Performance Improvement of a Honeycomb Battery Thermal Management System Based on Fin–Casing Synergistically Enhanced Heat Transfer

Abstract

With the continuous rise in the energy density of power batteries, battery heat generation has become an increasingly severe issue. Particularly under extreme conditions combining high summer temperatures and high discharge rates, battery thermal safety is facing tremendous challenges. To address this problem, this study proposes a honeycomb liquid cooling–PCM hybrid battery thermal management system (BTMS) based on fin–casing synergistic heat transfer enhancement. We analyzed the effects of the longitudinal fins and thermal conductive casing on the thermal characteristics of the system, further investigated the influence patterns of key factors including fin number, battery spacing and contact thermal resistance on the thermal performance of the honeycomb BTMS, and clarified the action mechanisms of each structure and parameter on battery temperature rise and temperature uniformity. The results show that the fin structure enhances longitudinal heat conduction, improves liquid cooling efficiency, and effectively reduces the maximum battery temperature, while the thermal conductive casing significantly improves battery temperature uniformity. The BTMS with fin–casing synergistic heat transfer enhancement can control the maximum battery temperature and temperature difference within 60 °C and 5 °C, respectively, under extreme operating conditions.

1. Introduction

The performance of lithium-ion batteries is closely correlated with operating temperature. Typically, the operational temperature range for lithium-ion batteries spans from −20 °C to 60 °C [1], with strict control required to maintain a maximum temperature difference ($∆T_{\mathrm{m}\mathrm{a}\mathrm{x}}$) within 5 °C across individual cells in a module and on cell surfaces [2,3]. Exceeding thermal safety thresholds may lead to irreversible battery degradation, underscoring the criticality of efficient battery thermal management systems (BTMSs) for ensuring the safety and performance of electric vehicles (EVs).

Current BTMS methodologies encompass air cooling [4], liquid cooling [5], phase change material (PCM) cooling [6], and hybrid cooling systems [7,8]. Air cooling, characterized by a simple structure, low cost, and zero leakage risk, has been widely adopted in early EV models and low-power energy storage devices [9]. However, air’s low specific heat capacity and thermal conductivity render air-cooled systems insufficient for dissipating high heat flux during high-rate charge/discharge cycles, often resulting in poor thermal uniformity [10].

Liquid cooling regulates battery temperature through the convective heat transfer of liquid media, primarily implemented via cold-plate channel structures [11] or immersion designs [12,13]. The mini-channel design for liquid cooling systems can significantly increase the heat transfer specific surface area [14]. Nevertheless, liquid cooling systems rely on active pump-driven circulation, leading to complex system architectures, a delayed transient response to high thermal loads, and challenges in achieving efficient thermal uniformity within confined spaces [15].

PCM, as a passive cooling medium, suppresses temperature rise by absorbing latent heat during solid–liquid phase transitions [16]. However, pure PCM suffers from low thermal conductivity [17], limiting heat diffusion into deeper regions and reducing material utilization efficiency. Additionally, once melted, pure PCM requires auxiliary cooling mechanisms to re-solidify and restore its thermal storage capacity.

To address PCM’s low thermal conductivity, three enhancement strategies are commonly employed: incorporating high-conductivity nanosuspensions, integrating porous media composites [18,19], and embedding metal foams or fins to construct high-conductivity pathways. Swamy et al. [20] demonstrated that CuO or Al₂O₃ nanoparticles significantly improve PCM thermal conductivity, albeit at the cost of exponential viscosity increases. Vali et al. [21] successfully maintained battery module peak temperatures within safety limits during 3C discharge by introducing multi-walled carbon nanotubes and graphene nanoplatelets. Zhou et al. [22] found that 20 wt% expanded graphite (EG) forms a continuous thermal conduction framework, enhancing thermal conductivity by 418.5%. Yang et al. [23] developed a composite PCM–liquid cooling honeycomb structure with 12% EG content, reducing the maximum battery temperature by 4.39 °C compared to pure PCM-based BTMS. Zang et al. [24] achieved a thermal conductivity of 4.75 W/(m·K) using paraffin–EG and aluminum foam composites. Keyhani-Asl et al. [25] confirmed that copper foam can multiply the equivalent thermal conductivity, with over 90% porosity effectively suppressing temperature rise during phase transitions. However, nanoparticle agglomeration and sedimentation, potential corrosion risks from sulfur residues in carbon-based material synthesis, and the high manufacturing costs/weight of metal foams remain critical challenges.

Fin-integrated structures for constructing thermal conduction frameworks focus on fin geometry innovation and hybrid system topology optimization. Luo et al. [26] designed fractal snowflake fins with high perimeter–area ratios, enabling deep PCM matrix penetration and achieving temperature differences below 3 °C under 3C conditions. Shen et al. [27] and Wu et al. [28] proposed web-like and gear-shaped fins, respectively, which enhanced isotropic thermal networks and expanded solid–liquid phase interfaces to eliminate hotspots and accelerate phase-transition propagation. Esmaeili et al. [29] demonstrated the irreplaceable advantages of fin–PCM systems over air cooling in high-load pulse scenarios, combining conductive pathways with PCM thermal buffering capacity. Wang et al. [30] integrated fins with heat pipes, leveraging the latter’s ultra-high thermal conductivity to instantaneously transfer heat from battery surfaces to remote fin regions, ideal for space-constrained prismatic battery modules. These studies confirm that a fin-enhanced BTMS offers structural simplicity, cost control, and stable thermal performance across diverse battery configurations.

