The global transition towards sustainable energy infrastructure has created an unprecedented demand for large-scale, efficient, and reliable energy storage solutions. Central to these solutions is the lithium-ion battery energy storage system, prized for its high energy density and long cycle life. However, the thermal behavior of these systems presents a significant challenge. Poor thermal stability, accelerated side reactions at elevated temperatures, and the latent risk of thermal runaway are critical barriers to their safe and durable operation. Effective thermal management is therefore not merely an enhancement but a fundamental requirement, aiming to maintain the battery pack within its optimal operating window of 20–35°C while minimizing temperature gradients between cells. This work employs a numerical simulation approach, calibrated with experimental data, to investigate and optimize key design factors influencing the thermal performance of a commercial-scale battery energy storage system. The focus is on three pivotal aspects: the geometry of the cooling channel, the strategy for deploying liquid cooling plates, and the operational parameter of coolant flow rate. The overarching goal is to develop a holistic understanding that leads to a cooling design offering superior temperature uniformity, constrained maximum temperature rise, and lower auxiliary energy consumption for the cooling system itself.
The performance and longevity of a battery energy storage system are intrinsically linked to its operating temperature. Excessive heat degrades components, while large temperature differences between cells lead to unbalanced aging and reduced usable capacity. Consequently, the design of the thermal management system (TMS) is a cornerstone of battery energy storage system engineering. Liquid cooling, particularly indirect cooling via cold plates, has become the industry standard for large-scale applications due to its high heat transfer efficiency. This study delves into advanced liquid cooling strategies, moving beyond conventional designs to explore geometries and configurations that can unlock higher performance. By systematically varying critical parameters through computational fluid dynamics (CFD) and heat transfer simulations, we quantify their impact on the thermal state of a representative battery pack. The findings provide actionable insights for designing next-generation thermal management systems that ensure the safety, reliability, and economic viability of stationary battery energy storage system deployments.
1. Numerical Methodology and Model Foundation
To ensure the fidelity of the numerical study, a rigorous model was developed and validated against experimental measurements. The physical domain represents a single module of a larger battery energy storage system, configured with a 1P52S arrangement of semi-solid lithium-ion cells. Each cell has a nominal capacity of 280 Ah. The assembly includes essential components such as thermal interface materials (TIMs), insulating polycarbonate (PC) boards, and the aluminum liquid cooling plates. The material properties critical for the conjugate heat transfer simulation are summarized in the table below.
| Component | Density, ρ (kg/m³) | Specific Heat, cp (J/(kg·K)) | Thermal Conductivity, k (W/(m·K)) |
|---|---|---|---|
| Battery Cell | 2152 | 1051.1 | kth=1.04, kw, kh=21.05 |
| Liquid Cooling Plate (Al) | 2680 | 880 | 237 |
| Thermal Silicone Pad (TIM) | 2420 | 967 | 2.1 |
| Coolant (Water-Glycol) | 1065 | 3394 | 0.419 |
| PC Insulation Board | 1200 | 1340 | 0.194 |
1.1 Governing Equations for Heat and Fluid Flow
The thermal behavior of the battery energy storage system is governed by the principles of conservation of energy, mass, and momentum. The following simplified assumptions were applied: material properties are temperature-independent, contact resistances are negligible relative to other thermal resistances in the system, and radiation heat transfer is insignificant compared to convection and conduction. The core equations are:
Energy Conservation for the Battery Cell:
$$ \rho c_p \frac{\partial T}{\partial t} = \nabla \cdot (k \nabla T) + Q_{gen} $$
Here, $Q_{gen}$ represents the volumetric heat generation rate within the cell, which is the most critical input for the thermal model of any battery energy storage system.
Navier-Stokes Equations for Coolant Flow:
$$ \frac{\partial}{\partial t} (\rho \vec{v}) + \nabla \cdot (\rho \vec{v} \vec{v}) = -\nabla p + \mu \nabla^2 \vec{v} + \vec{F} $$
$$ \frac{\partial \rho}{\partial t} + \nabla \cdot (\rho \vec{v}) = 0 $$
These equations describe the motion and pressure drop of the coolant within the cold plate channels, directly influencing the convective heat removal capability.
