In my research, I focus on the thermal management of an electric vehicle battery pack, because the safety, driving range, and cycle life of lithium-ion batteries are strongly affected by temperature. A lithium-ion battery performs best within a narrow temperature window, typically between 293 K and 313 K. When an electric vehicle battery pack operates at a high discharge rate, a large amount of heat is generated inside each cell. If this heat cannot be removed efficiently, the temperature of the electric vehicle battery pack rises rapidly, and the temperature difference among cells increases. Such non-uniform temperature distribution accelerates aging, reduces usable capacity, and can even trigger thermal runaway. Therefore, an efficient and compact cooling system is essential for any electric vehicle battery pack.
Among the available cooling strategies, liquid cooling has become the dominant solution because of its high heat transfer coefficient and controllable energy consumption. Air cooling is simple but limited by the low thermal conductivity and low heat capacity of air. Phase change material cooling and heat pipe cooling offer passive or semi-passive advantages, but they suffer from cost, weight, and limited heat dissipation capacity. Liquid cold plates, especially those with mini-channels or pin fins, are therefore attractive for a high-power electric vehicle battery pack. In my work, I propose a hybrid cold plate that combines straight channels with pin-fin structures. This design aims to enhance heat transfer while keeping pressure drop and pump power at acceptable levels.
I begin by developing a reliable battery heat generation model. The Bernardi model is used to describe the volumetric heat generation rate. The model is expressed as
$$q = \frac{\beta I}{V_c} \left[ (E_0 – U) – T \frac{dE_0}{dT} \right]$$
where \(q\) is the volumetric heat generation rate, \(I\) is the current, \(V_c\) is the cell volume, \(E_0\) is the open-circuit voltage, \(U\) is the terminal voltage, \(T\) is the temperature, and \(\beta\) is a correction coefficient. The term \(dE_0/dT\) represents the entropic heat coefficient. In practice, the reversible and polarization heat contributions are small compared with the ohmic heat, so I simplify the model to
$$q = \frac{I^2 R_T}{V_c}$$
where \(R_T\) is the internal resistance. The internal resistance depends on the state of charge (SOC) and temperature. I measured the internal resistance at a 1C discharge rate and fitted the data with a polynomial. The resistance changes slowly when the SOC is between 0.2 and 0.8, but it increases sharply when the SOC falls below 0.2 or rises above 0.8. Therefore, I limit the depth of discharge to 80% in my simulations, meaning the final SOC is 0.2.
The energy conservation equation for the battery cell is
$$\rho_b c_b \frac{\partial T}{\partial t} = k_{x,b} \frac{\partial^2 T}{\partial x^2} + k_{y,b} \frac{\partial^2 T}{\partial y^2} + k_{z,b} \frac{\partial^2 T}{\partial z^2} + Q$$
where \(\rho_b\), \(c_b\), and \(k_{x,b}, k_{y,b}, k_{z,b}\) are the density, specific heat capacity, and anisotropic thermal conductivities of the battery. The thermal properties of the battery are calculated using mass-weighted averages. For the density and specific heat capacity,
$$\rho_{cell} = \frac{\sum L_i \rho_i}{\sum L_i}, \quad c_{cell} = \frac{\sum (\rho_i L_i) c_i}{\sum \rho_i L_i}$$
For the thermal conductivity, the through-plane direction is treated as a series of thermal resistances, while the in-plane directions are treated as parallel resistances:
$$k_{T,x} = \frac{\sum L_i}{\sum L_i / k_{T,i}}, \quad k_{T,y} = k_{T,z} = \frac{\sum L_i k_{T,i}}{\sum L_i}$$
I used a 100 Ah prismatic lithium iron phosphate cell with dimensions 160 mm × 116 mm × 50 mm and a mass of 1.65 kg. The thermal properties are summarized in Table 1.
| Parameter | Value | Unit |
|---|---|---|
| Nominal capacity | 100 | Ah |
| Operating voltage | 2.5–3.65 | V |
| Dimensions (L × W × H) | 160 × 116 × 50 | mm |
| Thermal conductivity (x/y/z) | 22.5 / 22.5 / 1.5 | W·m⁻¹·K⁻¹ |
| Mass | 1.65 | kg |
| Density | 2050 | kg·m⁻³ |
| Specific heat capacity | 1088.745 | J·kg⁻¹·K⁻¹ |
I validated the battery heat generation model by comparing the simulated temperature rise with experimental data at 1C discharge. The maximum error at key points was less than 5%. This confirms that the model is accurate enough for conjugate heat transfer simulations of an electric vehicle battery pack.
