In this study, we propose a bidirectional counter-flow heat exchange plate for liquid cooling of energy storage batteries, aiming to address the challenges of excessive temperature and large temperature differences under high-rate charge/discharge conditions. Through numerical simulation, we compare three cooling configurations: bottom liquid cooling, side unidirectional liquid cooling, and side bidirectional counter-flow liquid cooling. The thermal management performance is evaluated under conventional (1 C), high (3 C), and ultra-high (5 C) charge/discharge rates. Our results demonstrate that the proposed bidirectional counter-flow plate achieves superior temperature uniformity and control, maintaining the maximum temperature below 299 K and the temperature difference within 4.8 K even under 3 C conditions. This work provides a reliable technical solution for the safe operation of grid-scale energy storage battery systems.
1. Introduction
As the global energy structure transitions toward low-carbon paradigms, renewable energy sources such as wind and solar power have become the core drivers of power system innovation. However, their inherent intermittency and distributed characteristics lead to unstable grid power quality. Energy storage technology, particularly lithium-ion battery-based electrochemical storage, has emerged as a key enabler for balancing supply and demand through “peak shaving” and “valley filling”. Nevertheless, the heat generated during charge/discharge cycles, exacerbated by internal resistance and chemical reactions, poses significant risks to the performance and safety of energy storage batteries. The optimal operating temperature range for lithium-ion batteries is 293–313 K, with a temperature uniformity requirement of within 5 K. Exceeding these limits can accelerate capacity degradation, cause thermal runaway, and compromise system reliability.
High-rate charging and discharging are frequently required in grid applications to rapidly store excess renewable energy or provide power during peak demand. Without effective thermal management, these conditions cause sharp temperature rises and large gradients. Among various cooling technologies, liquid cooling stands out due to its high heat transfer coefficient and stability. However, most existing studies focus on conventional low-rate conditions. In this work, we design a novel bidirectional counter-flow heat exchange plate that can effectively dissipate heat and maintain temperature uniformity even under extreme high-rate scenarios. By comparing three cooling configurations, we demonstrate the advantages of our proposed design for energy storage battery thermal management.
2. Numerical Model
2.1 Geometric Configuration
We construct a battery pack model consisting of 10 prismatic lithium iron phosphate (LFP) cells connected in series. Each cell has a capacity of 135 Ah, nominal voltage of 3.2 V, and dimensions of 945 mm × 14 mm × 90 mm. The anisotropic thermal conductivity of the cell is 18.3 W/(m·K) in the planar direction (along the electrode surface) and 1.1 W/(m·K) in the through-plane direction. The density is 1715 kg/m³, and the specific heat capacity is 1100 J/(kg·K).
Three cooling schemes are modeled:
- Scheme A: A conventional bottom liquid cooling plate with eight parallel straight channels. Coolant enters from the left inlet and exits from the right outlet. The plate is placed under the battery pack.
- Scheme B: Side-mounted unidirectional liquid cooling plates placed between adjacent cells. Coolant flows in a single direction from bottom to top.
- Scheme C: Side-mounted bidirectional counter-flow liquid cooling plates. Each plate has two inlets at the bottom and two outlets at the top, with coolant flowing in opposite directions to realize counter-flow heat exchange. An air gap is included between the plates for insulation.
2.2 Mathematical Model
The heat generation rate of the battery is calculated using the Bernardi model:
$$
q = \frac{I}{V_b} \left[ (E_0 – U) – T \frac{dE_0}{dT} \right] = \frac{1}{V_b} \left( I^2 R – I T \frac{dE_0}{dT} \right)
$$
where $V_b$ is the cell volume (m³), $E_0$ is the open-circuit voltage (V), $U$ is the terminal voltage (V), $T$ is the temperature (K), $I$ is the current (A), and $R$ is the internal resistance (Ω). The anisotropic heat conduction within the cell is described by:
$$
\rho C_p \frac{\partial T}{\partial t} = \lambda_x \frac{\partial^2 T}{\partial x^2} + \lambda_y \frac{\partial^2 T}{\partial y^2} + \lambda_z \frac{\partial^2 T}{\partial z^2} + q
$$
where $\rho$ is density, $C_p$ is specific heat, and $\lambda_x,\lambda_y,\lambda_z$ are thermal conductivities in the three orthogonal directions.