1.2 Battery Heat Generation Model
The heat generation term $Q_{gen}$ is calculated using the Bernardi model, which accounts for both irreversible (Joule) heating and reversible (entropic) heating effects:
$$ Q_{gen} = \frac{1}{V_b} \left( I^2 R(T, SOC) + I T \frac{dE_{OC}}{dT} \right) $$
where $I$ is the discharge current (0.5C rate), $R$ is the internal resistance, $T$ is the cell temperature, $V_b$ is the cell volume, and $\frac{dE_{OC}}{dT}$ is the entropy coefficient, taken as 0.00049 V/K. The internal resistance $R$ is not a constant but a function of both State of Charge (SOC) and temperature. Accurate characterization of this dependency is vital for a trustworthy simulation of a battery energy storage system under dynamic loads. The 10-second pulse discharge resistance (DCR) map used in this study is tabulated below.
| SOC (%) | -10°C (mΩ) | 0°C (mΩ) | 25°C (mΩ) | 45°C (mΩ) |
|---|---|---|---|---|
| 0 | – | – | – | – |
| 20 | 3.11 | 1.50 | 0.49 | 0.32 |
| 40 | 2.46 | 1.34 | 0.44 | 0.30 |
| 60 | 2.18 | 1.25 | 0.42 | 0.29 |
| 80 | 2.07 | 1.21 | 0.41 | 0.29 |
| 100 | 1.97 | 1.17 | 0.39 | 0.29 |
The SOC is updated during the simulation according to:
$$ SOC(t) = 1 – \frac{I t}{C_N} $$
where $C_N$ is the nominal capacity.
1.3 Boundary Conditions, Validation, and Performance Metrics
The computational domain includes the solid components and the fluid domain within the cold plates. A constant ambient temperature of 25°C was set. The external surfaces of the battery energy storage system module exchange heat with the environment via natural convection, defined by a heat transfer coefficient of 5 W/(m²·K). The coolant inlet was defined as a mass flow inlet (e.g., 0.1791 kg/s for a 10 L/min case) with a temperature of 22°C, and the outlet was set as a pressure outlet. A 2-mm thick thermal pad was modeled between the cells and the cold plate to account for contact resistance.
Model Validation: Prior to full pack simulation, the electro-thermal model for a single cell was validated against experimental data at 0.5C and 1C discharge rates. The maximum error in temperature prediction was below 2.9%, confirming the model’s accuracy for studying the complete battery energy storage system pack.
Key Performance Indicators (KPIs): The analysis focuses on two primary thermal KPIs and one hydraulic KPI:
1. Maximum Cell Temperature Rise (ΔTmax): The highest temperature on any cell surface minus the initial temperature.
2. Global Maximum Temperature Difference (ΔTglobal): The difference between the absolute highest and lowest temperatures anywhere within the entire battery pack. This is a stricter metric than the common “top surface”温差 and is critical for assessing uniformity in a battery energy storage system.
3. Cooling System Energy Consumption (Ecool): The pumping power required, calculated by integrating the pressure drop across the cold plate over time:
$$ E_{cool} = \int_0^t \Delta p \cdot \dot{V}_{in} \, dt $$
where $\Delta p$ is the total pressure drop and $\dot{V}_{in}$ is the inlet volumetric flow rate.
2. Parametric Investigation and Results
The discharge simulation was run for 6480 seconds (0.5C rate down to ~10% SOC). The effects of three independent design and operational variables were investigated sequentially and then in combination.
2.1 Influence of Cold Plate Channel Aspect Ratio (W/H)
The first design variable is the cross-sectional geometry of the serpentine channels within the cold plate. Three rectangular channel types with a fixed height (H=6 mm) but varying width-to-height ratios (W/H = 3, 5, 7) were analyzed under a constant inlet flow rate of 10 L/min and a bottom-only cooling layout.
Thermal Impact: The effect on the peak cell temperature was minimal but discernible. A smaller aspect ratio (W/H=3) yielded the lowest temperature, while W/H=7 resulted in the highest. The relative difference at the end of discharge was less than 0.7%. This minor effect is due to competing factors: a smaller W/H reduces the contact surface area for heat transfer but increases the flow velocity (for a constant flow rate and height), thereby increasing the convective heat transfer coefficient. The near-net result is a slight thermal advantage for the more compact channel design.
Hydraulic and Energy Impact: The impact on system pressure drop and pumping energy was profound and is summarized below.
| Channel Aspect Ratio (W/H) | ΔTmax at 6480s (°C) | Δp (kPa) | Ecool at 6480s (kJ) | Relative Δp vs. W/H=3 |
|---|---|---|---|---|
| 3 | 7.30 | 4.8 | ~23.5 | Base |
| 5 | 7.32 | 6.6 | ~32.3 | +37.5% |
| 7 | 7.35 | 7.8 | ~34.5 | +62.5% |
The data clearly shows that while thermal performance is similar, the hydraulic penalty for larger aspect ratio channels is significant. The channel with W/H=3 offers the best trade-off, providing marginally better cooling with substantially lower pressure drop and energy consumption for the battery energy storage system’s auxiliary systems.