However, most published studies have focused primarily on ambient temperatures below 40 °C, while the National Standard GB38031-2020 [31] Safety Requirements for Traction Battery Packs and Systems of Electric Vehicles stipulates that the maximum allowable discharge temperature for cylindrical lithium-ion batteries is 60 °C. In actual driving processes, vehicles operate under working conditions such as rapid acceleration, climbing and full load, all of which demand high-rate discharge of batteries. In high-temperature regions in summer, the ground surface temperature can usually exceed 60 °C; during charging and discharging, batteries generate a large amount of heat on the one hand, and are exposed to baking by high-temperature ambient air from the ground on the other. If heat fails to be dissipated in a timely manner, the battery temperature will most likely exceed the safe temperature [32,33]. Furthermore, although the incorporation of metal fins into PCM can enhance heat transfer, most relevant studies have focused primarily on the regulation of the maximum or average battery temperature by fin structures in cylindrical BTMSs, whereas relatively little attention has been paid to their effects on novel configurations such as honeycomb-structured BTMSs and battery temperature uniformity. Honeycomb-structured BTMSs not only feature a compact layout, but also can effectively suppress the leakage of liquid PCM [34] and mitigate the hazards caused by battery thermal runaway [35], thus exhibiting prominent application advantages. In addition, a large internal temperature difference in the battery pack is prone to induce uneven current distribution and the generation of thermal stress, which in turn degrades battery performance and shortens its service life [36].

According to the background, this study addresses the extreme harsh working condition of 5C high-rate battery discharge at an ambient temperature of 50 °C and proposes a honeycomb-structured liquid cooling–PCM hybrid BTMS based on the synergistic heat transfer enhancement of longitudinal fins and thermal conductive casing. Based on the COMSOL Multiphysics 6.2 platform [37], four numerical models of a honeycomb BTMS with different combinations of longitudinal fins and a thermally conductive casing were developed. Numerical simulations were conducted to investigate the effects of the fin structures and conductive casing on the system heat dissipation performance, battery temperature uniformity, and PCM phase change behavior under ambient temperatures of 30, 40, and 50 °C and counter-flow coolant configurations. Subsequently, the single-factor analysis method was adopted to investigate the influences of battery spacing and contact thermal resistance on the thermal performance, battery temperature uniformity and module grouping efficiency of the honeycomb-structured BTMS with synergistic heat transfer enhancement by longitudinal fins and a thermal conductive casing.

2. Numerical Model

2.1. Geometric Model and Thermophysical Parameters

Figure 1 illustrates the structure and simplified model of the honeycomb BTMS. Cylindrical battery cells are arranged in a staggered configuration within the module, with each cell surrounded by six liquid-cooled tubes (inner diameter: 3 mm, wall thickness: 1 mm). Adjacent liquid-cooled tubes are connected via thermal conductive fins, and the internal cavities of the tubes are filled with PCM, forming a honeycomb unit structure, as shown in Figure 1a.

Based on the symmetry of the battery module, the module is simplified to a single-cell model, which comprises a cylindrical battery, PCM, liquid-cooled tubes, and their thermal conductive plates. The implementation of staggered counter-flow cooling fluid enhances temperature uniformity [14], as depicted in Figure 1b.

In this study, on the basis of the simplified honeycomb BTMS, longitudinal fins with a thickness of 1 mm and a thermal conductive casing are introduced to strengthen the internal heat transfer of the module, and four models are constructed, as presented in Figure 1c. The INR18650/25P cylindrical lithium-ion battery is adopted in this research, which has a rated capacity of 2500 mAh, supports high-rate discharge, and exhibits an allowable operating temperature range of −20 °C to 60 °C for discharge. High-purity paraffin RT54HC was adopted in this study, with a solid–liquid phase change temperature of 53–54 °C. The detailed physical properties of the battery, PCM and aluminum alloy are summarized in

Table 1. Thermophysical properties of the battery and key materials.

Material	Property	Value	Unit
INR18650/25P	Nominal Capacity	2500	mAh
	Nominal Voltage	3.6	V
	Battery Height	65.00 ± 0.15	mm
	Battery Diameter	18.35 ± 0.10	mm
	Battery Weight	48	g
	Density	2755.9	kg/m3
	Specific Heat Capacity	1129.95	J/(kg·K)
	Thermal Conductivity	Radial 1.6; Axial 27	W/(m·K)
RT-54HC	Melting Point	53–54	°C
	Density	800	kg/m3
	Specific Heat Capacity	2000	J/(kg·K)
	Thermal Conductivity	0.2	W/(m·K)
	Latent Heat	200	kJ/kg
Water	Density Liquid	998	kg/m3
	Dynamic Viscosity	1.01 × 10−3	kg/(m·s)
	Specific Heat Capacity	4180	J/(kg·K)
	Thermal Conductivity	0.599	W/(m·K)
Aluminum	Density	2719	kg/m3
	Specific Heat Capacity	871	J/(kg·K)
	Thermal Conductivity	238	W/(m·K)

.

2.2. Governing Equations

2.2.1. Heat Generation and Heat Transfer Model of Lithium-Ion Batteries

In this study, the battery is assumed to be a homogeneous solid with a constant specific heat capacity. The thermal conductivities in the radial, axial, and circumferential directions are anisotropic but constant, and the internal heat generation rate of the battery is assumed to be uniform.

The battery heat generation model adopted in this study is based on the heat generation rates obtained from the research group’s previous work, which were used to simulate the internal heat generation during battery discharge. Specifically, the dynamic internal resistance of the battery was measured using the hybrid pulse power characterization (HPPC) method under various discharge rates, including 5C, over a state-of-charge (SOC) range from 0.1 to 1.0. The nonlinear dependence of internal resistance on discharge rate and SOC was thus characterized. Subsequently, constant-current discharge temperature-rise experiments were conducted, during which real-time data of the current, terminal voltage, and surface temperature were recorded throughout the entire discharge process at different discharge rates.

