Electric vehicle battery performance is strongly dependent on temperature. Both high and low temperature extremes can degrade capacity, accelerate aging, and even cause safety hazards. To ensure reliable operation across a wide range of ambient conditions, a thermal management system (BTMS) must provide efficient cooling, uniform temperature distribution, and rapid preheating when needed. In this work, I propose a novel BTMS based on a honeycomb-like structure that couples a mini-channel liquid cooling network with a composite phase change material (PCM). The system is designed to achieve integrated cooling, thermal insulation, and preheating for a cylindrical lithium-ion battery pack.
The lithium-ion battery is the core energy storage component of electric vehicles. Under high discharge rates, the heat generation rate can increase rapidly. If this heat is not removed effectively, the maximum temperature of the electric vehicle battery pack may exceed the safe limit, and local hot spots can cause thermal runaway. On the other hand, at sub-zero temperatures, the electrolyte conductivity drops and the internal resistance rises, which reduces the available capacity and may induce lithium plating. Therefore, I have developed a numerical model to evaluate both cooling and preheating performance of the proposed system. In this paper, I systematically investigate the effects of channel number, coolant flow velocity, graphite mass fraction in the composite PCM, and flow arrangement on the thermal behavior of the electric vehicle battery pack.
Model of the Electric Vehicle Battery and the Proposed BTMS
The battery under study is a commercial 18650 cylindrical lithium-ion cell with a nominal capacity of 1.5 Ah and a nominal voltage of 3.2 V. The cell geometry is simplified to a homogeneous cylinder with a diameter of 18 mm and a height of 65 mm. The average thermophysical properties are calculated by volume weighting of the materials inside the cell. The values are listed in the following table.
| Property | Value |
|---|---|
| Density | 2180 kg/m³ |
| Specific heat capacity | 1066 J/(kg·K) |
| Radial thermal conductivity | 1.1 W/(m·K) |
| Circumferential thermal conductivity | 1.1 W/(m·K) |
| Axial thermal conductivity | 12.5 W/(m·K) |
I adopted the simplified Bernardi model to compute the volumetric heat generation rate of the electric vehicle battery. The heat generation during discharge is expressed as
$$ q = \frac{I}{V_0} \left[ (E_{OC} – E) + T \frac{\partial E_{OC}}{\partial T} \right] $$
where \(I\) is the current, \(V_0\) is the cell volume, \(E_{OC}\) is the open-circuit voltage, \(E\) is the operating voltage, and \(T\) is the temperature. The term \(\frac{\partial E_{OC}}{\partial T}\) is related to the entropy change and is treated as a constant for a given cell chemistry. The ohmic and polarization resistances are combined as a function of temperature:
$$ R_{\Omega} + R_P = -0.0001 T^3 + 0.0134 T^2 – 0.5345 T + 12.407 $$
At an ambient temperature of 35 °C, this equation gives heat generation rates of 5362 W/m³ at 1C, 30586 W/m³ at 2C, 60156 W/m³ at 3C, and 126758 W/m³ at 4C discharge rates. I validated the heat generation model by comparing the simulated surface temperature rise with an experimental reference. The maximum difference was only 0.64 °C, which confirms the accuracy of my model.
The proposed BTMS consists of a 4×4 staggered array of cylindrical cells embedded in a composite PCM. The whole assembly is placed inside an aluminum frame with vertical mini-channels of 1 mm diameter. The distance between the frame and the cell surface is 1.25 mm. The height of the frame is 65 mm and its thickness is 1.5 mm. The baseline design contains 63 channels, one at the center of each side of the hexagonal cells. The PCM is a paraffin/expanded graphite composite. The temperature of the phase change is 37 °C and the latent heat depends on the graphite mass fraction.