2.2 Influence of Liquid Cooling Plate Layout
The second and most impactful design variable is the spatial arrangement of the cooling plates. Two configurations were compared:
1. Baseline (Bottom-Only): A single cold plate placed underneath the battery cells.
2. Enhanced (Bottom + Two Sides): Cold plates placed underneath and along the two large side faces of the cell stack (covering approximately the upper half of the cell height). The total coolant flow rate was kept constant at 20 L/min (10 L/min per bottom plate, distributed equally to side plates).
The results demonstrate a transformative improvement in the thermal management of the battery energy storage system module.
| Cooling Layout | ΔTmax at 6480s (°C) | ΔTglobal at 6480s (°C) | Peak Heat Transfer Power (kW) |
|---|---|---|---|
| Bottom-Only | 7.62 | 8.80 | ~0.23 |
| Bottom + Two Sides | 0.51 | 3.00 | ~1.00 |
| Improvement | -93.3% / -7.11°C | -65.9% / -5.80°C | +335% |
The bottom+side cooling strategy dramatically reduces the maximum temperature rise, keeping the cells well within the ideal range. More importantly, it drastically improves temperature uniformity. The large vertical gradient characteristic of bottom-only cooling is effectively mitigated by the side plates, which directly extract heat from the mid-to-upper sections of the cells. This superior uniformity is crucial for reducing inter-cell aging divergence and maximizing the lifespan and reliable performance of the battery energy storage system.
2.3 Influence of Coolant Inlet Flow Rate
The third variable is an operational parameter: the volumetric flow rate of coolant through the bottom cold plate (in the bottom-only configuration for this parametric study). Flow rates from 5.0 L/min to 15.0 L/min were tested to identify the point of diminishing returns.
Thermal Impact: Increasing the flow rate improves both peak temperature and uniformity, but the benefit saturates. The relationship between flow rate, maximum temperature, and global temperature difference is captured in the following data and trend analysis.
| Flow Rate (L/min) | ΔTmax at 6480s (°C) | ΔTglobal at 6480s (°C) | Δp (kPa) | Ecool at 6480s (kJ) |
|---|---|---|---|---|
| 5.0 | 9.21 | 2.80 | 1.2 | ~5.9 |
| 7.5 | 8.31 | 2.09 | 2.7 | ~13.5 |
| 10.0 | 7.62 | 1.80 | 4.8 | ~23.5 |
| 12.5 | 7.15 | 1.55 | 7.4 | ~37.0 |
| 15.0 | 6.95 | 1.38 | 10.6 | ~53.0 |
The improvement in ΔTmax from 12.5 to 15.0 L/min is only 0.20°C, while the pressure drop and pumping energy increase by approximately 43% and 43%, respectively. This illustrates the law of diminishing returns: beyond a certain point, increased convective capability is offset by a dramatically thicker hydraulic boundary layer that is difficult to thin further, while the pressure drop, which is generally proportional to the square of the flow velocity ($\Delta p \propto v^2$), escalates rapidly. For this specific battery energy storage system configuration, a flow rate in the range of 10.0–12.5 L/min represents an optimal trade-off, providing effective cooling and good uniformity without excessive parasitic energy consumption.
2.4 Multi-Factor Synergistic Analysis
Finally, the optimal settings from the individual parametric studies were combined to evaluate the potential for synergistic performance gains. The combined configuration consisted of:
– Cold plate channels with W/H = 3.
– Bottom + Two Side cooling plate layout.
– A bottom plate coolant flow rate of 12.5 L/min (with proportional flow to side plates).