Based on these measurements, the classical Bernardi heat generation equation [38] was employed to calculate the total heat generation, which was decomposed into two components: irreversible heat generated by ohmic and polarization effects, and reversible heat associated with the entropy change in electrochemical reactions. By integrating the resistance–voltage correlation data obtained from the HPPC tests with the experimentally measured transient data from the temperature-rise experiments, and further utilizing the conversion relationship between SOC and discharge time, the volumetric heat generation rate of the battery over the entire SOC range at different discharge rates was accurately determined. Finally, sixth-order polynomial expressions describing the variation in battery heat generation rate with discharge time under different discharge rates were obtained by curve fitting [23].

The governing equations for the heat transfer process of the battery are as follows:

$${\rho }_{\mathrm{b}}{C}_{p,\mathrm{b}}{\displaystyle \frac{\partial T_{\mathrm{b}}}{\partial t}}=\nabla ·\left(\left(\begin{array}{ccc}{\lambda }_{\mathrm{b}\mathrm{x}}& 0& 0\\ 0& {\lambda }_{\mathrm{b}\mathrm{y}}& 0\\ 0& 0& {\lambda }_{\mathrm{b}\mathrm{z}}\end{array}\right)\nabla T_{\mathrm{b}}\right)+Q_v$$

(1)

where ${\rho }_{\mathrm{b}}$ denotes the density of the battery, in units of kg/m³; $C_{p,\mathrm{b}}$ represents the specific heat capacity of the battery, in units of J/(kg·K); ${\lambda }_{\mathrm{b}\mathrm{x}}$, ${\lambda }_{\mathrm{b}\mathrm{y}}$, and ${\lambda }_{\mathrm{b}\mathrm{z}}$ are the thermal conductivities of the battery in three orthogonal directions; and $Q_v$ denotes the heat generated by the battery, in units of W/m³.

2.2.2. PCM Model

In this study, PCM is assumed to be a homogeneous medium with thermal conductivity in all directions. The thermophysical properties of the solid and liquid phases, including density, specific heat capacity, and thermal conductivity, are assumed to be identical and temperature-independent constants. The liquid phase region is simplified as an incompressible fluid, and natural convection is neglected [39].

The energy conservation equation is expressed as follows:

$${\rho }_{\mathrm{P}\mathrm{C}\mathrm{M}}{C}_{p,\mathrm{P}\mathrm{C}\mathrm{M}}{\displaystyle \frac{\partial T_{\mathrm{PCM}}}{\partial t}}=\nabla ·\left({\lambda }_{\mathrm{P}\mathrm{C}\mathrm{M}}\nabla T_{\mathrm{P}\mathrm{C}\mathrm{M}}\right) - {\rho }_{\mathrm{P}\mathrm{C}\mathrm{M}}L{\displaystyle \frac{\partial \beta }{\partial t}}$$

(2)

$$\beta =\left\{\begin{array}{cc}\hfill 0,& T\le T_{\mathrm{s}\mathrm{o}\mathrm{l}}\hfill \\ \hfill {\displaystyle \frac{T-T_{\mathrm{s}\mathrm{o}\mathrm{l}}}{T_{\mathrm{l}\mathrm{i}\mathrm{q}}-T_{\mathrm{s}\mathrm{o}\mathrm{l}}}},& T_{\mathrm{s}\mathrm{o}\mathrm{l}}<T<T_{\mathrm{l}\mathrm{i}\mathrm{q}}\hfill \\ \hfill 1,& T\ge T_{\mathrm{l}\mathrm{i}\mathrm{q}}\hfill \end{array}\right.$$

(3)

where $T_{\mathrm{P}\mathrm{C}\mathrm{M}}$ denotes the temperature of PCM, in units of K; ${\rho }_{\mathrm{P}\mathrm{C}\mathrm{M}}$ is the density of the PCM, in units of kg/m³; $C_{p,\mathrm{P}\mathrm{C}\mathrm{M}}$ represents the specific heat capacity of the PCM, in units of J/(kg·K); ${\lambda }_{\mathrm{PCM}}$ and $L$ are the thermal conductivity (W/(m·K)) and latent heat (J/kg) of the PCM, respectively. The last term on the right-hand side of Equation (2) characterizes the release or absorption of latent heat during the phase change process of the PCM. The liquid volume fraction β is defined as a function of temperature, as illustrated in Equation (3).

2.2.3. Liquid Cooling Model

In this study, a pipe-type liquid cooling structure is adopted, with water as the coolant, which is treated as an incompressible fluid. Calculations show that the Reynolds number (Re) of the fluid in all simulation cases is lower than the critical laminar–turbulent value (2300) [23]; thus, the flow in the fluid domain is solved using the laminar flow model. The fluid flow and heat transfer processes are governed by the equations of mass conservation, momentum conservation, and energy conservation, as presented below:

$${\displaystyle \frac{\partial {\rho }_{\mathrm{c}}}{\partial t}}+\nabla ·\left({\rho }_{\mathrm{c}}\overrightarrow{u}\right)=0$$

(4)

$${\rho }_{\mathrm{c}}{\displaystyle \frac{\partial \overrightarrow{u}}{\partial t}}+{\rho }_{\mathrm{c}}\left(\overrightarrow{u}·\nabla \right)\overrightarrow{u}=-\nabla p+\nabla \left[\mu \left(\nabla \overrightarrow{u}+{ \left(\nabla \overrightarrow{u}\right)}^T\right)\right]$$

(5)

$${\rho }_{\mathrm{c}}{C}_{p,\mathrm{c}}{\displaystyle \frac{\partial T}{\partial t}}+\nabla ·\left({\rho }_{\mathrm{c}}{c}_{p,\mathrm{c}}\overrightarrow{u}T\right)=\nabla ·\left({\lambda }_{\mathrm{c}}\nabla T\right)$$

(6)

where ${\rho }_{\mathrm{c}}$ is the density of the coolant (kg/m³), $\overrightarrow{u}$ is the velocity vector (m/s), and $p$ is the static pressure (Pa); $C_{p,\mathrm{c}}$ and ${\lambda }_{\mathrm{c}}$ are the specific heat capacity (J/(kg·K)) and thermal conductivity (W/(m·K)) of the coolant, respectively; $\mu$ denotes the dynamic viscosity(N·s/m²).

