With the continuous advancement of the “dual carbon” objective, the development and application of renewable energy sources such as wind and solar energy have achieved remarkable results. However, the intermittent and unstable nature of wind power and photovoltaic power generation often gives rise to the phenomena of “abandoned wind” and “abandoned light”, which substantially affects the efficiency of electric power utilization. Under these circumstances, energy storage technology has become the critical solution to the integration challenge of renewable energy. Among various energy storage technologies, lithium-ion batteries are extensively used in energy storage systems owing to their high energy density, long cycle life, and mature manufacturing process. Nevertheless, during actual operation, lithium-ion batteries generate a significant amount of heat. If this heat is not removed in a timely manner, the increasing temperature of the cells will cause adverse effects on their electrochemical performance, service lifetime, and safety. In severe cases, thermal runaway may be triggered, which can lead to fire or explosion accidents. Therefore, an effective battery thermal management system (BTMS) is essential for energy storage battery packs to ensure the cells operate within the optimum temperature range. Among the existing cooling strategies, the air-cooling system is widely adopted in battery thermal management because of its simple structure, low cost, convenient maintenance, and high reliability. However, the air-cooling system still has some deficiencies in terms of cooling efficiency and temperature uniformity under high-rate charge-discharge conditions and complex operating scenarios. Hence, further research is necessary to optimize the air-cooling system for energy storage battery packs, aiming at improving the system efficiency and operational safety, thereby enhancing the performance and extending the cycle life of the batteries.
In this dissertation, I first established a detailed numerical model of an energy storage battery pack, including cells, positive and negative electrodes, and busbars, based on the actual electrical connection configuration of the energy storage battery packs. A novel MSMD-NTGK thermoelectric coupling battery model was employed to simulate the discharge process of a single prismatic lithium iron phosphate (LiFePO₄) battery under different discharge rates. The simulation results were compared with the experimental data reported in the literature to verify the reliability of the model. On this basis, I explored the effects of different airflow organizations and battery pack arrangement modes on the cooling effectiveness of the air-cooling system and the energy density of the battery pack. Specifically, three battery arrangement types were analyzed: linear arrangement, staggered arrangement, and aligned arrangement. Through a series of multi-condition numerical simulations, the study not only verified the fundamental theoretical rules in the field of battery thermal management, but also yielded a series of innovative findings with engineering guidance value. For instance, the discharge rate was found to be positively correlated with the heat generation rate; in the range of 0.5C to 3C, every 1C increase in the discharge rate raised the maximum cell temperature by 3–5 ℃. In terms of airflow organization optimization, the adoption of a short-flow-channel and multi-channel design reduced the maximum temperature of the battery pack by 5.7%. An inlet air velocity of 5 m/s was found to balance the cooling effectiveness and energy consumption. The optimization of the number of outlets indicated that the configuration with four outlets reduced the pressure drop by 90.6% while keeping the temperature within the acceptable range. The investigation of the battery arrangement showed that the optimal staggered distance and the optimal battery spacing were 10 mm and 5 mm, respectively, for the staggered arrangement, which lowered the maximum temperature difference by 4.5%. In the aligned arrangement, the installation of a deflector plate increased the amount of cold air flowing into the battery gaps, raised the internal air velocity, and further improved the cooling effect. When the deflector height was 25 mm, the maximum temperature difference was reduced by 5.6% compared with the case without the deflector. Comparing the air-cooling systems under the three battery arrangements, I observed that the aligned arrangement delivered the best cooling performance, but the energy density of the battery pack was relatively lower. Finally, to further improve the cooling performance of the battery pack, I established a battery pack composed of 12 batteries at a discharge rate of 3C in a linear arrangement. The battery pack incorporated both a deflector and a constant-temperature radiation plate, and a more detailed optimization study was conducted on the dimensions of the deflector. The research results revealed that the deflector size had a significant impact on the cooling effect of the battery pack. Installing a deflector with an appropriate size effectively modified the airflow distribution inside the pack, enhanced the heat exchange between the air and the batteries, and optimized the temperature distribution of the battery pack. After coupling the optimized deflector with the constant-temperature radiation plate, the temperature uniformity was further improved, and the maximum temperature difference was reduced by 12.1% compared with the case without the deflector and radiation plate. At the same time, the energy consumption of the air-cooling system was also reduced, and the pressure drop between the inlet and outlet was decreased by 11.3%. The research results of this dissertation provide theoretical support and practical reference for the performance improvement and optimized design of air-cooling systems for energy storage battery packs.
