The pursuit of carbon neutrality has positioned new energy technologies, particularly lithium-ion battery energy storage systems (BESS), at the forefront of the global energy transition. These systems are pivotal for grid stabilization, renewable energy integration, and electric mobility due to their high energy density and extended cycle life. However, the operational performance, longevity, and, most critically, the safety of a battery energy storage system are intrinsically linked to its operating temperature. During charge and discharge cycles, lithium-ion cells generate substantial heat. Inadequate heat dissipation can lead to elevated temperatures, accelerated degradation, and in severe cases, thermal runaway—a catastrophic failure mode. Therefore, an efficient thermal management system (TMS) is not an accessory but a fundamental component for a reliable battery energy storage system.
Among the various cooling technologies—including phase change materials and heat pipes—forced air cooling (air-cooled) and indirect liquid cooling (liquid-cooled) are the most technologically mature and widely deployed in commercial battery energy storage system applications. Air-cooled systems leverage the convective heat transfer of air blown across battery modules. Their advantages include structural simplicity, low weight, and minimal maintenance. Conversely, their primary limitation is the relatively low heat transfer coefficient of air, which can result in significant temperature gradients within a large battery energy storage system pack, especially under high load conditions. Liquid-cooled systems, typically employing coolant circulated through plates or tubes in contact with the cells, offer a significantly higher heat transfer coefficient. This leads to superior temperature uniformity and cooling capacity, albeit at the cost of increased system complexity, weight, and potential leakage risks.
While extensive research exists on optimizing channel designs for both methods, a holistic, quantitative comparison focusing on comprehensive performance metrics is essential for informed design choices in battery energy storage system engineering. This analysis employs a numerical simulation framework to dissect and compare the thermal-hydraulic performance of parallel-configuration air-cooled and liquid-cooled modules for a battery energy storage system. The evaluation extends beyond mere maximum temperature to encompass flow resistance, combined performance factors, and robustness against environmental fluctuations.
1. Modeling Framework and Methodology
1.1 Geometrical Configuration
The analysis is based on a representative battery energy storage system module comprising 26 prismatic lithium-ion cells (50 Ah, 3.7 V nominal), arranged in two columns of 13 cells each. Each cell is treated as a homogeneous anisotropic heat source with dimensions 205 mm × 174 mm × 72 mm. The heat generation rate is set to a constant value corresponding to a 1C charging scenario.
Air-Cooled Module Design: The design adopts a parallel air flow configuration. A lower plenum (diffusion chamber) distributes incoming air into individual channels between adjacent cells. After cooling the cells, the air converges into an upper plenum before exiting. The inter-cell channel width is 10 mm, with inlet/outlet dimensions of 408 mm × 20 mm.
Liquid-Cooled Module Design: The design features cold plates with parallel direct channels attached to the large faces of the cells. A 50% ethylene glycol-water mixture serves as the coolant. The channels have a thickness of 3.5 mm, and 1.5 mm thick thermal pads provide insulation between adjacent cells, isolating their thermal profiles.
1.2 Mathematical Model and Assumptions
The numerical simulation is conducted using the Finite Volume Method (FVM). To render the problem tractable, the following assumptions are made:
- The flow is steady-state, incompressible, and turbulent.
- The battery cells are isotropic, uniform heat generation sources.
- Radiation heat transfer is negligible compared to convection and conduction.
- Minor components (busbars, small fixtures) are omitted from the geometry.
The governing equations for fluid flow and heat transfer are solved using the pressure-based SIMPLE algorithm with first-order upwind discretization schemes.
Governing Equations:
Conservation of Mass (Continuity):
$$ \nabla \cdot (\rho \vec{u}) = 0 $$
Conservation of Momentum (Navier-Stokes):
$$ \rho (\vec{u} \cdot \nabla) \vec{u} = -\nabla p + \mu \nabla^2 \vec{u} $$
Energy Conservation for Fluid:
$$ \rho c_p (\vec{u} \cdot \nabla T) = \nabla \cdot (k \nabla T) $$
Energy Conservation for Solid (Battery Cell):
$$ \nabla \cdot (k_s \nabla T_s) + \dot{q}_g = 0 $$
where \( \rho \) is density, \( \vec{u} \) is velocity vector, \( p \) is pressure, \( \mu \) is dynamic viscosity, \( c_p \) is specific heat capacity, \( T \) is temperature, \( k \) is thermal conductivity, \( \dot{q}_g \) is volumetric heat generation rate, and subscripts \( s \) denote solid properties.
Turbulence is modeled using the standard \( k-\epsilon \) model.
1.3 Boundary Conditions, Materials, and Grid Independence
The baseline ambient and coolant inlet temperature is set to 20°C. The coolant is a 50% ethylene glycol-water solution. Aluminum alloy is assigned for the casing and cold plates. The cell’s anisotropic thermal conductivity is defined with different values for the in-plane and through-plane directions. A constant heat flux boundary condition representing the total heat generation is applied to the cell volumes. Adiabatic conditions are set on the outer casing walls.