Governing Equations for Fluid and PCM
The coolant (water) flow in the mini-channels is assumed to be laminar and incompressible. The mass, momentum, and energy conservation equations are:
$$ \frac{\partial \rho_w}{\partial t} + \nabla \cdot (\rho_w \mathbf{v}) = 0 $$
$$ \frac{\partial (\rho_w \mathbf{v})}{\partial t} + \nabla \cdot (\rho_w \mathbf{v} \mathbf{v}) = -\nabla p + \nabla \cdot (\mu \nabla \mathbf{v}) $$
$$ \frac{\partial (\rho_w c_{p,w} T)}{\partial t} + \nabla \cdot (\rho_w c_{p,w} \mathbf{v} T) = \nabla \cdot (\lambda_w \nabla T) $$
For the PCM, I used the enthalpy-porosity method. The energy equation is
$$ \frac{\partial (\rho_{PCM} H)}{\partial t} = \nabla \cdot (\lambda_{PCM} \nabla T) $$
where the total enthalpy \(H\) is the sum of sensible enthalpy \(H_1\) and latent enthalpy \(\Delta H\):
$$ H = H_1 + \Delta H $$
$$ H_1 = \int_{T_0}^{T} c_{p,PCM} dT $$
$$ \Delta H = \beta \gamma $$
Here, \(\gamma\) is the latent heat of fusion and \(\beta\) is the liquid fraction, defined as
$$ \beta = \begin{cases} 0 & T < T_s \\ \frac{T – T_s}{T_l – T_s} & T_s \le T \le T_l \\ 1 & T > T_l \end{cases} $$
The numerical simulations were performed using the ANSYS Fluent software. The boundary conditions include a velocity inlet, a pressure outlet, and convective heat transfer on the outer surfaces. The temperature-dependent resistances were included through user-defined functions. I carried out grid independence tests with four mesh sizes. The final mesh size was about 1.97 million cells, which offered a good balance between accuracy and computational cost. The time step was set to 2 s with 20 iterations per step.
Cooling Performance of the Electric Vehicle Battery
Comparison of PCM-only and PCM-microchannel cooling
To highlight the advantage of the coupled system, I first compared a pure PCM cooling scheme with the proposed PCM-microchannel liquid cooling scheme. The ambient temperature was 35 °C and the battery discharge rate was 4C. The inlet velocity of water in the mini-channels was 0.1 m/s and the inlet temperature was 35 °C. The results show that the maximum temperature after complete discharge was 41.98 °C for the PCM-only system and 39.83 °C for the coupled system. The maximum temperature difference decreased from 3.50 °C to 2.95 °C. The liquid fraction of the PCM was 0.90 in the pure PCM case and 0.33 in the coupled case, which indicates that a significant portion of the heat was carried away by the coolant instead of being stored in the PCM. Therefore, the coupled system is more effective for high-rate discharge of an electric vehicle battery pack.
| Scheme | Maximum temperature (°C) | Temperature difference (°C) | Liquid fraction |
|---|---|---|---|
| PCM only | 41.98 | 3.50 | 0.90 |
| PCM + microchannels | 39.83 | 2.95 | 0.33 |
Effect of the number of channels
With a constant total flow rate, I designed three channel configurations: Case 1 with 63 channels, Case 2 with 111 channels, and Case 3 with 174 channels. The inlet velocities were adjusted to maintain the same mass flow rate. At the end of discharge, the maximum temperatures were 39.83 °C, 39.82 °C, and 40.09 °C, respectively. The temperature differences were 2.94 °C, 3.14 °C, and 3.51 °C. Increasing the number of channels enlarged the heat transfer area, but it also reduced the velocity in each channel, weakening the local convective heat transfer. Consequently, the case with the smallest number of channels still offered the best overall cooling performance. The liquid fraction data followed the same trend, confirming that a moderate channel count is preferable.