The performance of this multi-factor optimized design was compared against the single best-performing factor from earlier (the Bottom+Side layout with default W/H=5 channel and 10 L/min flow). The results are revealing.
| Configuration | ΔTmax at 6480s (°C) | ΔTglobal at 6480s (°C) | Primary Contributor |
|---|---|---|---|
| Optimal Single Factor (Bottom+Side Layout) | 0.51 | 3.00 | Cooling Strategy |
| Multi-Factor Optimized Design | 0.40 | 2.90 | All Factors |
| Difference | 0.11°C | 0.10°C | – |
| W/H=3 Only | ~7.30 | >8.70 | Minor |
| 12.5 L/min Only | ~7.15 | Minor |
The analysis leads to a critical conclusion: the cooling plate layout strategy (Bottom+Two Sides) is the overwhelmingly dominant factor in determining the thermal performance of this battery energy storage system. Its individual contribution accounts for the vast majority of the improvement in both maximum temperature and uniformity. The multi-factor optimization provides only a marginal further enhancement of about 0.1°C. This indicates that while channel geometry and flow rate fine-tuning are valuable for minimizing pressure drop and auxiliary energy use, the primary thermal challenge is addressed by fundamentally rethinking how and where cooling is applied to the battery pack.
3. Discussion and Implications for Battery Energy Storage System Design
The findings from this comprehensive numerical study offer clear directives for the engineering of thermal management systems in large-scale battery energy storage system applications. The primary objective—maintaining cell temperature within a safe, optimal range while ensuring exceptional uniformity—is most effectively achieved through a multi-pronged yet strategically prioritized approach.
First, the choice of cooling architecture is paramount. The traditional sole reliance on bottom cooling is insufficient for high-capacity prismatic cells, as it inevitably creates large vertical thermal gradients. Incorporating strategic side cooling, particularly targeting the upper regions of the cells where heat accumulates, is a highly effective method to “flatten” the temperature field. This study quantitatively demonstrates improvements exceeding 65% in global temperature uniformity, a factor directly linked to system longevity and performance consistency. For designers of battery energy storage systems, this implies a potential shift from simple cold plate designs to more integrated, three-dimensional cooling envelopes, even if it adds some complexity to the module assembly.
Second, while the cooling strategy dominates thermal outcomes, secondary factors like channel geometry and flow rate are crucial for system-level efficiency. An optimized channel with a lower aspect ratio (e.g., W/H=3) can reduce pressure drop by over 25% compared to common designs, directly lowering the parasitic load of the cooling pumps. This contributes to a higher overall round-trip efficiency for the battery energy storage system. Similarly, operating at a flow rate that balances cooling performance with pumping power (identified here as 10–12.5 L/min) is essential for economic operation. Oversizing pumps and operating at excessive flow rates incur an energy penalty with minimal thermal benefit.
The synergistic analysis further reinforces a hierarchical design philosophy: (1) Select the most effective cooling topology to manage core thermal loads and gradients. (2) Optimize the component details (like channel shape) to minimize the energy cost of delivering that cooling. (3) Tune operational parameters to the sweet spot on the performance-efficiency curve. For the battery energy storage system industry, this means that significant performance gains are available not from incremental tweaks to existing bottom-cooled designs, but from adopting more holistic cooling strategies.
4. Conclusion
This numerical investigation provides a detailed framework for analyzing and optimizing the thermal management of lithium-ion battery energy storage systems. Through validated modeling and systematic parametric studies, the work elucidates the distinct roles played by channel design, cooling layout, and flow rate. The key conclusions are:
- Channel aspect ratio has a minor influence on the final cell temperature but a major impact on system hydraulic losses. A channel with a width-to-height ratio (W/H) of 3 is recommended for its optimal balance of adequate heat transfer and low pressure drop.
- The cooling plate layout is the most critical design factor. A configuration combining bottom and side cooling plates drastically outperforms traditional bottom-only cooling. It reduces the maximum cell temperature rise by over 90% (from 7.6°C to 0.5°C) and improves overall pack temperature uniformity by approximately 66% (from 8.8°C to 3.0°C), fundamentally enhancing the safety and durability profile of the battery energy storage system.
- Coolant flow rate must be carefully selected. A moderate flow rate (10–12.5 L/min in this study) provides the majority of the cooling benefit. Higher flows yield diminishing thermal returns while disproportionately increasing pumping energy, making them inefficient for continuous operation.
- In a multi-factor optimization, the cooling layout strategy exerts a dominant effect on thermal performance. The combined optimal design (W/H=3, Bottom+Side layout, 12.5 L/min) delivers the best possible result, but its performance is virtually matched by the single implementation of the Bottom+Side layout. This underscores the primary importance of the macro cooling strategy over micro-optimizations in determining the thermal state of the system.
In summary, for engineers and developers of commercial and industrial battery energy storage systems, this research strongly advocates for the adoption of enhanced, multi-surface liquid cooling layouts. Such an approach, coupled with sensible component and operational optimization, is the most effective pathway to achieving the stringent thermal control required for the next generation of safe, long-lasting, and efficient grid-scale energy storage.