A grid independence study was conducted to ensure the accuracy of the solution. The key monitoring parameters were the module’s maximum temperature (\(T_{max}\)) and the system pressure drop (\(\Delta P\)). The results confirmed that solutions become mesh-independent beyond approximately 5.64 million poly-hexcore cells, with changes in target parameters below 0.5% upon further refinement. The selected mesh maintained an average orthogonal quality above 0.7.
2. Performance Evaluation Metrics
A comprehensive comparison requires moving beyond single-point metrics. The following parameters and derived factors are used to evaluate the battery energy storage system thermal management systems.
2.1 Core Thermal Metrics
| Metric | Symbol | Description | Significance for BESS |
|---|---|---|---|
| Maximum Temperature | \(T_{max}\) | Highest temperature within the battery module. | Primary indicator of safety risk and potential for thermal runaway. |
| Average Temperature | \(T_{avg}\) | Spatial average temperature of all cells. | Related to overall electrochemical performance and aging rate. |
| Maximum Temperature Difference | \(\Delta T_{cell}\) | Difference between \(T_{max}\) and \(T_{min}\) in the module. | Critical for cell-to-cell uniformity, balancing, and lifespan. |
2.2 Derived Thermo-Hydraulic Performance Factors
To quantitatively compare the intrinsic heat transfer and flow resistance characteristics, dimensionless factors are employed.
Heat Transfer Performance (J-factor): Analogous to the Colburn \(j\)-factor, it evaluates the heat transfer efficiency relative to the pumping power incurred.
$$ J = \frac{Nu}{Re \cdot Pr^{1/3}} $$
where \(Nu\) is the Nusselt number, \(Re\) is the Reynolds number, and \(Pr\) is the Prandtl number. A higher \(J\) indicates better heat transfer performance for a given flow condition.
Flow Resistance Performance (F-factor): Represents the dimensionless pressure loss or friction factor.
$$ F = \frac{2 \Delta P}{\rho u^2} $$
where \(\Delta P\) is the pressure drop across the cooling system and \(u\) is the characteristic flow velocity. A lower \(F\) is desirable, indicating lower pumping power requirement.
Comprehensive Thermo-Hydraulic Performance (θ-factor): This ratio provides a single metric balancing heat transfer against flow resistance.
$$ \theta = \frac{J}{F} $$
A higher θ-factor signifies a more efficient cooling design, delivering better heat transfer per unit of pressure loss. This is a crucial figure of merit for optimizing the energy efficiency of the overall battery energy storage system, as the cooling system’s parasitic load (e.g., pump or fan power) directly impacts system efficiency.
3. Simulation Results and Comparative Analysis
The simulations were performed for a range of coolant inlet-to-outlet temperature rises (\(\Delta T_f\)), which directly correlates to the required volumetric flow rate (\(Q\)) via the energy balance: \(\dot{q} = \rho c_p Q \Delta T_f\). The analysis covers \(\Delta T_f\) from 1°C to 10°C.
3.1 Thermal Performance Analysis
The core thermal metrics for both cooling strategies across different flow rates are summarized below.
| \(\Delta T_f\) (°C) | Cooling Method | \(T_{max}\) (°C) | \(T_{avg}\) (°C) | \(\Delta T_{cell}\) (°C) |
|---|---|---|---|---|
| 1 | Air-Cooled | 24.5 | 23.1 | 3.2 |
| Liquid-Cooled | 27.0 | 26.8 | 0.4 | |
| 4 | Air-Cooled | 32.8 | 29.5 | 6.8 |
| Liquid-Cooled | 28.7 | 28.5 | 0.5 | |
| 8 | Air-Cooled | 41.5 | 36.2 | 10.6 |
| Liquid-Cooled | 31.5 | 31.2 | 0.6 | |
| 10 | Air-Cooled | 46.0 | 39.5 | 12.9 |
| Liquid-Cooled | 33.0 | 32.7 | 0.7 |
Key Observations on Thermal Performance:
- Crossover Point: There exists a critical \(\Delta T_f\) (approximately 1.76°C) below which the air-cooled system yields a lower \(T_{max}\). This is because at very high air flow rates (low \(\Delta T_f\)), forced convection can be effective. However, beyond this point, the liquid-cooled system consistently maintains a lower and safer maximum temperature for the battery energy storage system module.
- Temperature Uniformity: The superiority of liquid cooling in ensuring cell-to-cell temperature uniformity is stark. For \(\Delta T_f = 6°C\), the \(\Delta T_{cell}\) for liquid cooling is about 0.5°C, compared to over 6°C for air cooling. This exceptional uniformity is critical for minimizing imbalance and maximizing the cycle life of a series-connected battery energy storage system pack.
- Heat Transfer Coefficient (J-factor): The calculated J-factor for the liquid-cooled system (\(9.67 \times 10^{-3}\)) is approximately 3.3 times higher than that of the air-cooled system (\(2.91 \times 10^{-3}\)), quantitatively confirming its intrinsically superior heat transfer capability.