| Case | Number of channels | Inlet velocity (m/s) | Maximum temperature (°C) | Temperature difference (°C) |
|---|---|---|---|---|
| Case 1 | 63 | 0.100 | 39.83 | 2.94 |
| Case 2 | 111 | 0.0567 | 39.82 | 3.14 |
| Case 3 | 174 | 0.0362 | 40.09 | 3.51 |
Effect of coolant flow velocity
Using the Case 1 geometry, I tested inlet velocities of 0.05 m/s, 0.1 m/s, and 0.2 m/s. The maximum temperature after discharge was 40.25 °C, 39.83 °C, and 39.63 °C, respectively. The temperature difference increased slightly with velocity during most of the discharge process, but after about 810 s the differences became small. A higher flow velocity enhances the heat transfer coefficient, but it also causes a faster temperature rise near the inlet, which increases the local thermal gradient. Increasing the velocity from 0.1 m/s to 0.2 m/s only reduced the maximum temperature by about 0.2 °C, while doubling the pumping power. Thus, 0.1 m/s is a reasonable compromise for the electric vehicle battery cooling application.
Effect of graphite mass fraction in the composite PCM
The composite PCM was prepared by mixing paraffin with expanded graphite. The thermal conductivity increases significantly with the graphite content, while the latent heat and specific heat decrease. I evaluated seven mass fractions: 0%, 2%, 5%, 8%, 10%, 15%, and 20%. The thermophysical properties of the composite PCM are listed in the table below.
| Graphite fraction (%) | Thermal conductivity (W/(m·K)) | Density (kg/m³) | Specific heat (J/(kg·K)) | Latent heat (J/g) | Phase change temperature (°C) |
|---|---|---|---|---|---|
| 0 | 0.245 | 800 | 2300 | 200.66 | 37 |
| 2 | 0.324 | 800 | 2260 | 196.65 | 37 |
| 5 | 0.907 | 800 | 2200 | 191.37 | 37 |
| 8 | 1.374 | 800 | 2140 | 184.61 | 37 |
| 10 | 1.687 | 800 | 2100 | 180.60 | 37 |
| 15 | 2.499 | 800 | 2000 | 170.56 | 37 |
| 20 | 3.355 | 800 | 1900 | 160.53 | 37 |
The simulation results for the maximum temperature show a monotonic decrease with increasing graphite fraction. At 0% graphite, the maximum temperature was about 40.9 °C, while at 20% it was slightly above 39.4 °C. The temperature difference, however, increased slightly as the graphite fraction increased, because the enhanced thermal conductivity promotes heat transfer from the cell surface to the frame, but the core temperature remains higher. The maximum difference between the 0% and 20% cases was only 0.23 °C, so the temperature uniformity is not seriously affected. The liquid fraction decreased with increasing graphite percentage, since more heat was conducted to the channels. From a safety perspective, a higher graphite content increases the electrical conductivity of the PCM, which increases the risk of short circuits. Therefore, 5% graphite was selected as the optimal mass fraction for the electric vehicle battery thermal management system.
Effect of inlet/outlet flow arrangement
In the baseline design, all channels have the same flow direction from the top to the bottom. This creates a temperature gradient along the flow direction. To improve uniformity, I proposed a counter-flow arrangement where adjacent channels have opposite flow directions: half of the channels carry water from top to bottom, and the other half carry water from bottom to top. I compared the two arrangements at the same inlet velocity of 0.1 m/s. The results showed that the counter-flow arrangement reduces the maximum temperature from 39.83 °C to 39.80 °C, and more importantly, the temperature difference drops from 3.15 °C to 2.77 °C. The liquid fraction increases from 0.25 to 0.29, indicating that the PCM is more uniformly utilized. The counter-flow pattern creates a more symmetric temperature field, which is beneficial for the electric vehicle battery pack in both safety and durability.
| Flow arrangement | Maximum temperature (°C) | Temperature difference (°C) | Liquid fraction |
|---|---|---|---|
| Co-current | 39.83 | 3.15 | 0.25 |
| Counter-current | 39.80 | 2.77 | 0.29 |
Preheating Performance of the Electric Vehicle Battery
In cold climates, the electric vehicle battery must be preheated before charging or discharging. The proposed BTMS can also operate as a preheater by circulating warm water through the microchannels. The heat is transferred from the water through the aluminum frame and the composite PCM to the cells. I evaluated the preheating performance by measuring the time required for the lowest cell surface temperature to reach 20 °C from an initial temperature of -20 °C. The ambient temperature was also -20 °C. The inlet water temperature was 25 °C unless otherwise specified. I used the rate of temperature rise (RTR) as an evaluation metric:
$$ RTR = \frac{20 – (-20)}{\tau} $$
where \(\tau\) is the preheating time in minutes.