Next, I built a three-dimensional conjugate heat transfer model for the electric vehicle battery pack and the cold plate. The battery module consists of 24 prismatic cells arranged in two rows of twelve. The cells are sandwiched between cold plates, and thermal silicone pads fill the gaps. The cold plate is 655 mm long, 134 mm wide, and 14 mm high. Each flow channel is 12 mm wide and 1.5 mm high. There are three cold plates, each with one inlet and two outlets. The coolant flows from the central inlet to the two outlets, which improves flow distribution. Inside each channel, I place pin fins in a staggered or inline arrangement to enhance heat transfer.
The governing equations for the coolant are the mass, momentum, and energy conservation equations:
$$\frac{\partial \rho_f}{\partial t} + \nabla \cdot (\rho_f \vec{v}) = 0$$
$$\frac{\partial}{\partial t}(\rho_f \vec{v}) + \nabla \cdot (\rho_f \vec{v}\vec{v}) = -\nabla P$$
$$\frac{\partial}{\partial t}(\rho_f c_f T_f) + \nabla \cdot (\rho_f c_f \vec{v} T_f) = \nabla \cdot (k_f \nabla T_f)$$
where \(\rho_f\), \(c_f\), \(k_f\), \(T_f\), and \(\vec{v}\) are the density, specific heat, thermal conductivity, temperature, and velocity vector of the coolant. The energy equations for the solid regions, including the silicone pad and the cold plate, are
$$\frac{\partial}{\partial t}(\rho_s c_s T_s) = \nabla \cdot (k_s \nabla T_s)$$
$$\frac{\partial}{\partial t}(\rho_c c_c T_c) = \nabla \cdot (k_c \nabla T_c)$$
For the straight channel, the flow remains laminar, so I use a laminar model. For the pin-fin channels, the pin fins induce turbulence, so I use the standard \(k\)-\(\epsilon\) turbulence model. The transport equations for turbulent kinetic energy \(k\) and its dissipation rate \(\epsilon\) are
$$\frac{\partial(\rho k)}{\partial t} + \frac{\partial(\rho u_j k)}{\partial x_i} = \frac{\partial}{\partial x_j}\left[ \left( \mu + \frac{\mu_t}{\sigma_k} \right) \frac{\partial k}{\partial x_j} \right] + P_k – \rho \epsilon$$
$$\frac{\partial(\rho \epsilon)}{\partial t} + \frac{\partial(\rho u_j \epsilon)}{\partial x_i} = \frac{\partial}{\partial x_j}\left[ \left( \mu + \frac{\mu_t}{\sigma_\epsilon} \right) \frac{\partial \epsilon}{\partial x_j} \right] + C_{\epsilon 1} \frac{\epsilon}{k} P_k – C_{\epsilon 2} \rho \frac{\epsilon^2}{k}$$
with the standard constants \(C_{\epsilon 1} = 1.44\), \(C_{\epsilon 2} = 1.92\), \(\sigma_k = 1.0\), and \(\sigma_\epsilon = 1.2\). The boundary conditions are as follows: each cell generates heat at a rate corresponding to a 1C discharge; the ambient and coolant inlet temperatures are 293 K; the mass flow rate ranges from 0.01 kg/s to 0.03 kg/s; the inlet is a mass-flow inlet; the outlet is a pressure outlet; and all exposed surfaces have a convective heat transfer coefficient of 10 W·m⁻²·K⁻¹. The material properties are listed in Table 2.