3.2 Flow Resistance and Combined Performance
The pressure drop and derived performance factors reveal the trade-offs involved.
| \(\Delta T_f\) (°C) | Cooling Method | Pressure Drop \(\Delta P\) (Pa) | F-factor (\( \times 10^{-3} \)) | θ-factor (\( \times 10^{-3} \)) |
|---|---|---|---|---|
| 1 | Air-Cooled | ~7800 | 1.26 | 2.30 |
| Liquid-Cooled | ~4200 | 6.57 | 1.47 | |
| 4 | Air-Cooled | ~490 | 1.98 | 1.47 |
| Liquid-Cooled | ~265 | 10.34 | 0.94 | |
| 10 | Air-Cooled | ~50 | 2.15 | 1.35 |
| Liquid-Cooled | ~43 | 12.25 | 0.79 |
Key Observations on Hydraulic and Combined Performance:
- Flow Resistance (F-factor): The liquid-cooled system exhibits a significantly higher F-factor (5 to 17 times greater across the range), indicating a higher flow resistance characteristic. This is attributable to the smaller hydraulic diameter of liquid channels and the higher viscosity of the coolant compared to air.
- Comprehensive Performance (θ-factor): Despite its higher flow resistance, the liquid-cooled system’s superior heat transfer capability results in a higher θ-factor across all operating conditions. This means that per unit of pumping power expended, the liquid-cooled system removes more heat. The performance advantage of liquid cooling becomes more pronounced at higher flow rates (lower \(\Delta T_f\)), where its heat transfer advantage outweighs its flow resistance penalty.
3.3 Impact of Ambient Temperature Variation
The robustness of a battery energy storage system TMS against fluctuating environmental conditions is vital for outdoor deployments. Simulations were conducted with ambient temperatures (\(T_{amb}\)) varying from 0°C to 30°C at a fixed \(\Delta T_f = 6°C\).
| \(T_{amb}\) (°C) | Cooling Method | \(T_{max}\) (°C) | \(\Delta T_{cell}\) (°C) | \(\Delta T_{max}\) vs. 20°C case |
|---|---|---|---|---|
| 0 | Air-Cooled | 30.2 | 11.2 | -6.3 |
| 0 | Liquid-Cooled | 28.9 | 0.4 | -1.1 |
| 20 | Air-Cooled | 36.5 | 7.1 | 0.0 |
| 20 | Liquid-Cooled | 30.0 | 0.5 | 0.0 |
| 30 | Air-Cooled | 40.7 | 3.0 | +4.2 |
| 30 | Liquid-Cooled | 31.1 | 0.5 | +1.1 |
Key Observations on Environmental Robustness:
- Stronger Insulation for Liquid Cooling: The liquid-cooled battery energy storage system module demonstrates remarkable insensitivity to ambient temperature swings. Its \(T_{max}\) increased by only 1.1°C when \(T_{amb}\) rose from 20°C to 30°C, and its temperature uniformity (\(\Delta T_{cell}\)) remained virtually unchanged.
- High Sensitivity of Air Cooling: The air-cooled module’s temperature is heavily influenced by \(T_{amb}\), with \(T_{max}\) increasing by 4.2°C for the same 10°C ambient rise. Furthermore, its temperature gradient changes dramatically, indicating that its cooling performance and cell balance are highly dependent on the inlet air temperature.
4. Conclusions and Implications for BESS Design
This detailed numerical investigation provides quantitative insights for selecting and optimizing thermal management systems for large-scale battery energy storage system applications.
- Performance Dominance of Liquid Cooling: For the majority of operating scenarios, especially those requiring high heat flux dissipation or superior temperature uniformity, indirect liquid cooling is the unequivocally superior choice. Its high heat transfer coefficient ensures lower maximum temperatures and exceptional cell-to-cell uniformity (\(\Delta T_{cell} < 1°C\)), which is paramount for the safety and longevity of a battery energy storage system.
- The Trade-off and Crossover: While liquid cooling incurs a higher flow resistance (higher F-factor), its overall thermo-hydraulic efficiency (θ-factor) is greater. A critical design crossover point exists: only when the system can operate with an exceptionally low coolant temperature rise (very high flow rate) might a highly optimized air-cooled system compete on maximum temperature alone. However, it would still fail to match the temperature uniformity of liquid cooling.
- Environmental Robustness: Liquid-cooled battery energy storage system designs exhibit significantly greater resilience to variations in external ambient temperature. This makes them particularly suitable for applications without a controlled environment, as their performance is more predictable and stable.
- Design Optimization Focus: For air-cooled systems, the primary optimization challenge lies in improving flow distribution among parallel channels to reduce temperature gradients. For liquid-cooled systems, effort should focus on minimizing the pressure drop in manifolds and channel designs (lowering the F-factor) to enhance the already favorable θ-factor and reduce parasitic pumping power.
In conclusion, for mission-critical, high-density, or long-duration battery energy storage system projects where safety, lifespan, and reliable performance under varying conditions are non-negotiable, the added complexity and cost of a liquid-cooled thermal management system are justified by its demonstrably superior thermal control, efficiency, and environmental robustness. The methodologies and comparative metrics presented here provide a framework for the ongoing optimization of thermal management in next-generation battery energy storage systems.