Thermal insulation capability
Before studying the active preheating process, I evaluated the passive thermal insulation capability of the PCM. The battery module was initially at 50 °C and placed in an environment at -10 °C. I monitored the maximum temperature for 10 hours. Without PCM, the battery temperature dropped below 0 °C within about 1 hour. With pure paraffin (0% graphite), the temperature remained above 0 °C for more than 8 hours. As the graphite fraction increased, the insulation time decreased, because the higher thermal conductivity allowed heat to escape more quickly to the surroundings. This demonstrates that the composite PCM can serve as a thermal buffer, which is essential for maintaining the electric vehicle battery in a warm state during short parking periods.
Influence of PCM composition on preheating
I simulated seven different graphite mass fractions for the preheating process. The preheating time and the final temperature difference are shown below.
| Graphite fraction (%) | Preheating time (s) | Temperature difference (°C) | RTR (°C/min) |
|---|---|---|---|
| 0 | 405 | 1.93 | 5.93 |
| 2 | 360 | 2.30 | 6.67 |
| 5 | 272 | 2.85 | 8.82 |
| 8 | 256 | 3.00 | 9.38 |
| 10 | 250 | 3.10 | 9.60 |
| 15 | 240 | 3.15 | 10.00 |
| 20 | 235 | 3.22 | 10.21 |
Increasing the graphite fraction from 0% to 5% reduced the preheating time by 133 s, while increasing it from 5% to 20% only reduced the time by another 37 s. The temperature difference increased gradually with the graphite fraction, but it remained below 3.5 °C for all cases. The improvement in preheating speed diminishes significantly beyond 5%, so I selected 5% as the optimal graphite mass fraction for the entire BTMS.
Influence of preheating fluid velocity
With the 5% graphite composite PCM and the Case 1 channel layout, I tested inlet velocities of 0.05, 0.1, 0.25, 0.5, and 1 m/s. The results are summarized below.
| Inlet velocity (m/s) | Preheating time (s) | Temperature difference (°C) | RTR (°C/min) |
|---|---|---|---|
| 0.05 | 380 | 2.10 | 6.32 |
| 0.1 | 272 | 2.85 | 8.82 |
| 0.25 | 211 | 3.32 | 11.37 |
| 0.5 | 186 | 3.53 | 12.90 |
| 1 | 169 | 3.70 | 14.20 |
The preheating time decreases as the velocity increases, but the benefit becomes smaller. Raising the velocity from 0.05 to 0.25 m/s shortens the time by 169 s, while raising it from 0.25 to 1 m/s only shortens it by 42 s. At the same time, the temperature difference grows with velocity because the inlet-side cells are heated more intensely. To balance preheating speed, temperature uniformity, and pumping cost, I chose 0.25 m/s as the optimal preheating fluid velocity for the electric vehicle battery pack.
Influence of the number of channels on preheating
I also examined the effect of the channel count while keeping the total flow rate constant. The inlet velocities for Case 1, Case 2, and Case 3 were 0.25 m/s, 0.1418 m/s, and 0.0905 m/s, respectively. The preheating times were 211 s, 192 s, and 181 s. The temperature differences were 3.32 °C, 3.65 °C, and 4.50 °C. More channels reduce the preheating time because the heat transfer area is enlarged. However, the temperature difference also increases, since the additional channels enhance the heat delivery near the inlet region. The Case 2 layout offers a compromise between a short preheating time and acceptable temperature uniformity. Therefore, I selected Case 2 as the final structure for the BTMS.
