cooling is cost-effective and structurally simple but suffers from low thermal conductivity and specific heat capacity, limiting its cooling efficiency . PCM cooling offers superior temperature control and high heat dissipation rates but is hindered by structural complexity, poor mechanical properties, and high costs. Liquid cooling can be further divided into indirect (cold plate) and direct (immersion) cooling. While indirect liquid cooling effectively removes heat, direct immersion cooling offers advantages such as minimal contact thermal resistance, large heat transfer areas, high cooling efficiency, compact structures, and enhanced thermal runaway prevention . However, most existing research on immersion cooling focuses on small-capacity batteries used in electric vehicles, whereas large-capacity batteries for energy storage require a more thorough investigation .
This paper presents a numerical simulation study on the cooling performance of an immersion liquid cooling system for 280Ah large-capacity LIB packs. The effects of battery spacing, coolant inlet/outlet configurations, inlet flow rates, and coolant types are systematically analyzed. Furthermore, the influence weights of coolant thermophysical properties on the cooling effect are determined.
2. Physical and Mathematical Models
2.1 Physical Model
The studied system consists of a battery pack immersed in a dielectric coolant, with the coolant circulating between the battery enclosure and a heat exchanger driven by a pump. The battery pack comprises 4×13 square aluminum-cased LFP (lithium iron phosphate) cells with dimensions of 204mm x 174mm x 72mm (H x W x T) arranged in series and parallel. The battery enclosure height is 230mm, with a 25mm gap between the pack and enclosure walls. The vertical cell spacing (d2) is fixed at 10mm, while the horizontal spacing (di) varies from 0 to 10mm. Nine coolant inlet/outlet configurations are considered, as illustrated in Table 1.
Table 1: Coolant inlet and outlet heights from the bottom of the battery case (mm)
| Configuration | Inlet Height (mm) | Outlet Height (mm) |
|---|---|---|
| Case 1 | 182 | 182 |
| Case 2 | 182 | 102 |
| Case 3 | 182 | 20 |
| Case 4 | 102 | 182 |
| Case 5 | 102 | 102 |
| Case 6 | 102 | 20 |
| Case 7 | 20 | 182 |
| Case 8 | 20 | 102 |
| Case 9 | 20 | 20 |
2.2 Mathematical Model
2.2.1 Battery Heat Generation Model
Assuming uniform material distribution, internal heating, and constant material properties with temperature, the battery’s heat generation is modeled using the following partial differential equation in a Cartesian coordinate system:
rhobcb∂t∂Tb=λx∂x2∂2Tb+λy∂y2∂2Tb+λz∂z2∂2Tb+qv
where ρb and cb are the battery density and specific heat capacity, respectively; Tb is the battery temperature; λx, λy, and λz are the thermal conductivities along the x, y, and z axes; t is time; and qv is the volumetric heat generation rate calculated as:
qv=Vb1[I2Rt−IT∂T∂U0]
where Vb is the battery volume, I is the operating current, Rt is the internal resistance, T is the operating temperature, and ∂T∂U0 is the temperature coefficient of voltage, assumed to be 11.16mV.
Table 2: Battery parameters
| Parameter | Value |
|---|---|
| Geometry (mm³) | 204x174x72 (H x W x T) |
| Capacity (Ah) | 280 |
| Nominal Voltage (V) | 3.2 |
| Charging/Discharging Voltage (V) | 2.5-3.65 |
| Mass (g) | 5435 |
| Specific Heat Capacity (J/kg·K) | 1029.4 |
| Density (kg/m³) | 2118 |
| Internal Resistance (mΩ) | 0.43 |
| Thermal Conductivity (W/m·K) | X/Y/Z: 21.6/2.1/21.6 |
2.2.2 Control Equations
The system’s behavior is governed by conservation equations for mass, momentum, and energy in both the solid (battery) and fluid (coolant) domains. The battery energy equation is:
frac∂(ρbcbTb)∂τ=−∇⋅(λb∇Tb)+qv
The coolant domain equations are:
- Continuity equation:frac∂ρ∂τ+∇⋅(ρν)=0
- Momentum equation:frac∂(ρν)∂τ+∇⋅(ρνν)=−∇p+(μ∇2ν)+ρg
- Energy equation:rhocp∂τ∂T+∇⋅(ρcpνT)=∇⋅(λ∇T)
where λb is the battery thermal conductivity, ρ, ν, p, cp, and λ are the coolant density, velocity, pressure, specific heat capacity, and thermal conductivity, respectively.