1. Introduction
In the context of global energy transition, the intensive consumption of fossil fuels has caused severe greenhouse gas emissions, energy crises, and global warming. Therefore, the exploitation of renewable energy sources has become a common consensus. According to the International Energy Agency (IEA) data, global energy demand is expected to increase by 0.7% in 2024, while electricity demand is projected to grow by 4.3%. By 2024, renewable energy has already accounted for about 30% of the global electricity generation structure. Nevertheless, the characteristics of renewable energy, including volatility, randomness, intermittency, and regional dependence, are strongly affected by natural conditions. These features lead to unstable power supply and the occurrence of “abandoned wind” and “abandoned light”, which restrict the sustainable development of the industry. To maintain the balance between energy supply and demand, energy storage systems (ESS) are regarded as the most practical solution.
According to the forms of energy conversion, energy storage systems can be classified into chemical, electrochemical, electrical, mechanical, and thermal energy storage categories. Among these, electrochemical energy storage technology occupies a dominant position. In the field of electrochemical energy storage, lithium-ion battery energy storage systems demonstrate numerous advantages, such as high energy density, relatively light weight, long lifespan, and good adaptability. These characteristics make them one of the most competitive options in current energy storage infrastructure. In January 2022, the National Development and Reform Commission and the National Energy Administration jointly issued the “14th Five-Year Plan for the Development of New Energy Storage”, explicitly proposing to promote the high-quality and large-scale development of new energy storage, with emphasis on advancing relatively mature technologies like lithium-ion batteries to achieve continuous cost reduction and commercial scale application.
During the charge-discharge cycles of lithium-ion energy storage batteries, a large amount of heat is generated. If this heat cannot be dissipated rapidly, the pack temperature will continuously rise, which affects the capacity and reliability of the battery. This problem is especially prominent in large-scale energy storage battery packs, because batteries are densely arranged and numerous, resulting in large internal heat generation but limited space for heat dissipation, which easily leads to heat accumulation. High temperature accelerates the decomposition of internal materials, accelerates battery aging, and increases capacity decay. In severe cases, it may cause combustion or explosion accidents. In recent years, numerous fire accidents in energy storage power stations caused by thermal runaway of lithium batteries have been reported. For instance, in February 2025, a 20 MWh energy storage power station in Wuwei, Gansu, caught fire due to thermal runaway of LiFePO₄ batteries. In February 2025, the Moss Landing energy storage power station in the United States experienced a fire, which was the fourth time that the station had been involved in a fire incident. Also, between 2017 and 2019, more than 30 lithium-ion battery energy storage fires occurred in South Korea. These accidents reflect both the widespread application of energy storage batteries and the safety hazards of thermal runaway under certain conditions. Therefore, it is crucial to develop an efficient battery thermal management system to ensure timely heat dissipation, prevent thermal runaway, improve battery performance, extend the lifespan, and guarantee operational safety.
Lithium-ion battery thermal management is not only beneficial for improving battery performance and extending cycle life in high-power applications, but also effectively solves the thermal safety problems of the battery. Many researchers have revealed that the optimal operating temperature range of lithium-ion batteries is 25 ℃–40 ℃, and the temperature difference between cells should be maintained within 5 ℃. Hence, an effective BTMS is an indispensable component for the efficient, stable, and safe operation of battery packs. According to the cooling media, BTMS can be divided into air cooling, liquid cooling, and phase-change cooling. The air-cooling system utilizes air as the cooling medium, removing heat generated by the battery through airflow. Liquid cooling uses a liquid as the medium, transferring heat from the battery through liquid circulation. Phase-change cooling relies on the large latent heat absorbed or released during the solid-liquid phase transition of phase-change materials (PCMs) to regulate battery temperature. Each type of thermal management system has its specific merits and limitations, and their applications have improved the performance and safety of lithium-ion battery packs to varying degrees.
1.1 Air-Cooling Thermal Management System
The air-cooling system mainly depends on air movement to take away the heat generated by the battery. Based on the driving force of airflow, it can be divided into natural convection cooling and forced convection cooling. Natural convection cooling relies on the natural movement of air, such as the convective heat exchange between air in the battery box and the battery module, to realize cooling. This method requires no additional power equipment and is energy-saving; however, its cooling efficiency is relatively low and it is significantly affected by ambient temperature. Forced convection cooling uses fans or other equipment to force air movement, increasing airflow velocity and enhancing the cooling effect. Forced convection cooling can control the battery temperature more effectively and is suitable for scenarios with high cooling demand. Because of its small volume, simple structure, high compactness, light weight, high design flexibility, low cost, little maintenance, and high reliability, the air-cooling system has been widely applied in energy storage battery thermal management.