| Component | Density (kg·m⁻³) | Specific heat (J·kg⁻¹·K⁻¹) | Thermal conductivity (W·m⁻¹·K⁻¹) | Viscosity (Pa·s) |
|---|---|---|---|---|
| Silicone pad | 2094.96 | 2684 | 4 | – |
| Cold plate (aluminum) | 2719 | 871 | 202.4 | – |
| 50% ethylene glycol | 1065 | 3281 | 0.38 | 0.0069 × (T/273)⁻⁸·³ |
I performed a mesh independence study for both the straight channel and the circular pin-fin channel. The maximum temperature and pressure drop became stable when the mesh count reached about 12.5 million for the straight channel and 13.2 million for the circular pin-fin channel. Therefore, I used these mesh sizes for the simulations. To validate the numerical model, I built an experimental platform with a pin-fin channel cold plate. The pressure drop from the standard \(k\)-\(\epsilon\) model agreed best with the experimental data, with an average error of 7.1%. The average temperature error was within 0.8 K. These results confirm that the standard \(k\)-\(\epsilon\) model is suitable for my study.
I then compared four cold plate designs: a straight channel (SC), a square pin-fin channel (S-Pin-fin), a circular pin-fin channel (C-Pin-fin), and a triangular pin-fin channel (T-Pin-fin). The battery module maximum temperature \(T_{\max}\), temperature difference \(\Delta T\), and pressure drop \(\Delta P\) are used as performance indicators. The definitions are
$$T_{\max} = \max\{T_1, T_2, \ldots, T_n\}$$
$$\Delta T = T_{\max-mid} – T_{\min-mid}$$
$$\Delta P = P_{in} – P_{out}$$
Table 3 summarizes the performance at a mass flow rate of 0.02 kg/s and 0.03 kg/s. At 0.03 kg/s, the C-Pin-fin reduces \(T_{\max}\) by 2.9 K and \(\Delta T\) by 0.7 K compared with the straight channel. The S-Pin-fin gives the lowest temperature but causes the highest pressure drop. The T-Pin-fin performs between the S-Pin-fin and C-Pin-fin. The C-Pin-fin achieves the best balance between heat transfer and energy consumption.
| Channel type | \(T_{\max}\) at 0.02 kg/s (K) | \(\Delta T\) at 0.02 kg/s (K) | \(\Delta P\) at 0.02 kg/s (kPa) | \(T_{\max}\) at 0.03 kg/s (K) | \(\Delta T\) at 0.03 kg/s (K) | \(\Delta P\) at 0.03 kg/s (kPa) |
|---|---|---|---|---|---|---|
| SC | 309.0 | 5.9 | 1.2 | 308.1 | 5.4 | 2.4 |
| S-Pin-fin | 307.8 | 5.2 | 3.8 | 306.9 | 4.6 | 6.5 |
| C-Pin-fin | 308.1 | 5.4 | 2.9 | 307.2 | 4.7 | 4.8 |
| T-Pin-fin | 308.3 | 5.5 | 3.2 | 307.3 | 4.8 | 5.2 |
To understand the heat transfer enhancement mechanism, I analyzed the velocity field, heat transfer coefficient, vortex structures using the \(Q\)-criterion, and turbulent kinetic energy. In the straight channel, the thermal boundary layer grows along the flow direction, and the heat transfer coefficient decreases after the entrance region. In the pin-fin channels, the pins periodically disturb the flow, break the boundary layer, and generate vortices. The circular pin fins produce regular ring-like vortices that are aligned with high heat transfer regions. The square pin fins create fragmented vortices and high turbulence dissipation, which increases pressure drop. The triangular pin fins generate separation vortices at the front and reattachment at the rear, giving moderate enhancement. The circular pin fin is therefore the best choice for an electric vehicle battery pack because it enhances heat transfer without excessive pressure loss.