The coolant flow is governed by the continuity, momentum, and energy equations:
$$
\frac{\partial \mathbf{v}}{\partial t} + (\mathbf{v} \cdot \nabla) \mathbf{v} = -\frac{\nabla p}{\rho_f} + \frac{\mu}{\rho_f} \nabla^2 \mathbf{v} + \mathbf{g}
$$
$$
\frac{\partial \rho_f}{\partial t} + \nabla \cdot (\rho_f \mathbf{v}) = 0
$$
$$
\frac{\partial}{\partial t}(\rho_f C_{p,f} T_f) + \nabla \cdot ( -k_f \nabla T_f) + \rho_f C_{p,f} \mathbf{v} \cdot \nabla T_f = 0
$$
where $\mathbf{v}$ is velocity, $p$ is pressure, $\rho_f$ is coolant density, $\mu$ is dynamic viscosity, $C_{p,f}$ is coolant specific heat, $k_f$ is coolant thermal conductivity, and $T_f$ is coolant temperature. The energy conservation for the solid heat exchange plate is:
$$
\frac{\partial}{\partial t}(\rho_s C_{p,s} T_s) + \nabla \cdot ( -k_s \nabla T_s) = 0
$$
2.3 Boundary Conditions and Mesh Independence
The initial temperature of the battery pack and ambient is 298 K. The coolant inlet is set to a velocity inlet condition with a total flow rate of 12 L/min and an inlet temperature of 293.15 K. The coolant is 50% ethylene glycol solution. Outlet boundary condition is pressure outlet. Three heat generation rates corresponding to 1 C, 3 C, and 5 C are applied as volumetric heat sources, with parameters derived from literature.
A mesh independence study is performed using six mesh densities from 1.33 million to 5.13 million elements. The difference in pressure drop between the 4.23 million and 5.13 million grids is less than 0.2% (14.90 kPa vs 14.93 kPa). Therefore, the mesh with 4.23 million elements is adopted for all subsequent simulations.
3. Results and Discussion
3.1 Conventional Rate (1 C) Case
Under 1 C discharge, all three schemes maintain the maximum battery temperature below 301 K. However, Scheme A (bottom cooling) exhibits a maximum temperature difference of 6.5 K, exceeding the 5 K requirement, primarily due to the large vertical temperature gradient. In contrast, both side-mounted schemes (B and C) achieve excellent uniformity, with temperature differences of 1.5 K and 1.0 K, respectively. The results are summarized in Table 1.
| Scheme | Max Temperature (K) | Temperature Difference (K) |
|---|---|---|
| A (Bottom) | 301.0 | 6.5 |
| B (Side Unidirectional) | 294.7 | 1.5 |
| C (Side Bidirectional Counter-flow) | 294.0 | 1.0 |
3.2 High Rate (3 C) Case
At 3 C, the performance gap becomes significant. Scheme A produces a maximum temperature of 357 K, approaching the thermal runaway risk threshold of 393 K, and a temperature difference of 52 K, making it completely unacceptable. Scheme B reduces the maximum temperature to 307 K but still has a 13 K difference, which is beyond the 5 K limit. Scheme C, however, maintains the maximum temperature at only 299 K and a temperature difference of 4.8 K, meeting both criteria. Detailed data are shown in Table 2.
| Scheme | Max Temperature (K) | Temperature Difference (K) |
|---|---|---|
| A (Bottom) | 357 | 52 |
| B (Side Unidirectional) | 307 | 13 |
| C (Side Bidirectional Counter-flow) | 299 | 4.8 |
The superior performance of Scheme C stems from the counter-flow design. In the bidirectional heat exchange plate, the coolant in the two channels flows in opposite directions, creating a thermal compensation effect that prevents heat accumulation along the flow path. The air gap further reduces thermal cross-talk between adjacent batteries.
3.3 Ultra-High Rate (5 C) Case
Under extreme 5 C conditions, Scheme A fails catastrophically, with maximum temperature reaching 470 K, far exceeding the safety limit. Scheme B still yields a maximum temperature of 332 K and a temperature difference of 39 K, which is unacceptable for long-term operation but may be tolerated during short-duration events. Scheme C achieves a maximum temperature of 308 K and a temperature difference of 14 K. Although the difference exceeds 5 K, the maximum temperature remains well below the thermal runaway threshold. Table 3 summarizes the results.
| Scheme | Max Temperature (K) | Temperature Difference (K) |
|---|---|---|
| A (Bottom) | 470 | 107 |
| B (Side Unidirectional) | 332 | 39 |
| C (Side Bidirectional Counter-flow) | 308 | 14 |
A comprehensive comparison of the three schemes across all rates is presented in Figure 1 (visualized through the data tables). The bidirectional counter-flow design consistently provides the best thermal management performance for energy storage batteries.