2.3. Initial and Boundary Conditions

The honeycomb BTMS exhibits geometric symmetry in its arrangement. Thus, symmetric boundary conditions are applied to the side surfaces of both the solid and fluid domains in the simplified model. The top and bottom surfaces of the model were treated as adiabatic boundaries, and the effect of natural convection with the ambient air was neglected. The heat exchange between the battery, PCM, fin structure, and cooling water is dominated by thermal conduction and thermal convection, with the specific expressions provided below:

$$-{\lambda }_{\mathrm{b}}{\displaystyle \frac{\partial T_{\mathrm{b}}}{\partial n_{\mathrm{b}}}}=-{\lambda }_{\mathrm{PCM}}{\displaystyle \frac{\partial T_{\mathrm{PCM}}}{\partial n_{\mathrm{PCM}}}}$$

(7)

$$-{\lambda }_{\mathrm{b}}{\displaystyle \frac{\partial T_{\mathrm{b}}}{\partial n_{\mathrm{b}}}}=-{\lambda }_{\mathrm{A}\mathrm{l}}{\displaystyle \frac{\partial T_{\mathrm{A}\mathrm{l}}}{\partial n_{\mathrm{A}\mathrm{l}}}}$$

(8)

$$-{\lambda }_{\mathrm{PCM}}{\displaystyle \frac{\partial T_{\mathrm{PCM}}}{\partial n_{\mathrm{PCM}}}}=-{\lambda }_{\mathrm{A}\mathrm{l}}{\displaystyle \frac{\partial T_{\mathrm{A}\mathrm{l}}}{\partial n_{\mathrm{A}\mathrm{l}}}}$$

(9)

$$-{\lambda }_{\mathrm{A}\mathrm{l}}{\displaystyle \frac{\partial T_{\mathrm{A}\mathrm{l}}}{\partial n_{\mathrm{A}\mathrm{l}}}}=h_c\left(T_{\mathrm{A}\mathrm{l}}-T_{\mathrm{c}}\right)$$

(10)

where ${\lambda }_{\mathrm{b}}$, ${\lambda }_{\mathrm{PCM}}$, and ${\lambda }_{\mathrm{A}\mathrm{l}}$ are the thermal conductivities of the battery, PCM, and fin structure, respectively, in W/(m·K); $h_c$ is the convective heat transfer coefficient between cooling water and the liquid-cooled tube wall, in W/(m²·K); $T_{\mathrm{b}}$, $T_{\mathrm{PCM}}$, $T_{\mathrm{A}\mathrm{l}}$, and $T_{\mathrm{c}}$ denote the temperatures of the battery, PCM, fin structure, and cooling water, respectively, in K.

In addition, the initial conditions were set as $T =T_{am}$ and $v ={ v}_0$, where $T_{am}$ denotes the initial temperature of the BTMS and $v_0$ represents the inlet velocity of the coolant. A velocity-inlet boundary condition was applied at the inlet, while a pressure-outlet boundary condition was specified at the outlet. According to relevant studies, a coolant flow velocity of 0.1 m/s can generally meet the heat dissipation requirement with low pump power consumption [27,40]. Therefore, 0.1 m/s is selected as the inlet flow velocity of the coolant.

2.4. Model Verification

The simplified baseline honeycomb BTMS model adopted in this study is derived from the pre-validated work of the research group [23]. The simulation results based on the COMSOL platform demonstrate that the proposed model can accurately reproduce the dynamic temperature-rise behavior of the battery under discharge rates ranging from 1C to 5C. In comparison with the experimental data, the maximum relative error of the simulation results is approximately 4%, and the root mean square errors (RMSEs) at different discharge rates are maintained at low levels of 0.45, 0.60, 0.75, and 0.38, respectively, indicating that the model accuracy satisfies the requirements for numerical simulation.

In addition, grid independence verification was conducted for the six-fin model with a thermal conductive casing under an ambient temperature of 50 °C, involving six groups of meshes with the number ranging from 5940 to 175,000. As shown in Figure 2, the maximum battery temperature exhibits a convergence trend of first increasing and then stabilizing with the increase in mesh number: in the sparse mesh regime (mesh number < 109,200), the calculation results are significantly affected by the mesh density. When the mesh number increases from 109,200 to 121,600, the maximum battery temperature rises by only 0.002 °C, with a relative deviation of less than 0.01%, indicating that the numerical solution satisfies the grid independence criteria. Further mesh refinement to 175,000 thereafter results in only negligible temperature variations. Considering both solution accuracy and computational cost, the meshing scheme with 109,200 mesh elements was selected for the subsequent studies in this work, and the mesh distribution is presented in Figure 3.

3. Results and Discussion

3.1. Thermal Performance Analysis of Honeycomb BTMS with Fins

Adding fins can improve the heat dissipation performance of the BTMS. In this section, the honeycomb BTMS without fins is taken as the baseline (Case 0). Case 1 and Case 2 are equipped with three and six longitudinal aluminum fins, respectively, and all fins are directly connected to the side surfaces of the battery cells and the liquid-cooled tube wall, as shown in Figure 4. The effects of fin configuration on the system heat dissipation performance, battery temperature uniformity, and PCM phase-transition process are investigated. Three ambient temperatures of 30 °C, 40 °C, and 50 °C are set for the simulations, with the battery discharged at a constant current of 5C and the coolant flow rate fixed at 0.1 m/s.

Figure 5 displays the temperature contour plots of the central cross-sections for the fin configuration models under three ambient temperatures. Case 0 exhibits concentric circular isotherms, with heat accumulating at the battery center to form a high-temperature region. Case 1 forms three low-temperature nodes at the fin connections, presenting a “triangular” contour pattern; in contrast, Case 2 displays a “petal-like” temperature field, where six low-temperature nodes are created at the interfaces between the six fins and the battery, dividing the high-temperature region into six smaller areas. Compared with Case 0, both Case 1 and Case 2 have a higher proportion of low-temperature color scales in their contour plots, indicating a reduction in battery temperature.

