In my research, I investigated a coupled spray and ventilation cooling method for an electric vehicle battery pack. The growing use of electric vehicles has made the thermal management of the electric vehicle battery pack one of the most important factors limiting performance, safety, and service life. Lithium-ion cells have high energy density, low memory effect, and long cycle life, but they still generate substantial heat during high-rate charging and discharging. If this heat is not removed efficiently, the temperature of the electric vehicle battery pack can rise rapidly, accelerating degradation and increasing the risk of thermal runaway. My objective was to combine the simplicity of air cooling with the high latent heat of water evaporation, thereby creating a cooling strategy that can reduce both the maximum temperature and the temperature difference inside an electric vehicle battery pack.
I selected prismatic lithium-ion cells because they offer high space utilization, stable mechanical behavior, and good grouping efficiency. In my numerical model, I arranged 24 prismatic cells in a module and injected fine water droplets directly into the airflow channels between the cells. The droplets absorbed heat from both the air and the cell surfaces, and the resulting vapor was carried away by the ventilation stream. I used experiments to obtain the heat generation rate of the cells, built a computational fluid dynamics model for the coupled spray and ventilation process, and validated the model against measured temperatures. After validation, I studied how inlet velocity, inlet temperature, nozzle type, nozzle spacing, spray mass flow rate, droplet diameter, cell spacing, discharge rate, ambient temperature, and ambient humidity affect the cooling performance of the electric vehicle battery pack.
Thermal Characteristics of the Electric Vehicle Battery Pack
I first characterized the thermal behavior of the selected prismatic cell. The cell had a nominal voltage of 3.2 V, a rated capacity of 27 Ah, a mass of 0.588 kg, and dimensions of 20 mm by 100 mm by 140 mm. I placed the cell in a constant-temperature chamber at 300 K and measured its surface temperature using five thermocouples connected to a data acquisition unit. I tested the cell at 1C, 2C, and 3C discharge rates. The temperature rose gradually at the beginning of discharge and increased more sharply near the end, especially at higher discharge rates. This behavior confirmed that the heat generation rate of the electric vehicle battery pack is strongly dependent on the discharge current.
| Cell parameter | Value |
|---|---|
| Cell type | Prismatic lithium-ion |
| Dimensions | 20 mm x 100 mm x 140 mm |
| Mass | 0.588 kg |
| Nominal voltage | 3.2 V |
| Operating voltage | 2.0 V to 3.65 V |
| Rated capacity | 27 Ah |
| AC resistance | 0.8 mΩ to 1.2 mΩ |
| DC resistance | No more than 3.0 mΩ |
I used a simplified heat generation model for the electric vehicle battery pack. The total heat generation can be written as the sum of reversible reaction heat, Joule heat, polarization heat, and side reaction heat. For lithium-ion cells, side reactions are usually negligible during normal operation. I therefore focused on the main irreversible and reversible contributions. The general energy balance can be expressed as
$$Q_{gen}=I\left(U_{OC}-U\right)-IT_b\frac{dU_{OC}}{dT_b}$$
where I is the discharge current, U_OC is the open-circuit voltage, U is the terminal voltage, T_b is the cell temperature, and dU_OC/dT_b is the entropic coefficient. The first term represents irreversible heat caused by internal resistance and polarization, while the second term represents reversible entropic heat. I also used the equivalent form
$$Q_{gen}=I^2R_j+IT_b\frac{\partial U_{OC}}{\partial T_b}$$
where R_j is the equivalent internal resistance. By fitting the measured temperature rise data, I obtained a practical heat generation correlation for the electric vehicle battery pack:
$$Q_{gen}=1.06\times10^{-3}I^2+0.085I$$
For a uniformly heated cell, the volumetric heat generation rate becomes
$$q_{gen}=3.42I^2+274.19I$$
The resulting volumetric heat source values used in my simulations are summarized below.
| Discharge rate | Volumetric heat source |
|---|---|
| 1C | 9893 W/m³ |
| 2C | 24777 W/m³ |
| 3C | 44626 W/m³ |
These values formed the thermal input for my later simulations. I treated each cell as a homogeneous heat source because the internal layered structure of the electric vehicle battery pack would require extremely fine meshes and large computational resources. This simplification was reasonable for studying module-level cooling, where the main concern is the temperature of the electric vehicle battery pack rather than the local temperature inside each electrode stack.