Extensive optimization studies have been carried out on air-cooling systems by researchers all over the world. Shi et al. established a battery module model to predict battery temperature variations with module design and operating conditions as well as system power consumption. Under low discharge rates, the air-cooling system exhibited lower power consumption compared with liquid-cooling systems. Wang et al. optimized the position and number of the inlet and outlet, and found that reducing the height of the inlet and increasing the number of inlets made the temperature distribution more uniform. Cai et al. studied the effects of the arrangement of battery packs and battery spacing on the cooling effect of air-cooled systems, reporting that the best cooling effect was obtained with an aligned arrangement and 32 mm spacing in the Y direction and 4 mm spacing in the X direction. Wang et al. investigated the cooling effect of battery modules under different cell arrangements and different installation positions of fans, and found the top fan position gave the best cooling effect. Du et al. installed guide plates in the battery compartment to optimize the cooling system, observing that reasonable installation of guide plates made the air and battery modules exchange heat more fully, thereby improving the cooling effect. Chen et al. studied the influence of the positions of inlets and outlets, indicating that the optimized “Z-type” battery thermal management system possessed superior cooling performance. Zhang et al. discovered that the heat generated in the discharge process was far greater than that generated in the charge process, that top-positioned inlets gave the best cooling effect, and that an optimal battery spacing existed for maximizing the cooling effect. Fan et al. found that reducing the spacing between batteries reduced the maximum temperature, but deteriorated temperature uniformity to some extent. Yang et al. theoretically investigated a control strategy with numerical simulation to balance the temperature distribution across the module, demonstrating that adding controllable valves to a typical “Z-type” model alleviated thermal imbalance. Severino et al. proposed a general framework for optimal BTMS design using a multi-objective particle swarm optimization algorithm. Sun et al. introduced a “Z-type” airflow channel with tapered inlet and outlet, showing that the tapered duct effectively alleviated airflow fluctuation, reduced temperature variation, and lowered the total pressure drop.
1.2 Liquid-Cooling Thermal Management System
Liquid cooling takes advantage of the high heat transfer coefficient of liquids to remove heat generated during battery operation, ensuring that the battery operating temperature is maintained within a certain range and guaranteeing temperature uniformity. However, the liquid-cooling system requires additional pumps, valves, condensers and other auxiliary equipment, which increases system cost and complexity. At the same time, the complex liquid circuit increases the risk of leakage; hence, stricter sealing requirements are necessary. Liquid cooling systems are mainly divided into direct-contact and indirect-contact types. In direct-contact systems, the battery pack is directly in contact with insulating liquids, such as mineral oil. Indirect-contact systems use a cooling plate or jacket, where the coolant does not directly contact the battery cells. Chen et al. employed a roll-bond liquid cooling plate with good cooling performance and low manufacturing cost, and demonstrated that the plate controlled the cell temperature below 35 ℃ and the temperature difference within 5 ℃ at a 2C discharge rate, at the expense of a relatively small pressure drop. Zhang et al. developed a thermal management system based on sodium polyacrylate (PAAS) hydrogel, showing high energy efficiency, easy fabrication, compactness, and low cost. Panchal et al. proved that water is a good cooling medium that effectively reduces the maximum temperature and ensures temperature uniformity. Satyanarayana et al. compared the performance of a direct-contact liquid cooling system using mineral oil and therminal oil at different discharge rates, and reported that immersion cooling technology showed better performance in controlling the maximum temperature of the battery module. Tan et al. proposed a novel direct liquid cooling system based on hydrofluoroether (HFE-6120) coolant, and numerically studied key parameters with a CFD approach. Basu et al. developed a three-dimensional electrochemical-thermal coupling model to evaluate the influence of operating conditions on pack temperature, and designed a novel temperature correlation to predict the temperature of all individual cells from a given measurement. Wang et al. proposed a liquid-cooled BTMS based on serpentine microchannels, demonstrating that interleaved placement of inlet and outlet provided the best cooling performance.