Influence of flow direction on preheating
I compared co-current and counter-current arrangements for the preheating process using the Case 2 structure. The counter-current design has adjacent channels with opposite flow directions. The preheating time for the counter-current case was 194 s, just 2 s longer than the co-current case. However, the temperature difference was reduced from 3.65 °C to 3.25 °C, and the temperature distribution in the aluminum frame and the cells became much more uniform. The counter-current arrangement is therefore beneficial to the temperature consistency of the electric vehicle battery, and the slight increase in preheating time is acceptable.
Influence of preheating fluid temperature
I finally investigated the effect of the preheating fluid temperature on the preheating performance. The ambient temperatures were -40, -30, -20, and -10 °C, while the inlet water temperatures were 25, 35, 40, and 45 °C. The results show that increasing the inlet water temperature drastically shortens the preheating time. For example, at an ambient temperature of -40 °C, raising the water temperature from 25 °C to 35 °C reduced the preheating time from 223 s to 146 s. However, the temperature difference also increased dramatically. When the inlet temperature was above 37 °C, the PCM began to melt, and the latent heat absorption helped to slow down the temperature rise of the cells. As a result, the reduction in preheating time became less pronounced, and the rate of increase in the temperature difference was mitigated. The phase change of the PCM thus serves as a natural buffer that prevents excessive thermal gradients during preheating.
| Ambient temperature (°C) | Inlet temperature 25 °C time (s) | Inlet temperature 35 °C time (s) | Inlet temperature 40 °C time (s) | Inlet temperature 45 °C time (s) |
|---|---|---|---|---|
| -40 | 223 | 146 | 119 | 101 |
| -30 | 208 | 135 | 110 | 94 |
| -20 | 192 | 125 | 102 | 86 |
| -10 | 175 | 114 | 94 | 79 |
Although a higher preheating fluid temperature reduces the preheating time, it also increases the maximum temperature difference across the electric vehicle battery pack. Therefore, the fluid temperature should be chosen based on the specific operating requirements. For normal cold-climate preheating, 25 °C water is a safe and effective choice, while 35 °C can be used when faster preheating is required and the temperature uniformity is less critical.
Conclusions
In this work, I developed and numerically evaluated a microchannel-PCM coupled BTMS for electric vehicle battery packs. The main conclusions are as follows:
- The coupled system provides better cooling than a PCM-only system, reducing the maximum temperature by 2.15 °C and the temperature difference by 0.55 °C during a 4C discharge at 35 °C ambient temperature.
- For cooling, the optimal number of channels is 63 (baseline Case 1) when the total flow rate is fixed, because additional channels reduce the per-channel velocity and slightly degrade the temperature uniformity.
- The optimal coolant velocity for cooling is 0.1 m/s, since velocities above this value yield only marginal temperature reductions but significantly increase pumping power.
- A graphite mass fraction of 5% in the composite PCM is recommended for both cooling and preheating. This value provides a good balance between thermal conductivity, latent heat, safety, and cost.
- A counter-current flow arrangement improves temperature uniformity both during cooling and preheating, with almost no penalty on the preheating time.
- For preheating, the optimal fluid velocity is 0.25 m/s, and the Case 2 channel configuration offers the best trade-off between preheating speed and temperature uniformity.
- The PCM provides a significant thermal insulation effect. Pure paraffin can keep the electric vehicle battery above 0 °C for several hours in a -10 °C environment, while composite PCMs with high graphite content reduce the insulation time.
- Increasing the preheating fluid temperature accelerates preheating but also increases the temperature difference. The PCM phase change acts as a thermal buffer that moderates the gradient when the fluid temperature exceeds the phase change temperature.
The proposed honeycomb-like microchannel-PCM coupled thermal management system is a promising solution for electric vehicle battery thermal management. It integrates high-temperature cooling, low-temperature preheating, and passive thermal insulation in a single compact structure. Future work should consider dynamic operating conditions and advanced optimization algorithms to further improve the energy efficiency and adaptability of the system.