Numerical Model for Spray and Ventilation Coupling
I developed a three-dimensional numerical model in which air was treated as the continuous phase and water droplets were treated as the discrete phase. The airflow and droplet motion were coupled through source terms in the mass, momentum, energy, and species equations. This Eulerian-Lagrangian approach allowed me to capture droplet evaporation, heat absorption, and vapor transport inside the electric vehicle battery pack. I used the realizable k-epsilon turbulence model because it performs well for jets, recirculation, and complex channel flows. I also used an enhanced wall treatment to resolve the near-wall heat transfer without excessive mesh refinement.
The continuous-phase mass conservation equation was
$$\frac{\partial \rho}{\partial t}+\nabla\cdot\left(\rho\vec{v}\right)=S_m$$
where S_m is the mass source term caused by droplet evaporation. The momentum equation was
$$\frac{\partial\left(\rho\vec{v}\right)}{\partial t}+\nabla\cdot\left(\rho\vec{v}\vec{v}\right)=-\nabla p+\nabla\cdot\tau+\rho\vec{g}+\vec{F}$$
where τ is the stress tensor, ρg is the gravitational body force, and F represents additional forces. The energy equation was
$$\frac{\partial\left(\rho E\right)}{\partial t}+\nabla\cdot\left[\vec{v}\left(\rho E+p\right)\right]=\nabla\cdot\left[k_{eff}\nabla T-\sum_j h_j\vec{J}_j\right]+\Phi+S_h$$
where k_eff is the effective thermal conductivity, E is total energy, h_j is the enthalpy of species j, J_j is the diffusion flux, Φ is viscous dissipation, and S_h is the energy source term from droplet evaporation and heat exchange. The species transport equation was
$$\frac{\partial\left(\rho Y_m\right)}{\partial t}+\nabla\cdot\left(\rho\vec{v}Y_m\right)=\nabla\cdot\left(\rho D_m\nabla Y_m\right)+S_i$$
where Y_m is the mass fraction of species m, D_m is the diffusion coefficient, and S_i is the source term due to water vapor production by droplet evaporation.
For turbulence, I used the realizable k-epsilon model. The turbulent kinetic energy equation was
$$\frac{\partial\left(\rho k\right)}{\partial t}+\frac{\partial\left(\rho k u_i\right)}{\partial x_i}=\frac{\partial}{\partial x_j}\left[\left(\mu+\frac{\mu_t}{\sigma_k}\right)\frac{\partial k}{\partial x_j}\right]+P_k+G_b-\rho\epsilon-Y_M$$
and the dissipation rate equation was
$$\frac{\partial\left(\rho\epsilon\right)}{\partial t}+\frac{\partial\left(\rho\epsilon u_i\right)}{\partial x_i}=\frac{\partial}{\partial x_j}\left[\left(\mu+\frac{\mu_t}{\sigma_\epsilon}\right)\frac{\partial\epsilon}{\partial x_j}\right]+C_1S\epsilon-C_2\rho\frac{\epsilon^2}{k+\sqrt{\nu\epsilon}}+C_3\frac{G_b\epsilon}{k}$$
These equations allowed me to capture the turbulent mixing of air and vapor inside the electric vehicle battery pack.