Figure 6 presents the temperature variation curves of the battery under different ambient temperatures. After adding fins, the maximum temperature of the battery decreases significantly, with Case 2 (six fins) achieving the largest reduction. Conversely, $∆T_{\mathrm{m}\mathrm{a}\mathrm{x}}$ of the battery increases sharply with the introduction of fins, rapidly exceeding the uniformity requirement of 5 °C. Notably, $∆T_{\mathrm{m}\mathrm{a}\mathrm{x}}$ of Case 2 is lower than that of Case 1, demonstrating that increasing the number of fins can partially mitigate the non-uniformity of battery temperature distribution, yet fails to meet the minimum uniformity requirement.

Figure 7 summarizes the maximum temperature and the maximum temperature difference in the batteries in three honeycomb BTMS configurations at the end of discharge. As shown in Figure 7a, under conventional operating conditions with ambient temperatures of 30 °C and 40 °C, all three configurations can maintain the maximum battery temperature within the safe range. In contrast, under the extreme condition of 50 °C, the maximum temperature of Case 0 increases to 64.8 °C, exceeding the discharge safety limit of 60 °C specified by GB 38031-2020.

With the introduction of fin structures, the maximum temperature of Case 1 under the 50 °C condition is reduced to 59.1 °C, corresponding to a temperature decrease of 5.7 °C, which represents an approximately 8.8% reduction. Case 2 further lowers the maximum temperature to 57.3 °C, achieving a decrease of 7.5 °C, equivalent to a reduction of about 11.6%. These results indicate that the fin structures significantly enhance the heat dissipation capability of the system, with Case 2 exhibiting the most pronounced cooling performance.

The results of the maximum temperature difference shown in Figure 7b reveal that, under the 30 °C and 40 °C conditions, the maximum temperature differences in Case 1 and Case 2 increase by approximately 7.0 °C and 4.3 °C, respectively, compared with Case 0, indicating that the fin structures adversely affect temperature uniformity. Under the 50 °C condition, although the maximum temperature difference decreases slightly, it still increases by 5.5 °C and 3.6 °C relative to Case 0, corresponding to increments of as high as 196% and 129%, respectively. Further comparison shows that both the maximum temperature and the maximum temperature difference in Case 2 are consistently lower than those of Case 1, suggesting that increasing the number of fins can enhance heat dissipation while partially alleviating temperature non-uniformity. Overall, although fin structures can effectively reduce the maximum battery temperature, they simultaneously lead to a significant deterioration in the temperature uniformity of the system.

For Case 0 without fins, the heat transfer pathway consists of the battery outer surface, PCM, honeycomb framework, liquid cooling tube wall, and coolant, forming a purely series thermal resistance network. By comparing the thermal conductivities of each component, it can be found that the PCM contributes to the dominant thermal resistance, resulting in a relatively high overall thermal resistance of the system. Consequently, the heat generated by the battery cannot be effectively transferred to the liquid cooling side in a timely manner, leading to limited heat dissipation performance. Under the extreme condition of 50 °C, this drawback becomes particularly pronounced, and the maximum temperature of Case 0 rises to 64.8 °C. On the other hand, since the battery is uniformly surrounded by homogeneous PCM in the circumferential direction, the radial thermal resistance distribution is nearly uniform, which contributes to good temperature uniformity. As a result, the maximum temperature difference in the battery does not exceed 3.4 °C under all three ambient temperature conditions.

For Case 1 and Case 2 with fins, parallel heat conduction pathways composed of fins and PCM are established between the battery surface and the honeycomb framework as well as the liquid cooling tubes. Owing to the much higher thermal conductivity of the fins compared with that of PCM, the equivalent thermal resistance of the parallel configuration is significantly reduced relative to that of pure PCM, leading to a substantial decrease in the overall thermal resistance of the BTMS. Consequently, the heat generated by the battery can be rapidly transferred through the low-resistance fin pathways, thereby effectively enhancing the heat dissipation performance. Under the 50 °C condition, the maximum temperatures of Case 1 and Case 2 decrease to 59.1 °C and 57.3 °C, respectively, corresponding to reductions of 5.7 °C and 7.5 °C compared with Case 0, which significantly improves the cooling performance under high ambient temperatures.

However, the parallel heat conduction structure also results in a non-uniform circumferential distribution of thermal resistance. Low-resistance pathways are formed in the regions connected to the fins, while the PCM regions between adjacent fins retain relatively high thermal resistance. The pronounced contrast between these regions generates large temperature gradients along the battery surface, thereby deteriorating temperature uniformity. As a result, the maximum temperature differences in Case 1 and Case 2 are consistently much higher than that of Case 0 under all operating conditions. In particular, under the 50 °C condition, they reach 8.3 °C and 6.4 °C, representing increases of 196% and 129%, respectively, relative to Case 0.

Under the ambient temperature of 50 °C, phase change occurs in the PCM of Case 0, Case 1, and Case 2, as illustrated in Figure 8. The PCM undergoes uniform phase change melting along the circumferential direction of the battery without fins. After adding fins, a large amount of heat is conducted out through the fins, and the region where the PCM is completely melted is significantly reduced. In addition, with the increase in the number of fins, the liquid fraction of PCM near the liquid-cooled tube increases, which reduces the distribution difference in the solid–liquid phases of the PCM and improves the distribution uniformity.