1.3 Phase-Change Cooling Thermal Management System
PCM-based battery thermal management is an efficient cooling technology that can effectively control battery temperature within the optimum operating range. PCM-based systems are mainly divided into passive (single) and active (hybrid) forms. The passive system uses only PCM for thermal management, having a simple structure and low energy consumption. The active system combines PCM with other cooling methods such as air cooling or liquid cooling, forming a hybrid system that possesses better temperature control performance, stability, and safety. However, selecting PCM with appropriate phase-change temperature and thermal characteristics is crucial for effective operation. When the ambient temperature approaches or exceeds the phase-change temperature of PCM, its ability to absorb or release extra heat is hindered. Huo et al. explored a passive thermal management strategy for a battery module based on non-uniform paraffin and glass fiber, examining the influence of glass fiber on the thermal conductivity and structural strength of paraffin. Ranjbaran et al. combined PCM with an air-cooling system, indicating that a higher inlet air velocity or a longer cooling pipe length through the PCM volume led to lower battery surface temperature. Li et al. proposed a novel BTMS based on PCM coupled with fractal fins, and studied the effect of latent heat, thickness, and thermal conductivity of PCM, as well as the number, aspect ratio, material, and structural morphology of fins. Chen et al. proposed a BTMS combining PCM and heating plates, and investigated the influence of structure, phase-change material, composite phase-change material, and battery spacing. Wu et al. simulated the thermal behavior under different gravity environments and inclination angles, showing the special significance of gravitational acceleration for PCM cooling effectiveness. Zhang et al. proposed a composite phase-change material with Na₂SO₄·10H₂O as the core and expanded graphite as the thermal conductivity enhancer, which demonstrated high latent heat, superior thermal conductivity, and two-stage temperature control function. Gu et al. proposed a coupled phase-change energy storage system based on composite PCM and variable-wall liquid cooling plates, and adopted a genetic algorithm to optimize the structure of the composite PCM. Su et al. proposed a composite thermal management scheme based on a harmonica plate coupled with phase-change material for prismatic lithium-ion batteries, and found that placing inlet and outlet on the same side gave better thermal management performance, while the “J-type” heat exchanger exhibited better overall temperature uniformity and lower energy consumption. Wang et al. proposed a hybrid cooling system combining PCM and wavy microchannel cooling plates, showing lower battery temperature than using PCM or air cooling alone.
Compared with air cooling, liquid cooling and phase-change cooling still exhibit significant limitations. Liquid cooling adds complexity, cost, and potential leakage risk. Phase-change cooling is restricted by the low thermal conductivity and thermal saturation of materials. Both are inferior in terms of system weight and maintenance convenience. In the field of energy storage battery thermal management, air cooling has been widely used because of its simplicity and reliability. Nevertheless, with the development of high-energy-density energy storage systems, the heat flux of batteries has been increasing, and the thermal management is facing new challenges. It is therefore essential to develop an efficient, reliable BTMS to improve the battery performance, extend its service life, and guarantee its safety.
In most existing numerical investigations, the battery is simplified as a uniformly heated stable heat source with either a cylindrical or prismatic shape. This simplification neglects the non-uniform heat generation inside the battery, and also ignores the influence of electrodes and busbars on heat generation. Such simplifications may lead to considerable simulation errors. Therefore, in this dissertation, I established a detailed numerical model of a lithium-ion battery pack including the cell, positive and negative electrodes, and busbars. Based on the MSMD-NTGK battery model of a prismatic LiFePO₄ battery available in the literature, I optimized the airflow organization and battery arrangement inside the battery pack to improve the cooling efficiency of the air-cooling system, providing a reference for the optimized design of air-cooling systems for energy storage battery packs.
2. Heat Generation Analysis and Modeling of Lithium-Ion Batteries
2.1 Structure and Working Principle of Lithium-Ion Batteries
An energy storage battery pack is composed of multiple single cells connected in series or parallel. Each single cell includes electrodes, separators, and electrolyte. The positive electrode material is the main source of lithium ions and directly determines the energy density, safety, and cycle life of the battery. Common positive materials include LiFePO₄, LiCoO₂, LiMn₂O₄, etc. The negative electrode allows lithium ions to deintercalate/intercalate freely, realizing the charge-discharge function. Common negative materials include graphite, hard carbon, and silicon-based materials. The separator is a microporous polymer membrane, usually made of polyethylene (PE), polypropylene (PP) or their composite materials. The separator physically isolates the positive and negative electrodes to prevent short circuits while allowing lithium ions to move through its micropores between the electrodes. The electrolyte is the ion transport medium in lithium-ion batteries, generally composed of lithium hexafluorophosphate dissolved in carbonate organic solvents. The electrolyte provides a channel for lithium ion transport and participates in the electrochemical reactions, ensuring smooth intercalation/deintercalation of lithium ions between the electrodes. The structure also includes the battery shell, top cover, terminal posts, and safety vent. Among various types of lithium-ion batteries, LiFePO₄ batteries have become the preferred technology route for energy storage systems because of their excellent thermal stability (decomposition temperature above 300 ℃), long cycle life (more than 2000 cycles), high safety (intrinsically stable olivine structure), and environmental friendliness.