For the discrete phase, I solved the droplet momentum equation:
$$\frac{d\vec{v}_p}{dt}=F_D\left(\vec{v}-\vec{v}_p\right)+\frac{\rho-\rho_p}{\rho_p}\vec{g}+\vec{F}$$
where the drag force was modeled using the Schiller-Naumann correlation for spherical droplets:
$$C_D=\frac{24}{Re}\left(1+0.15Re^{0.687}\right)$$
The droplet Reynolds number was
$$Re=\frac{\rho_p d_p\left|\vec{v}-\vec{v}_p\right|}{\mu}$$
For droplet heating before evaporation, I used the heat balance equation
$$T_p\left(t+\Delta t\right)=\alpha_p+\left[T_p\left(t\right)-\alpha_p\right]e^{-\beta_p\Delta t}$$
where
$$\alpha_p=\frac{hT_\infty+\epsilon_p\sigma\theta_R^4}{h+\epsilon_p\sigma T_p^3\left(t\right)}$$
and
$$\beta_p=\frac{A_p\left(h+\epsilon_p\sigma T_p^3\left(t\right)\right)}{m_pc_p}$$
When the droplet temperature exceeded the evaporation temperature but remained below the boiling point, I calculated the mass loss by
$$m_p\left(t+\Delta t\right)=m_p\left(t\right)-N_{H2O}A_pM_{w,H2O}\Delta t$$
The molar flux of water vapor was calculated from
$$N_{H2O}=k_{H2O}\left(C_{H2O,s}-C_{H2O,\infty}\right)$$
where C_H2O,s is the vapor concentration at the droplet surface and C_H2O,∞ is the vapor concentration in the surrounding air. The droplet size distribution followed the Rosin-Rammler model:
$$Y_d=e^{-\left(d/\bar{d}\right)^{n^*}}$$
The coupling between phases was achieved through momentum, energy, and mass source terms. The momentum source was
$$\vec{F}=\sum\left[\frac{18\mu C_DRe}{\rho_pd_p^224}\left(\vec{u}_p-\vec{u}\right)+\vec{F}_{other}\right]\Delta t$$
The heat source was
$$Q=\frac{\dot{m}_{p,o}}{m_{p,o}}\left[\left(m_{p,in}-m_{p,out}\right)\left(-H_{lat_{ref}}\right)-m_{p,out}\int_{T_{ref}}^{T_{p,out}}C_{p,p}dT+m_{p,in}\int_{T_{ref}}^{T_{p,in}}C_{p,p}dT\right]$$
and the mass source was
$$M=\frac{\Delta m_p}{m_{p,o}}\dot{m}_{p,o}$$
I used these equations to simulate the coupled heat and mass transfer in the electric vehicle battery pack. The model accounted for convective cooling, droplet evaporation, vapor diffusion, and latent heat absorption.
Mesh Independence and Experimental Validation
I divided the computational domain into structured hexahedral cells because the geometry contained regular channels and rectangular cells. To ensure that the results were independent of mesh size, I tested six meshes ranging from about 0.89 million to 5.74 million cells. The maximum temperature changed by only 0.15 K when the mesh count increased from 2.815 million to 3.455 million. Further refinement produced only marginal changes. I therefore selected the 2.815 million-cell mesh for all subsequent simulations. This mesh provided a good balance between accuracy and computational cost.
| Mesh cell count | Maximum temperature |
|---|---|
| 893,896 | 315.51 K |
| 1,433,159 | 314.95 K |
| 2,815,236 | 314.61 K |
| 3,455,368 | 314.46 K |
| 4,560,371 | 314.39 K |
| 5,744,296 | 314.31 K |
I validated the numerical model with an experimental platform. The platform consisted of a heated aluminum block, a cartridge heater, a nebulizer, a variable-speed fan, an anemometer, T-type thermocouples, and a real-time data acquisition unit. I used the aluminum block instead of a real cell because it was safer and easier to control, while still reproducing the same heat generation rate. I controlled the heater power with a voltage regulator and measured the block temperature at multiple points. I also measured the water consumption before and after each test to determine the spray mass flow rate. The uncertainty of the temperature measurement was estimated to be about 0.16 K.