Figure 9 presents the time evolution curve of the average liquid fraction of the PCM: after the addition of fins, the liquid fraction of the PCM is reduced by nearly half. This means that more PCM remains in the solid state, storing more latent heat, which enhances the thermal safety redundancy of the system and provides an effective buffer against battery temperature rise under extreme operating conditions or thermal accumulation during long-term cycling. Mechanistically, the increase in the number of fins expands the contact area between the fins and the PCM, and improves the heat transfer in the PCM along the circumferential direction, resulting in a more uniform temperature distribution of the PCM. This facilitates the stable and full utilization of the latent heat absorption capacity of the PCM under subsequent thermal loads.

In summary, the addition of fins alone gives rise to a contradiction between heat dissipation performance and temperature uniformity. While fins can effectively suppress the maximum battery temperature and delay the latent heat consumption of PCM, they significantly deteriorate the temperature uniformity of the BTMS. This phenomenon is primarily attributed to the non-uniform distribution of local thermal resistance between the fin-contact region and the non-contact region. Therefore, enhancing the circumferential thermal conductivity of the battery surface could, in theory, maintain the efficient heat dissipation of fins while balancing the circumferential thermal resistance distribution. This hypothesis will be verified in the subsequent section of this study.

3.2. Thermal Performance Analysis of Fin–-Casing Honeycomb BTMS

Previous research in the preceding section indicated that adding fins causes an excessive maximum temperature difference in the honeycomb BTMS. In this section, a high-thermal-conductivity aluminum alloy casing with a thickness of 1 mm is introduced to evaluate its effect on the comprehensive heat dissipation performance of the system in the fin-integrated honeycomb BTMS. As shown in Figure 10, the thermal casing is tightly wrapped around the surface of the cylindrical battery, with both ends of the fins connected to the casing and the liquid-cooled tube, respectively. All other simulation conditions remain consistent with those in the preceding section.

As shown in the temperature cloud maps of the central cross-section in Figure 11, after adding the high-thermal-conductivity casing, the temperature distribution of the BTMS exhibits a concentric circular pattern for both the 3-fin and 6-fin configurations, with the highest temperature at the battery center and heat transferring uniformly outward. This is attributed to the fact that the aluminum alloy casing constructs a fully contacted heat conduction path on the battery surface. Heat is transferred from the battery surface to the low-thermal-resistance aluminum casing, where it is rapidly distributed evenly, and then transmitted to the fins uniformly, eliminating the thermal resistance difference on the battery surface caused by direct fin contact. Compared with Case 3, Case 4 features a smaller temperature gradient and superior heat dissipation performance.

Figure 12 presents the temperature-time curves of four honeycomb BTMS configurations, namely Cases 1, 2 from Figure 6 and Cases 3, 4 in this section. After adding the casing, the maximum battery temperature in Cases 3 and 4 further decreases, and the maximum temperature difference is strictly controlled below 5 °C, verifying that the thermal conductive casing can significantly improve the temperature uniformity of the battery. In particular, the maximum temperature difference in Case 3 is lower than that of Case 2, indicating that the introduction of the thermal conductive casing has a better effect on improving uniformity than simply increasing the number of fins.

Figure 13a shows that Case 4 achieves the best cooling effect; however, the cooling gain brought by the configuration upgrade shows a decreasing trend, and the temperature drop amplitude of the maximum battery temperature from Case 1 to Case 4 gradually weakens, which indicates that the main thermal resistance affecting the system heat dissipation has shifted from the external heat dissipation path resistance to the internal radial thermal resistance of the cell. Figure 13b further illustrates that at an ambient temperature of 50 °C, the introduction of the thermal conductive casing reduces $∆T_{\mathrm{m}\mathrm{a}\mathrm{x}}$ from 6.4 °C in Case 2 to 4.3 °C in Case 3; while the difference in $∆T_{\mathrm{m}\mathrm{a}\mathrm{x}}$ between Case 3 and Case 4 is only within 0.5 °C, demonstrating that the introduction of the casing has a significant effect on reducing the maximum temperature difference in the battery, whereas the influence of the number of fins on the battery uniformity is significantly weakened.

In Case 3 and Case 4, the internally generated heat of the battery is first transferred from the battery surface to the low-thermal-resistance aluminum alloy casing. Owing to the high thermal conductivity of aluminum alloy and its excellent temperature equalization capability, heat can rapidly diffuse along the circumferential direction, resulting in a more uniform temperature distribution on the battery surface. Consequently, the temperature difference induced by the direct contact between the fins and the battery is substantially mitigated. Under the 50 °C condition, the maximum temperature differences in Case 3 and Case 4 are only 4.3 °C and 4.1 °C, respectively, which are reduced by approximately 48% and 36% compared with 8.3 °C in Case 1 and 6.4 °C in Case 2, indicating a remarkable improvement in temperature uniformity.

Similar trends are also observed under the 30 °C and 40 °C conditions, where the maximum temperature differences in Case 3 and Case 4 consistently remain below 4.6 °C, which are significantly lower than those of the configurations with direct fin–battery contact. These results convincingly demonstrate the effectiveness of the aluminum alloy casing in reducing circumferential thermal resistance disparities and enhancing the surface temperature uniformity of the battery.

Figure 14a integrates the time-course curves of the average PCM liquid fraction of the five honeycomb BTMS configurations, including those in Figure 9 (Cases 0, 1, and 2) and Cases 3 and 4 in this section. Figure 14b summarizes the liquid fraction values at the end of discharge. The results indicate that the introduction of fins significantly suppresses the melting process of PCM. Furthermore, when the thermal conductive casing is added, the liquid fraction can be further reduced, with Case 4 achieving the minimum liquid fraction value. Figure 14c presents the cross-sectional distribution of the PCM liquid fraction at the end of battery discharge. Case 4 exhibits a highly uniform distribution in the circumferential direction, with no local regions of high liquid fraction concentration. This result visually indicates that the design of Case 4, with the thermal conductive casing providing uniform circumferential heating and six fins efficiently dissipating heat, maximizes the retention of the solid-state latent heat storage capacity of PCM.