After identifying the circular pin-fin channel as the best configuration, I performed a multi-objective optimization. The design variables are the coolant mass flow rate \(A\), the channel height \(B\), and the pin-fin pitch \(C\). The objectives are to minimize the temperature difference \(\Delta T\) and the pressure drop \(\Delta P\). The range of each variable is given in Table 4.
| Design variable | Initial value | Minimum | Maximum |
|---|---|---|---|
| \(A\) (kg/s) | 0.02 | 0.01 | 0.03 |
| \(B\) (mm) | 1.5 | 1.5 | 4 |
| \(C\) (mm) | 15 | 4 | 30 |
I used the optimal Latin hypercube sampling method to generate 31 sample points. The numerical simulation results for these samples are given in Table 5. I then constructed a Kriging surrogate model. The coefficient of determination \(R^2\) is used to evaluate the accuracy:
$$R^2 = 1 – \frac{\sum_{i=1}^n (y_i – \hat{y}_i)^2}{\sum_{i=1}^n (y_i – \bar{y})^2}$$
The \(R^2\) values for \(\Delta P\) and \(\Delta T\) are 0.952 and 0.966, respectively, which are both above 0.9. This indicates that the surrogate model is accurate enough for optimization.
| Sample | \(A\) (kg/s) | \(B\) (mm) | \(C\) (mm) | \(T_{\max}\) (K) | \(\Delta T\) (K) | \(\Delta P\) (kPa) |
|---|---|---|---|---|---|---|
| 1 | 0.0260 | 2.16 | 23.6 | 307.2 | 4.4 | 2.2 |
| 2 | 0.0167 | 2.34 | 17.3 | 309.5 | 6.2 | 1.0 |
| 3 | 0.0127 | 1.92 | 22.2 | 311.3 | 7.6 | 0.9 |
| 4 | 0.0293 | 1.76 | 13.8 | 306.7 | 4.1 | 4.2 |
| 5 | 0.0153 | 1.50 | 15.2 | 309.8 | 6.6 | 2.1 |
| 6 | 0.0273 | 2.92 | 11.7 | 307.0 | 4.3 | 1.9 |
| 7 | 0.0140 | 2.66 | 5.4 | 310.5 | 7.0 | 0.8 |
| 8 | 0.0207 | 1.66 | 20.8 | 308.2 | 5.2 | 2.4 |
| 9 | 0.0160 | 3.16 | 14.5 | 309.8 | 6.2 | 0.7 |
| 10 | 0.0193 | 2.00 | 4.0 | 308.5 | 5.6 | 2.3 |
| 11 | 0.0187 | 2.50 | 24.3 | 308.8 | 5.6 | 1.1 |
| 12 | 0.0280 | 3.84 | 7.5 | 306.9 | 4.2 | 1.7 |
| 13 | 0.0120 | 2.84 | 21.5 | 311.8 | 7.7 | 0.5 |
| 14 | 0.0287 | 3.58 | 16.6 | 307.0 | 4.1 | 1.6 |
| 15 | 0.0147 | 4.00 | 11.0 | 310.6 | 6.8 | 0.5 |
| 16 | 0.0213 | 1.58 | 10.3 | 308.0 | 5.1 | 3.3 |
| 17 | 0.0220 | 3.66 | 12.4 | 308.1 | 5.0 | 1.1 |
| 18 | 0.0113 | 3.76 | 18.7 | 312.5 | 8.0 | 0.3 |
| 19 | 0.0173 | 3.50 | 22.9 | 309.4 | 5.9 | 0.7 |
| 20 | 0.0240 | 2.26 | 15.9 | 307.5 | 4.7 | 1.9 |
| 21 | 0.0300 | 2.76 | 19.4 | 306.7 | 4.0 | 2.1 |
| 22 | 0.0200 | 2.58 | 9.6 | 308.4 | 5.4 | 1.3 |
| 23 | 0.0227 | 3.08 | 18.0 | 307.9 | 4.9 | 1.2 |
| 24 | 0.0267 | 2.08 | 6.8 | 307.0 | 4.3 | 3.1 |
| 25 | 0.0247 | 3.00 | 4.7 | 307.4 | 4.6 | 1.8 |
| 26 | 0.0180 | 3.42 | 6.1 | 309.0 | 5.8 | 0.9 |
| 27 | 0.0107 | 2.42 | 13.1 | 312.7 | 8.4 | 0.5 |
| 28 | 0.0253 | 3.26 | 25.0 | 307.5 | 4.5 | 1.4 |
| 29 | 0.0133 | 1.84 | 8.2 | 310.8 | 7.3 | 1.3 |
| 30 | 0.0233 | 3.92 | 20.1 | 307.9 | 4.8 | 1.1 |
| 31 | 0.0100 | 3.34 | 8.9 | 313.3 | 8.7 | 0.3 |
Sensitivity analysis is used to quantify the influence of each design variable on the objectives. The sensitivity index is defined as
$$SA_i = \frac{f_{\max}(x_i) – f_{\min}(x_i)}{f_{\max}(x) – f_{\min}(x)} \times 100\%$$