I compared measured and simulated temperatures for six validation cases: three air velocities without spray and three air velocities with a spray rate of 0.15 g/s. The air velocities were 1 m/s, 1.5 m/s, and 2 m/s. The root-mean-square error was calculated as
$$RMSE=\sqrt{\frac{1}{n}\sum_{i=1}^{n}\left(y_i-\hat{y}_i\right)^2}$$
where y_i is the measured value, \hat{y}_i is the simulated value, and n is the number of data points. I obtained an RMSE of about 1.566 K, with a maximum deviation of about 1.9 K. The trends of the measured and simulated temperatures were consistent. This confirmed that my model was reliable for analyzing the electric vehicle battery pack cooling process.
| Validation condition | Value |
|---|---|
| Air velocities without spray | 1.0, 1.5, 2.0 m/s |
| Air velocities with spray | 1.0, 1.5, 2.0 m/s |
| Spray rate in spray cases | 0.15 g/s |
| Ambient temperature | 298.4 K |
| RMSE | 1.566 K |
| Maximum deviation | 1.9 K |
Ventilation-Only Cooling Behavior
I first studied the ventilation-only cooling of the electric vehicle battery pack to establish a baseline. I varied the inlet air velocity from 0.5 m/s to 3 m/s while maintaining a 3C discharge rate and an inlet air temperature of 300 K. As the inlet velocity increased, the maximum temperature decreased. At 0.5 m/s, the average surface temperature reached about 321.7 K. At 3 m/s, it decreased to about 307.3 K. Higher velocities increased turbulence, enhanced mixing, and supplied more cool air to the channel surfaces. However, the pressure drop increased sharply. It rose from about 2.4 Pa at 0.5 m/s to about 61.9 Pa at 3 m/s. This means that the fan must consume more power to maintain the flow. For practical design of an electric vehicle battery pack, I concluded that an inlet velocity between 1 m/s and 2 m/s provides a reasonable balance between cooling performance and pressure loss.
| Inlet velocity | Maximum temperature | Pressure drop |
|---|---|---|
| 0.5 m/s | 321.7 K | 2.4 Pa |
| 1.0 m/s | 318.0 K | 6.9 Pa |
| 1.5 m/s | 314.4 K | 15.5 Pa |
| 2.0 m/s | 311.7 K | 27.5 Pa |
| 2.5 m/s | 309.3 K | 43.0 Pa |
| 3.0 m/s | 307.3 K | 61.9 Pa |
I also studied the effect of inlet air temperature. I fixed the inlet velocity at 1 m/s and the discharge rate at 3C. The inlet temperature was set to 290 K, 295 K, 300 K, 305 K, and 310 K. The maximum temperature increased by about 20.2 K when the inlet temperature increased from 290 K to 310 K. A lower inlet temperature increases the temperature difference between the air and the cell surface, which improves convective heat transfer. The temperature difference inside the electric vehicle battery pack remained relatively stable, changing only from about 6.3 K to 6.6 K over this range. This indicates that inlet temperature strongly affects the absolute temperature level but has a weaker effect on temperature uniformity.
| Inlet temperature | Maximum temperature | Maximum temperature difference |
|---|---|---|
| 290 K | 306.0 K | 6.3 K |
| 295 K | 310.8 K | 6.4 K |
| 300 K | 315.6 K | 6.5 K |
| 305 K | 320.4 K | 6.5 K |
| 310 K | 326.2 K | 6.6 K |
Spray Cooling Parameters
I introduced spray into the ventilation channel to enhance cooling of the electric vehicle battery pack. I compared three nozzle types: a surface-type nozzle, a cone-back nozzle, and a cone nozzle. The surface-type nozzle injected droplets along the airflow direction over a broad area. The cone-back nozzle injected droplets toward the incoming air. The cone nozzle injected droplets primarily into the gaps between cells. I found that all spray cases reduced the maximum temperature compared with ventilation alone, but the temperature uniformity differed. The surface-type and cone-back nozzles produced uneven droplet coverage. Droplets deposited on the front cells and reduced the liquid available for the rear cells, which increased the temperature difference. The cone nozzle provided better uniformity because it directed droplets into the cell gaps. I then replaced each single cone nozzle with two vertically symmetric nozzles. This configuration further improved uniformity. I varied the vertical spacing between the two nozzles and found that 110 mm gave the best result. A smaller spacing produced droplet accumulation similar to a single nozzle, while a larger spacing caused droplets to impinge on the top and bottom walls and leave the middle region under-cooled.