As shown in Figure 15, when the fin thickness increases from 1 mm to 5 mm, the variation in the maximum battery temperature under both Case 3 and Case 4 conditions does not exceed 0.3 °C, and the overall temperature remains nearly unchanged, indicating that fin thickness has a negligible influence on the maximum battery temperature. Meanwhile, with increasing fin thickness, the variation in the maximum temperature difference under both operating conditions remains below 0.25 °C and exhibits no evident decreasing trend, suggesting that thickening the fins does not significantly improve the temperature uniformity of the system.

Based on the above results and considering the high thermal conductivity of aluminum alloy fins, which is 238 W/(m·K), it can be concluded that the heat conduction requirement can already be satisfied at relatively small fin thicknesses, and further increasing the thickness is unlikely to yield a noticeable improvement in the overall thermal performance of the BTMS. In addition, thicker fins occupy a larger volume of PCM, thereby reducing the effective latent heat storage capacity of the system, while also increasing the total mass, which is detrimental to lightweight design. Therefore, while ensuring adequate thermal performance, and considering structural mass, PCM utilization, as well as practical manufacturing constraints and mechanical strength requirements, a fin thickness of 1 mm represents a near-minimum yet reasonable design choice. Accordingly, 1 mm is selected as the optimal fin thickness in this study.

In summary, the introduction of a high-thermal-conductivity aluminum alloy casing with heat-conducting fins effectively improves the temperature uniformity of the fin-integrated honeycomb BTMS. Specifically, the maximum temperature difference in all casing-equipped configurations (Case 3 and Case 4) is controlled within 5 °C. However, restricted by the radial thermal resistance inside the battery, the inhibitory effect of increasing the number of fins on the maximum temperature exhibits a trend of diminishing marginal returns. Meanwhile, the fin–casing structure further reduces the liquid fraction of the PCM, thereby retaining more solid-phase latent heat storage capacity.

Under the working conditions of this study, Case 4 exhibits the optimal heat dissipation performance, the lowest battery temperature difference, and the smallest PCM liquid fraction. In contrast, Case 3 ensures satisfaction of temperature uniformity requirements, while the maximum battery temperature of Case 3 is slightly higher than that of Case 4. Moreover, Case 3 features a more simplified structure and achieves a favorable balance between thermal performance and manufacturing cost, making it equally worthy of consideration.

3.3. Effect of Battery Spacing on the Thermal Performance of the Fin–-Casing Honeycomb BTMS

In battery modules, the center-to-center spacing between adjacent cells is a key parameter governing the thermal management performance and space utilization efficiency of a Battery Thermal Management System. As the spacing increases, the storage capacity of PCM rises, while the heat conduction paths of the fins are elongated, leading to an increase in thermal resistance for heat conduction. In addition, the volumetric packing efficiency of the module decreases significantly. Therefore, under the operating conditions of an ambient temperature of 50 °C and a cooling fluid flow rate of 0.1 m/s, this section simulates Cases 3, 4 with cell spacings ranging from 25 mm to 32 mm to investigate the effects of cell center-to-center spacing on the thermal performance of the BTMS and volumetric packing efficiency.

Figure 16 presents the temperature contours along the fin cross-section of the two fin–casing honeycomb BTMS configurations. Increasing the cell spacing enlarges the thickness of the PCM filling layer and elongates the heat conduction paths of the fins. Both configurations maintain excellent temperature uniformity across all cell spacing conditions.

Figure 17 illustrates the influence of variations in cell spacing on the thermal performance and packing efficiency of the two honeycomb BTMS. The maximum temperatures of Cases 3 and 4 stabilize at approximately 55.8 °C and 55.5 °C, respectively. In contrast to the trend of the maximum temperature, the maximum temperature difference in the cells shows a monotonic decrease with increasing spacing. This phenomenon is attributed to two factors: first, the elongated heat conduction paths of the fins caused by larger cell spacing reduce the heat transfer efficiency of the liquid cooling loop; second, the increased thickness of the PCM elevates its thermal resistance. As the overall thermal resistance of the BTMS rises with increasing cell spacing and approaches that of the cell, the heat transfer efficiency between the battery’s interior and exterior is balanced, leading to a continuous decline in the maximum temperature difference. Meanwhile, the heat capacity of the PCM increases accordingly, enabling it to absorb more heat through latent heat storage and thus maintain the stability of the battery’s maximum temperature. This further indicates that increasing the PCM thickness strengthens its thermal buffering effect, which is conducive to improving the circumferential temperature uniformity of the cells.

As illustrated in Figure 17a,b, with increasing battery spacing, both the total volume and total mass of the honeycomb BTMS exhibit pronounced nonlinear growth trends, while the volumetric energy density correspondingly shows a nonlinear decline. Quantitative analysis indicates that when the battery spacing increases from 25 mm to 32 mm, the total volume of the thermal management system rises from 35.2 cm³ to 57.6 cm³, corresponding to a cumulative increase of 63.8%. Meanwhile, the volumetric energy density decreases cumulatively by 39%, demonstrating that enlarging the battery spacing significantly deteriorates the spatial utilization efficiency of the system.

In terms of mass, the total mass of both Case 3 and Case 4 increases with increasing battery spacing, with cumulative increments of 27.3% and 29.3%, respectively. Moreover, under identical battery spacing conditions, the total mass of Case 4 is consistently slightly higher than that of Case 3.

However, from the perspective of battery temperature uniformity, the performance enhancement achieved by increasing battery spacing is relatively limited. As the battery spacing increases, the maximum temperature differences in Case 3 and Case 4 decrease by only approximately 11%. This reduction is not only substantially lower than the growth rates of system volume and mass, but also smaller than the decline in volumetric energy density. Further analysis reveals that when the battery spacing exceeds 27 mm, each additional increase of 1 mm leads to an improvement in the maximum temperature difference of less than 0.1 °C, whereas the system volume and mass continue to increase rapidly in a nonlinear manner.