The results show that the mass flow rate has the strongest effect on \(T_{\max}\) and \(\Delta T\), with sensitivity values of 0.996 and 0.988, respectively. The channel height is the dominant factor for \(\Delta P\), with a sensitivity of 0.442. The pin-fin pitch has a very small effect on \(\Delta P\), with a sensitivity of only 0.028. The correlation analysis shows that \(T_{\max}\) and \(\Delta T\) are strongly positively correlated, with a correlation coefficient of 0.99. Therefore, I removed \(T_{\max}\) from the optimization objectives and kept only \(\Delta T\) and \(\Delta P\). The correlation coefficients between \(\Delta P\) and \(\Delta T\), and between \(\Delta P\) and \(T_{\max}\), are \(-0.67\) and \(-0.70\), respectively.
I used the non-dominated sorting genetic algorithm II (NSGA-II) to perform the multi-objective optimization. The population size is 120 and the number of generations is 50, generating 6000 solutions. The Pareto front is obtained, and I selected three representative optimized designs for validation. Table 6 compares the optimized results with the baseline model. The baseline model has \(A = 0.0200\) kg/s, \(B = 1.50\) mm, and \(C = 15\) mm. The optimized model 1 focuses on energy saving, reducing \(\Delta P\) by 59.0% and \(\Delta T\) by 16.5%. The optimized model 2 provides the best compromise, reducing \(\Delta T\) by 23.1% and \(\Delta P\) by 43.2%. The optimized model 3 achieves the best temperature uniformity, reducing \(\Delta T\) by 24.6% and \(\Delta P\) by 25.8%. I selected optimized model 2 as the final design because it balances thermal performance and energy consumption. Its parameters are \(A = 0.0285\) kg/s, \(B = 3.14\) mm, and \(C = 19\) mm.
| Model | \(A\) (kg/s) | \(B\) (mm) | \(C\) (mm) | \(T_{\max}\) (K) | \(\Delta T\) (K) | \(\Delta P\) (kPa) | \(\Delta T\) improvement | \(\Delta P\) improvement |
|---|---|---|---|---|---|---|---|---|
| Baseline | 0.0200 | 1.50 | 15 | 308.3 | 5.40 | 3.10 | – | – |
| Optimized 1 | 0.0253 | 3.66 | 22 | 307.5 | 4.51 | 1.27 | 16.5% | 59.0% |
| Optimized 2 | 0.0285 | 3.14 | 19 | 306.9 | 4.15 | 1.76 | 23.1% | 43.2% |
| Optimized 3 | 0.0294 | 2.52 | 18 | 306.7 | 4.07 | 2.30 | 24.6% | 25.8% |
I further evaluated the optimized cold plate using thermal resistance analysis. The total thermal resistance \(R_{total}\) is the sum of the convective resistance \(R_{conv}\), the enthalpy change resistance \(R_{heat}\), and the conduction resistance \(R_{cond}\):
$$R_{conv} = \frac{T_{ave} – T_{liq}}{q}, \quad R_{heat} = \frac{T_{out} – T_{in}}{q}, \quad R_{cond} = \frac{L_b}{k_s A_{cont}}$$
$$R_{total} = R_{conv} + R_{heat} + R_{cond}$$
At a mass flow rate of 0.02 kg/s, the optimized circular pin-fin cold plate reduces the total thermal resistance by 14% compared with the straight channel. At 0.03 kg/s, the reduction reaches 16%. The improvement mainly comes from the convective resistance, while the enthalpy change and conduction resistances change very little.