| Nozzle configuration | Maximum temperature | Maximum temperature difference |
|---|---|---|
| Ventilation only | 318.0 K | 6.6 K |
| Surface nozzle | 315.8 K | 5.9 K |
| Cone-back nozzle | 315.6 K | 5.7 K |
| Single cone nozzle | 315.3 K | 5.2 K |
| Double cone nozzle, 110 mm spacing | 314.6 K | 4.7 K |
I then analyzed the effect of spray mass flow rate. I tested 0.1 g/s, 0.2 g/s, 0.3 g/s, 0.4 g/s, and 0.5 g/s under a fixed inlet velocity of 1 m/s and a nozzle spacing of 110 mm. The maximum temperature decreased from about 315.1 K at 0.1 g/s to about 314.1 K at 0.5 g/s. The improvement became small after 0.2 g/s. At 0.2 g/s, the maximum temperature was about 314.6 K, only 0.5 K lower than at 0.1 g/s. Between 0.3 g/s and 0.5 g/s, the change was only about 0.15 K. This indicates that the air becomes close to saturated with vapor, and additional water cannot evaporate effectively. The temperature difference remained between about 4.6 K and 4.8 K. Therefore, increasing the spray rate beyond a certain point mainly increases water consumption and pressure loss without proportional cooling benefit.
| Spray rate | Maximum temperature | Maximum temperature difference |
|---|---|---|
| 0.1 g/s | 315.1 K | 4.8 K |
| 0.2 g/s | 314.6 K | 4.7 K |
| 0.3 g/s | 314.4 K | 4.7 K |
| 0.4 g/s | 314.2 K | 4.6 K |
| 0.5 g/s | 314.1 K | 4.6 K |
I also studied droplet diameter. I considered 20 μm, 40 μm, 60 μm, 80 μm, and 100 μm droplets while keeping the spray rate at 0.2 g/s. Smaller droplets produced lower maximum temperatures and smaller temperature differences. The reason is that smaller droplets have a larger specific surface area, which increases heat and mass transfer. They also follow the airflow more closely and distribute more uniformly inside the electric vehicle battery pack. Among the cases I tested, the 20 μm droplets gave the best cooling performance.
| Droplet diameter | Maximum temperature | Maximum temperature difference |
|---|---|---|
| 20 μm | 314.3 K | 4.5 K |
| 40 μm | 314.7 K | 4.7 K |
| 60 μm | 315.0 K | 4.9 K |
| 80 μm | 315.3 K | 5.1 K |
| 100 μm | 315.6 K | 5.3 K |
Coupled Effects of Airflow and Spray in the Electric Vehicle Battery Pack
The interaction between airflow and spray is critical in an electric vehicle battery pack. I studied spray rates from 0.1 g/s to 0.5 g/s and inlet velocities from 0.5 m/s to 3 m/s. The maximum temperature decreased with increasing velocity. In the ventilation-only case, increasing the velocity from 0.5 m/s to 3 m/s reduced the maximum temperature by about 19 K. When spray was added, the benefit depended strongly on velocity. At 0.5 m/s, increasing the spray rate from 0.1 g/s to 0.2 g/s reduced the maximum temperature by about 1 K. At 1 m/s, the same increase reduced the temperature by about 0.5 K. At 2 m/s, it reduced the temperature by only about 0.17 K. This means that high airflow velocity shortens the droplet residence time and reduces the time available for evaporation. The droplets are carried away before they can absorb a large amount of heat. Therefore, the spray cooling effect has a diminishing return at high velocity. In practical control of an electric vehicle battery pack, the spray rate should be matched to the airflow rate. At low airflow, a moderate spray rate can provide useful cooling. At high airflow, increasing the spray rate may not be efficient.