4. Heat Transfer Analysis of the Heat Exchange Plate
To understand the mechanism behind the superior performance, we analyze the coolant temperature distribution within the heat exchange plates at 3 C. In Scheme A, the coolant temperature rises by 15 K from inlet to outlet, with a pronounced hot spot in the central region due to flow maldistribution and limited heat transfer from the battery top. In Scheme B, the temperature rise is about 10 K, and the temperature gradient along the flow direction is clearly visible. In Scheme C, however, the bidirectional counter-flow design reduces the coolant temperature variation to less than 1 K across the entire plate. The two streams exchange heat with each other, effectively equalizing the temperature.
The air gap between the heat exchange plates further prevents heat transfer between adjacent cells, ensuring that the thermal behavior of each energy storage battery is independent. This design is particularly beneficial for large battery packs where thermal runaway propagation must be avoided.
We further quantify the cooling efficiency using the dimensionless Nusselt number and effectiveness-NTU method. For the bidirectional counter-flow plate, the heat transfer effectiveness is defined as:
$$
\varepsilon = \frac{T_{c,out} – T_{c,in}}{T_{b,max} – T_{c,in}}
$$
where $T_{c,in}$ and $T_{c,out}$ are the coolant inlet and outlet temperatures, and $T_{b,max}$ is the maximum battery temperature. At 3 C, the effectiveness of Scheme C is 0.85, compared to 0.52 for Scheme B and 0.21 for Scheme A. This demonstrates the significant improvement in heat exchange efficiency.
The thermal resistance network for the energy storage battery and cooling system can be expressed as:
$$
R_{total} = \frac{1}{h_{conv} A} + \frac{\delta}{k A} + R_{contact}
$$
where $h_{conv}$ is the convective heat transfer coefficient, $A$ is the contact area, $\delta$ is the thickness of the heat exchange plate wall, $k$ is the thermal conductivity of the plate, and $R_{contact}$ accounts for the thermal grease layer. In Scheme C, the large effective contact area (both sides of the cell) and the counter-flow arrangement increase the effective heat transfer coefficient, reducing the total thermal resistance.
Table 4 summarizes the key thermal performance parameters for the three schemes at 3 C.
| Parameter | Scheme A | Scheme B | Scheme C |
|---|---|---|---|
| Coolant temperature rise (K) | 15.0 | 10.0 | 1.0 |
| Effectiveness $\varepsilon$ | 0.21 | 0.52 | 0.85 |
| Contact area (m² per cell) | 0.085 | 0.170 | 0.340 |
| Pressure drop (kPa) | 8.2 | 14.9 | 15.3 |
| Pumping power (W) | 1.64 | 2.98 | 3.06 |
The slightly higher pressure drop and pumping power of Scheme C are justified by the dramatic improvement in temperature control. For grid-scale energy storage battery systems, the safety and longevity benefits far outweigh the marginal increase in parasitic power consumption.
5. Conclusion
In this work, we developed a bidirectional counter-flow heat exchange plate for liquid cooling of energy storage batteries and evaluated its thermal management performance through numerical simulation. Three cooling configurations were compared under 1 C, 3 C, and 5 C charge/discharge rates. The key conclusions are as follows:
- The traditional bottom-mounted liquid cooling plate (Scheme A) fails to meet the temperature uniformity requirement of 5 K even at 1 C, and becomes completely inadequate under high-rate conditions.
- The side-mounted unidirectional liquid cooling plate (Scheme B) can control maximum temperature at low and moderate rates, but the temperature difference exceeds 5 K at 3 C and 5 C, leading to accelerated degradation.
- The proposed side-mounted bidirectional counter-flow liquid cooling plate (Scheme C) successfully maintains both the maximum temperature and temperature difference within acceptable limits for all tested rates: 299 K and 4.8 K at 3 C, and 308 K and 14 K at 5 C. The design achieves this through symmetric flow channels that create a thermal compensation effect, along with an air gap for thermal isolation.
The bidirectional counter-flow heat exchange plate exhibits strong adaptability across a wide range of charge/discharge rates, making it an ideal candidate for grid-level energy storage battery systems that must operate reliably under extreme conditions. Future work will focus on experimental validation and optimization of channel geometry for further performance improvements.