Considering multiple factors, including system volume, mass, volumetric energy density, and battery temperature uniformity, both Case 3 and Case 4 satisfy the temperature uniformity requirements under all investigated operating conditions. Therefore, from the perspectives of overall performance and engineering practicality, a battery spacing of 25–27 mm is recommended as the optimal design range for the honeycomb BTMS.

This section demonstrates that for the fin–casing honeycomb BTMS (Cases 3 and 4), increasing the battery spacing can further improve the temperature uniformity inside the module, while simultaneously reducing the module packing efficiency. Considering the sensible heat and thermal buffering benefits of the (PCM as well as the requirements of volumetric energy density, this study recommends 25–27 mm as the optimal design range. Within this range, the volume increment of the system can be controlled within 20%, effectively avoiding volume redundancy and thereby maximizing the volumetric energy density of the system.

3.4. Effect of Contact Thermal Resistance on the Thermal Performance of the Fin–-Casing Honeycomb BTMS

Considering the inevitable microvoids at solid–solid contact interfaces in practical engineering applications, this section selects the contact thermal resistance R_c as the variable and sets five different values to investigate the effect of contact thermal resistance on the thermal performance of the fin–casing honeycomb BTMS. The cell spacing is set to 27 mm, and the cooling fluid flow velocity is set to 0.1 m/s for the simulation.

The contact thermal resistance values selected in this section cover a range from 10⁻³ to 10⁻⁶ m²·K/W, which can comprehensively represent practical operating conditions under different assembly tolerances and interfacial treatment methods. For a typical assembly gap of 50 μm, the corresponding contact thermal resistance is on the order of 10⁻³ m²·K/W when air is present at the interface, whereas it decreases to approximately 10⁻⁵ m²·K/W when thermal grease is used as the interfacial material [41]. Furthermore, when a graphite-based composite thermal interface material is applied, the contact thermal resistance can be further reduced to the order of 10⁻⁶ m²·K/W [42].

Figure 18 presents the temperature field cross-sections of the BTMS under four orders of magnitude. When the contact thermal resistance 10⁻³ K·m²/W, severe heat accumulation occurs inside the cell, forming an obvious temperature stratification between the interior and exterior of the cell, which seriously impairs the heat dissipation performance of the BTMS. As the contact thermal resistance decreases from 10⁻³ to 10⁻⁶ K·m²/W, the temperature distribution of the two honeycomb BTMS configurations becomes increasingly uniform with the reduction in the magnitude.

Figure 19a,b illustrate the cell temperatures at different contact thermal resistances R_c at the end of discharge. When the contact thermal resistance R_c is at the level of 10⁻⁴ K·m²/W, both the maximum cell temperature and the maximum temperature difference remain stable, and the BTMS maintains high-efficiency heat dissipation. In contrast, when the contact thermal resistance exceeds the level of 10⁻⁴ K·m²/W, the maximum cell temperature rises sharply, while the maximum temperature difference continues to decrease. Combined with the results in Figure 18, it can be concluded that when R_c is at the level of 10⁻⁴ K·m²/W or lower, the heat inside the cell is rapidly transferred to the liquid-cooled tube. However, when the contact thermal resistance increases to 10⁻³ K·m²/W, a large amount of heat accumulates inside the cell and cannot be transferred across the contact interface, which reduces the temperature gradient inside the cell and leads to an increase in the maximum temperature.

In summary, the magnitude of contact thermal resistance has a significant influence on the maximum battery temperature and the maximum temperature difference. Excessively high contact thermal resistance causes a sharp increase in the maximum battery temperature; meanwhile, it exhibits a temperature equalization effect, which leads to a reduction in the maximum temperature difference. The findings indicate that when the contact thermal resistance is at the order of 10⁻⁴ K·m²/W or lower, its adverse impact on the heat dissipation performance is negligible.

4. Conclusions

In this study, numerical simulations were conducted on the fin–casing honeycomb BTMS under conditions of 5C high-rate discharge and an ambient temperature of 50 °C. By comparing the thermal performance of five configurations, two optimized cases that meet the heat dissipation requirements were selected. Furthermore, the influence patterns of battery spacing and contact thermal resistance on the system’s heat transfer characteristics were analyzed. The main conclusions are as follows:

(1)

The addition of longitudinal fins can significantly suppress the temperature rise in the battery and delay the phase change process of PCM, with a more pronounced effect achieved as the number of fins increases. However, the installation of fins leads to an increase in the internal temperature difference in the battery due to the uneven distribution of circumferential thermal resistance.

(2)

The introduction of a battery thermal conductive casing effectively resolves the issue of excessive internal temperature difference in the battery caused by the fin structure, further enhances the heat dissipation capability of the system, and delays the phase change process of PCM.

(3)

The battery spacing affects the temperature uniformity and grouping efficiency of battery modules. Increasing spacing raises the latent heat of PCM and improves the temperature uniformity of the battery but has a negligible effect on the maximum temperature and leads to a decline in the grouping efficiency of the BTMS. Considered comprehensively, a battery spacing of 25–27 mm is optimal.

(4)

The contact thermal resistance between the battery and the thermal conductive casing is a key parameter influencing the heat dissipation efficiency and temperature uniformity of the battery. Excessively high contact thermal resistance exacerbates heat accumulation inside the battery and raises the maximum temperature. When the contact thermal resistance is controlled below 10⁻⁴ K·m²/W, its adverse impact on the system heat dissipation can be neglected.

The numerical simulations conducted in this study aim to investigate the effects of fin structures and a thermally conductive casing on the thermal performance of the honeycomb BTMS, thereby providing optimized design guidelines for subsequent experimental investigations. In the present work, a simple longitudinal fin configuration is adopted. In future studies, different fin geometries and advanced optimization strategies will be further explored. In addition, the natural convection of liquid-phase PCM will be taken into consideration, and a quantitative evaluation of the liquid-cooling pump power and BTMS cost will be conducted.