| Inlet velocity | Spray rate | Maximum temperature | Maximum temperature difference |
|---|---|---|---|
| 0.5 m/s | 0.0 g/s | 327.3 K | 8.5 K |
| 0.5 m/s | 0.1 g/s | 320.1 K | 8.2 K |
| 0.5 m/s | 0.2 g/s | 319.1 K | 7.9 K |
| 1.0 m/s | 0.0 g/s | 318.0 K | 6.6 K |
| 1.0 m/s | 0.1 g/s | 315.1 K | 4.8 K |
| 1.0 m/s | 0.2 g/s | 314.6 K | 4.7 K |
| 2.0 m/s | 0.0 g/s | 311.7 K | 4.4 K |
| 2.0 m/s | 0.1 g/s | 309.8 K | 3.6 K |
| 2.0 m/s | 0.2 g/s | 309.6 K | 3.5 K |
| 3.0 m/s | 0.0 g/s | 307.3 K | 3.3 K |
| 3.0 m/s | 0.1 g/s | 306.9 K | 2.8 K |
| 3.0 m/s | 0.2 g/s | 306.8 K | 2.7 K |
Effect of Cell Spacing in the Electric Vehicle Battery Pack
I studied cell spacing because it directly affects airflow velocity, pressure drop, droplet distribution, and temperature uniformity in an electric vehicle battery pack. I considered spacings of 5 mm, 10 mm, 15 mm, and 20 mm. When the spacing decreased, the channel cross-sectional area decreased, and the airflow velocity increased. This enhanced convective heat transfer and reduced the maximum temperature. However, the pressure drop increased greatly. At 0.5 m/s, the pressure drop was about 16.0 Pa at 5 mm spacing and only about 0.76 Pa at 20 mm spacing, a difference of about 20 times. At 3 m/s, the pressure drop reached about 258.1 Pa at 5 mm spacing and about 17.3 Pa at 20 mm spacing. Spray behavior also changed with spacing. In narrow channels, the airflow velocity was high, so spray droplets were carried away quickly. Increasing the spray rate in a 5 mm channel reduced both the maximum temperature and the temperature difference because additional droplets spread the cooling effect more evenly. In wider channels, such as 15 mm and 20 mm, the airflow velocity was lower, and droplets could accumulate. Under low airflow, increasing the spray rate in these wider channels sometimes increased the temperature difference. At higher airflow, this accumulation effect disappeared. After considering cooling performance, pressure loss, and volume, I found that a spacing of 10 mm was a good compromise for the electric vehicle battery pack.
| Cell spacing | Inlet velocity | Pressure drop | Maximum temperature | Maximum temperature difference |
|---|---|---|---|---|
| 5 mm | 0.5 m/s | 16.0 Pa | 315.2 K | 10.9 K |
| 10 mm | 0.5 m/s | 5.8 Pa | 318.4 K | 8.0 K |
| 15 mm | 0.5 m/s | 1.9 Pa | 321.0 K | 6.8 K |
| 20 mm | 0.5 m/s | 0.76 Pa | 323.6 K | 6.4 K |
| 5 mm | 3.0 m/s | 258.1 Pa | 305.0 K | 4.7 K |
| 10 mm | 3.0 m/s | 86.4 Pa | 307.1 K | 3.8 K |
| 15 mm | 3.0 m/s | 31.0 Pa | 308.7 K | 3.0 K |
| 20 mm | 3.0 m/s | 17.3 Pa | 310.0 K | 2.6 K |
Discharge Rate and Environmental Effects
I also evaluated the cooling performance of the electric vehicle battery pack under different discharge rates. I used a spray rate of 0.2 g/s and discharge rates of 1C, 2C, and 3C. At 0.5 m/s, the maximum temperature was about 320.1 K at 3C, 310.5 K at 2C, and 303.1 K at 1C. The corresponding temperature differences were about 8.2 K, 5.0 K, and 2.4 K. At 3 m/s, the maximum temperature was about 307.4 K at 3C, 303.9 K at 2C, and 301.4 K at 1C, while the temperature differences were about 3.7 K, 1.3 K, and 1.2 K. These results show that lower discharge rates require much less cooling. If the electric vehicle battery pack operates mainly at low rates such as 1C, I can reduce the spray rate or even use ventilation alone to save energy and water.