The effective heat transfer enhancement factor \(pf\) is used to evaluate the trade-off between heat transfer enhancement and pressure drop penalty:
$$pf = \frac{Nu / Nu_0}{(\Delta P / \Delta P_0)^{1/3}}$$
where \(Nu\) is the Nusselt number, defined as
$$Nu = \frac{h_{ave} D_h}{k_{liq}}$$
At 0.02 kg/s, the optimized model increases \(Nu\) by more than 6 times and \(pf\) by 7 times compared with the straight channel. Compared with the baseline circular pin-fin model, the optimized model increases \(Nu\) by 33.9% and \(pf\) by 97.5%. At 0.03 kg/s, the optimized model increases \(Nu\) by 43.7% and \(pf\) by 87.7% relative to the baseline. These results confirm that the optimized design achieves significant heat transfer enhancement at a moderate pressure drop cost.
The cooling efficiency factor \(\eta\) is another important indicator. It is defined as
$$Q(t)_{liq} = c \cdot m \cdot \Delta T$$
$$Q(t)_{cell} = \int_{t_0}^{t_1} I^2 R(t) dt$$
$$\eta = \frac{Q(t)_{liq} / t}{\Delta P \cdot q_v}$$
At 0.02 kg/s, the optimized model reaches a maximum \(\eta\) of 15452, which is much higher than the baseline model (4875) and the straight channel (7250). This means that the optimized cold plate removes heat efficiently with very low flow energy consumption. Although \(\eta\) decreases as the flow rate increases for all configurations, the optimized model always maintains the highest value. This demonstrates the superior comprehensive thermo-hydraulic performance of the optimized circular pin-fin cold plate for an electric vehicle battery pack.
After the optimization, I further investigated the effect of channel height on the performance. The channel height directly determines the cold plate thickness, which is critical for compact integration in an electric vehicle battery pack. I fixed the other optimized parameters and varied the channel height from 1.0 mm to 4.0 mm. At a mass flow rate of 0.0285 kg/s, the maximum temperature increases slightly by 0.3 K as the channel height increases. The temperature difference is smallest at 1.5 mm. The pressure drop decreases significantly as the channel height increases. When the height increases from 1.0 mm to 1.5 mm, the pressure drop decreases by 7.7 kPa, a reduction of 59.2%. Further increases in height yield smaller reductions. Therefore, I selected 1.5 mm as the optimal channel height because it provides a good balance between thermal performance and compactness.
I manufactured the circular pin-fin cold plate with a 1.5 mm channel height using subtractive manufacturing. The cold plate dimensions are 732 mm × 191 mm × 12 mm. The inlet and outlet diameters are 6 mm. The pin-fin pitch is 19 mm and the pin-fin diameter is 3 mm. I built a closed-loop experimental platform to test the flow resistance and heat transfer performance. The working fluid is a 50% ethylene glycol aqueous solution. The main instruments include a gear pump, a turbine flow meter, PT100 temperature sensors, pressure transmitters, a DC power supply, an infrared thermal imager, a data acquisition system, and a thermostatic water bath. The accuracy and range of the instruments are listed in Table 7.
| Instrument | Accuracy / Range |
|---|---|
| Multi-channel DC power supply | 0–200 W |
| Infrared thermal imager | ±2% (–20–2000 °C) |
| Liquid turbine flow meter | ±0.5% (0–10 L/min) |
| Thermostatic water bath | ±0.75% (–20–100 °C) |
| Temperature sensor | ±1% (–100–280 °C) |
| Inlet pressure transmitter | ±0.5% (0–500 kPa) |
| Outlet pressure transmitter | ±0.5% (0–130 kPa) |
I performed an uncertainty analysis. The relative uncertainties of the measured parameters are calculated using the error propagation method. The relative uncertainty of a derived parameter \(R\) is
$$\xi_R = \frac{\delta R}{R}, \quad \delta R = \sqrt{\sum_{i=1}^n \left( \frac{\partial R}{\partial v_i} \delta v_i \right)^2}$$
The relative uncertainties are summarized in Table 8. The maximum relative uncertainty of the heat flux is 3.1%, and that of the thermal resistance is 3.6%. These values are acceptable for experimental validation.