Table 3: Coolant thermophysical properties
| Coolant | Density (kg/m³) | Specific Heat Capacity (J/kg·K) | Thermal Conductivity (W/m·K) | Dynamic Viscosity (Pa·s) |
|---|---|---|---|---|
| Synthetic Oil | 807 | 2523 | 0.159 | 0.0070 |
| (Example) |
3. Numerical Methods and Boundary Conditions
The numerical simulations were performed using ANSYS Fluent software. The coolant flow and heat transfer processes were solved using a pressure-based solver with a second-order upwind scheme for discretization. The initial and ambient temperatures were set to 25°C. The battery was discharged at a 1C rate, and the coolant inlet was set to a velocity inlet with a temperature of 25°C. The outlet was modeled as a pressure outlet at 101,325Pa. Radiation heat transfer and contact thermal resistance were neglected. The battery enclosure exchanged heat with the ambient through natural convection, with a heat transfer coefficient of 5W/m²·K.
4. Grid Generation and Independence Validation
A polyhedral mesh was generated for both the coolant and battery domains. Grid independence was validated by varying the mesh density and monitoring T max and ΔT max at an inlet flow rate of 0.4m/s and a horizontal spacing of 5mm. the results converged with a mesh size of approximately 6.05 million elements, with errors below 1%. This mesh density was used for all subsequent simulations.
5. Results and Discussion
5.1 Effect of Battery Horizontal Spacing
Increasing the horizontal battery spacing (di) from 0 to 10mm initially reduced ΔT max and T max before increasing again beyond 5mm. At di=0mm, the heat transfer area was minimized, resulting in the highest ΔT max (10.94°C) and T max (37.19°C). As di increased to 5mm, ΔT max and T max decreased to 9.37°C and 35.35°C, respectively, due to increased heat transfer area and reduced coolant flow resistance. Beyond 5mm, ΔT max and T max increased slightly due to decreased coolant velocities.
5.2 Effect of Coolant Inlet/Outlet Configurations
Varying inlet and outlet positions significantly influenced ΔT max and T max . The inlet position had a greater impact than the outlet position, with Case 4 (inlet at 102mm, outlet at 182mm) exhibiting the lowest ΔT max and T max. Velocity and vorticity distributions at a cross-section revealed more pronounced coolant perturbations near the inlet compared to the outlet .
5.3 Effect of Inlet Flow Rate
Increasing the inlet flow rate from 0.2 to 1.6m/s steadily decreased ΔT max and T max, with diminishing returns at higher flow rates. Between 0.2 and 0.4m/s, ΔT max and T max reduced by 21.1% and 8.2%, respectively. Above 0.8m/s, the battery temperatures stabilized more rapidly with increasing flow rates.
5.4 Effect of Coolant Type
Among the tested coolants (synthetic oil, MIVOLT-DF7, FC-72, deionized water, and silicone oil), deionized water showed the best cooling performance, while silicone oil was the least effective . The coolant properties’ influence on cooling was ranked as density > specific heat capacity > thermal conductivity > dynamic viscosity.
6. Conclusion
This study systematically investigated the cooling performance of an immersion liquid cooling system for 280Ah large-capacity LIB packs through numerical simulations. Key findings include:
- Battery Spacing: An optimal horizontal spacing of 5mm minimizes ΔT max and T max, highlighting the importance of spacing optimization.
- Inlet/Outlet Configurations: The coolant inlet position significantly influences cooling performance, with the inlet height having a greater impact than the outlet height.
- Inlet Flow Rate: Increasing the inlet flow rate reduces ΔT max and T max, with diminishing returns at higher flow rates.
- Coolant Type: Deionized water exhibits the best cooling performance due to its low viscosity and high thermal conductivity and specific heat capacity.
- Coolant Properties: The influence of coolant properties on cooling performance is ranked as density > specific heat capacity > thermal conductivity > dynamic viscosity.
These findings provide valuable insights for the design and optimization of immersion liquid cooling systems for large-capacity energy storage batteries.