| Discharge rate | Inlet velocity | Maximum temperature | Maximum temperature difference |
|---|---|---|---|
| 1C | 0.5 m/s | 303.1 K | 2.4 K |
| 2C | 0.5 m/s | 310.5 K | 5.0 K |
| 3C | 0.5 m/s | 320.1 K | 8.2 K |
| 1C | 3.0 m/s | 301.4 K | 1.2 K |
| 2C | 3.0 m/s | 303.9 K | 1.3 K |
| 3C | 3.0 m/s | 307.4 K | 3.7 K |
I examined ambient temperature and humidity because they affect both convective heat transfer and droplet evaporation. I considered ambient temperatures of 290 K, 295 K, 300 K, 305 K, and 310 K. At 1 m/s, the maximum temperature increased from about 306.0 K at 290 K to about 323.2 K at 310 K. The maximum temperature difference increased from about 5.5 K to about 6.0 K. At 3 m/s, the maximum temperature increased from about 297.8 K to about 316.7 K, and the temperature difference increased from about 1.8 K to about 3.3 K. Higher ambient temperature reduces the temperature difference between the electric vehicle battery pack and the air, which weakens convective cooling. It also causes droplets to evaporate faster, which can create local nonuniformity.
| Ambient temperature | Inlet velocity | Maximum temperature | Maximum temperature difference |
|---|---|---|---|
| 290 K | 1.0 m/s | 306.0 K | 5.5 K |
| 300 K | 1.0 m/s | 314.6 K | 5.7 K |
| 310 K | 1.0 m/s | 323.2 K | 6.0 K |
| 290 K | 3.0 m/s | 297.8 K | 1.8 K |
| 300 K | 3.0 m/s | 306.8 K | 2.7 K |
| 310 K | 3.0 m/s | 316.7 K | 3.3 K |
For relative humidity, I tested 20%, 30%, 40%, 50%, 60%, 70%, and 80% at an ambient temperature of 300 K. The outlet vapor mass fraction increased from about 0.0092 at 20% relative humidity to about 0.0202 at 80% relative humidity. At 1 m/s, the maximum temperature increased from about 313.5 K to about 315.4 K, and the maximum temperature difference increased from about 4.2 K to about 5.3 K. At 3 m/s, the maximum temperature increased from about 306.9 K to about 307.5 K, and the temperature difference increased from about 2.8 K to about 3.3 K. Higher humidity reduces the concentration difference between the droplet surface and the surrounding air. This lowers the mass transfer driving force for evaporation and weakens spray cooling. Therefore, the electric vehicle battery pack cooling system should account for ambient humidity in its control strategy.