| Parameter | Maximum relative uncertainty |
|---|---|
| Inlet pressure \(P_{in}\) | 2.4% |
| Outlet pressure \(P_{out}\) | 0.6% |
| Volume flow rate \(V\) | 2.5% |
| Mass flow rate \(m\) | 2.5% |
| Inlet/outlet water temperature \(T_{in}, T_{out}\) | 1.6% |
| Pressure drop \(\Delta P\) | 2.5% |
| Heat flux \(q\) | 3.1% |
| Thermal resistance \(R\) | 3.6% |
In the flow resistance experiment, I measured the pressure drop across the cold plate at different flow rates. The pressure drop increases continuously with the flow rate. At low flow rates, the pressure drop increases slowly because viscous forces dominate and the flow separation behind the pin fins is small. At medium flow rates, the inertial effect becomes significant, the wake region behind the pin fins expands, and flow separation appears, causing a faster increase in pressure drop. At high flow rates, the inertial effect dominates, the wake region becomes large, and vortex shedding may occur, leading to a rapid increase in pressure drop. However, the pressure drop of the circular pin-fin cold plate remains within an acceptable range for the electric vehicle battery pack.
In the heat transfer experiment, I attached four heating films to the cold plate to simulate the heat load. The total heating power was set to 200 W, 280 W, 360 W, and 440 W. The coolant inlet temperature and ambient temperature were both 293 K. I measured the surface temperature of the heating films using an infrared thermal imager. The temperature increases along the flow direction because the coolant absorbs heat and its temperature rises. The temperature is lowest near the inlet and highest near the outlet. I also observed that the heating film close to the inlet has a lower temperature than the downstream heating film, which is due to the entrance effect that enhances the local heat transfer coefficient.
When the flow rate is constant, the heating film temperature increases with increasing heat flux. When the heat flux is constant, the heating film temperature decreases as the flow rate increases. This is because a higher flow rate increases the flow velocity and turbulence intensity, which enhances the convective heat transfer coefficient and allows more heat to be removed. The thermal resistance and pump power are analyzed as follows. As the flow rate increases, the pump power increases exponentially because both the flow rate and the pressure drop increase. The thermal resistance decreases as the flow rate increases because the convective heat transfer is enhanced. As the heating power increases, the thermal resistance decreases initially and then tends to stabilize. This is because the coolant properties change with temperature, and the enhancement in heat transfer reaches a limit. In practical applications, I need to balance the pump power and thermal resistance to avoid excessive energy consumption for a small reduction in thermal resistance.
In conclusion, I have designed and optimized a pin-fin channel cold plate for an electric vehicle battery pack. The circular pin-fin configuration provides the best thermo-hydraulic performance. Through multi-objective optimization, I obtained an optimal design with a mass flow rate of 0.0285 kg/s, a channel height of 3.14 mm, and a pin-fin pitch of 19 mm. This design reduces the temperature difference by 23.1% and the pressure drop by 43.2% compared with the baseline model. The optimized cold plate also achieves a 14–16% reduction in thermal resistance, a 6-fold increase in Nusselt number, and a cooling efficiency factor of 15452 at 0.02 kg/s. Experimental results confirm the accuracy of my numerical model and show that the cold plate maintains good heat transfer performance with acceptable pressure drop. My study provides a practical and efficient cooling solution for a high-power electric vehicle battery pack.
For future work, I plan to extend the battery thermal model to include non-uniform heat generation and anisotropic properties. I also intend to test the cold plate under different ambient temperatures, including low-temperature preheating and high-temperature cooling. Furthermore, I will study the performance of the electric vehicle battery pack under various discharge rates, such as 2C and 3C, to evaluate the robustness of the design. Finally, I will build a real battery module test platform to replace the heating films and investigate the transient thermal behavior of the electric vehicle battery pack during actual charge and discharge cycles.