| Relative humidity | Inlet velocity | Maximum temperature | Maximum temperature difference |
|---|---|---|---|
| 20% | 1.0 m/s | 313.5 K | 4.2 K |
| 40% | 1.0 m/s | 314.2 K | 4.7 K |
| 60% | 1.0 m/s | 314.9 K | 5.0 K |
| 80% | 1.0 m/s | 315.4 K | 5.3 K |
| 20% | 3.0 m/s | 306.9 K | 2.8 K |
| 40% | 3.0 m/s | 307.1 K | 3.0 K |
| 60% | 3.0 m/s | 307.3 K | 3.2 K |
| 80% | 3.0 m/s | 307.5 K | 3.3 K |
Design Implications for the Electric Vehicle Battery Pack
My results show that spray-assisted ventilation can significantly improve the thermal performance of an electric vehicle battery pack, but the design must be balanced. Inlet velocity is the strongest single factor for reducing maximum temperature, but it also causes the largest pressure drop. Spray cooling is most effective when the airflow is moderate and the droplets have enough residence time to evaporate. A cone nozzle placed in the cell gaps with a vertical spacing of about 110 mm gives better uniformity than a broad surface spray or a backward cone spray. A spray rate near 0.2 g/s is sufficient for the geometry I studied, and smaller droplets in the range of 20 μm to 40 μm provide better cooling than larger droplets.
Cell spacing is also a key design variable. Narrow spacing increases airflow velocity but raises pressure drop and can make droplet distribution uneven. Wider spacing reduces pressure drop but lowers airflow velocity and can allow droplets to accumulate. I found that 10 mm spacing offers a practical compromise for the electric vehicle battery pack. At low discharge rates, the cooling demand is much lower, so the spray rate can be reduced. At high ambient temperature and high humidity, the cooling performance degrades because both convective and evaporative driving forces are reduced. A control system for an electric vehicle battery pack should therefore adjust fan speed and spray rate according to discharge rate, ambient temperature, and humidity.
Conclusions
I performed a combined experimental and numerical study of spray and ventilation coupled cooling for an electric vehicle battery pack. I measured the heat generation of prismatic lithium-ion cells, developed a heat generation correlation, built a three-dimensional two-phase CFD model, and validated the model with experiments. I then analyzed the effects of ventilation and spray parameters on cooling performance and temperature uniformity. The main conclusions from my work are as follows.
First, ventilation alone can reduce the maximum temperature of the electric vehicle battery pack, but the pressure drop increases rapidly with inlet velocity. Increasing the inlet velocity from 0.5 m/s to 3 m/s reduced the maximum temperature by about 8.4 K in my 3C case, but the pressure drop increased from about 2.4 Pa to about 61.9 Pa. I therefore recommend an inlet velocity of 1 m/s to 2 m/s for a balanced design. Lower inlet temperature improves cooling, but the temperature difference remains relatively stable.
Second, spray injection substantially improves cooling. Among the nozzle types I tested, the cone nozzle directed into the cell gaps gave the best temperature uniformity. A vertically symmetric double-nozzle arrangement with 110 mm spacing further improved uniformity. The spray rate showed a saturation effect: after about 0.2 g/s, additional water provided little additional cooling. Smaller droplets, especially 20 μm droplets, improved both maximum temperature and temperature difference because of their larger specific surface area and better distribution.
Third, airflow and spray interact strongly in the electric vehicle battery pack. At low velocity, increasing the spray rate produced a clear temperature reduction. At high velocity, the benefit became much smaller because droplets were carried away before they could evaporate. This means that the fan speed and spray rate must be controlled together rather than optimized separately.
Fourth, cell spacing affects pressure drop, airflow velocity, and droplet distribution. Narrow spacing reduced maximum temperature but greatly increased pressure drop and often increased temperature difference. Wide spacing reduced pressure drop but weakened convective cooling and allowed droplet accumulation under low airflow. A spacing of about 10 mm was the best compromise in my study. Under low discharge rates, the electric vehicle battery pack can be cooled with less spray or ventilation alone. Under high ambient temperature and high humidity, both maximum temperature and temperature difference increase because convective and evaporative cooling are both weakened.
Overall, my study demonstrates that spray and ventilation coupled cooling is a promising method for thermal management of an electric vehicle battery pack. It combines the low cost and simple structure of air cooling with the high latent heat of water evaporation. The results provide guidance for nozzle selection, spray rate control, cell spacing, and operating strategy. Future work should consider dynamic driving cycles, long-term reliability, water recovery, and integration with other cooling methods for the electric vehicle battery pack.