The charge-discharge process of a lithium-ion battery is essentially an electrochemical reaction process in which lithium ions migrate between the positive and negative electrodes while electrons transfer in the external circuit. During charging, lithium ions are extracted from the positive electrode, pass through the electrolyte and separator, and insert into the negative electrode. Simultaneously, an equal number of electrons transfer through the external circuit to the negative electrode. The discharge process is the reverse. For a LiFePO₄/graphite cell, the electrode reactions can be expressed as:
$$ \text{Positive: } LiFePO_4 \rightarrow xLi^+ + xe^- + Li_{(1-x)}FePO_4 $$
$$ \text{Negative: } Li_xC_6 \rightarrow 6C + xLi^+ + xe^- $$
2.2 Heat Generation Mechanism
Lithium-ion batteries generate heat during operation due to the internal electrochemical reactions and physical processes. The total heat generation rate, Qtotal, can be divided into five components: reaction heat Qr, polarization heat Qp, ohmic internal resistance heat Qj, electrolyte decomposition heat Qe, and SEI film decomposition heat Qs.
The reaction heat Qr is the heat generated during charge-discharge due to the internal electrochemical reaction, belonging to reversible heat. It is negative during charging and positive during discharging. It can be expressed as:
$$ Q_r = nFT \frac{\partial(E_a)}{\partial T} $$
The polarization heat Qp arises from the electrochemical reaction heat and ohmic heat during operation. It is generally simplified as the product of current and voltage:
$$ Q_p = I^2 R_p $$
The ohmic internal resistance heat Qj refers to the heat generated when current passes through a resistance. In lithium-ion batteries, the resistance mainly comes from electrode materials, electrolyte, and internal contact resistance. It can be calculated as:
$$ Q_j = I^2 R_j $$
The electrolyte decomposition heat Qe and SEI film decomposition heat Qs occur under extreme conditions such as high temperature or internal short circuit. Under normal low-rate operation, these two components are negligible. The total heat generation rate can be expressed as:
$$ Q_{total} = Q_r + Q_p + Q_j + Q_s = I_L T \left( \frac{dE_0}{dT} \right) + Q_p + Q_{ohm} + Q_s $$
From the equation above, the heat generated during charge-discharge is mainly determined by the current, open-circuit voltage, polarization internal resistance, ohmic internal resistance, and battery temperature.
2.3 Heat Transfer Characteristics
Heat generated inside the battery is simultaneously transferred to the surroundings through three fundamental modes: heat conduction, heat convection, and heat radiation. Heat conduction occurs inside the battery, where the electrolyte, electrodes, separator, and shell are all conductors. The heat conduction rate is given by Fourier’s law:
$$ q = -k_n \frac{\partial T}{\partial n} $$
Heat convection occurs on the battery surface between the surface and the fluid medium. It is described by Newton’s law of cooling:
$$ q = h(T_1 – T_2) $$
Any object with a temperature above absolute zero emits thermal radiation. The radiation heat transfer follows the Stefan–Boltzmann law:
$$ q = F\varepsilon\sigma A_1(T_1^4 – T_2^4) $$
2.4 Thermophysical Parameters of the Battery
Because the cell core is a stack of multiple layers, the thermal conductivity is anisotropic. In the X and Y directions, the material distribution is uniform, so the thermal conductivities are the same. In the Z direction, the thermal conductivity is calculated by the series thermal resistance method:
$$ k_{T,x} = k_{T,y} = \frac{\sum L_i k_{T,i}}{\sum L_i} $$
$$ k_{T,z} = \frac{\sum L_i}{\sum (L_i / k_{T,i})} $$
The average density and specific heat capacity of the battery are calculated as the weighted averages of the component materials:
$$ \rho_{batt} = \frac{\sum L_i \rho_i}{\sum L_i} $$
$$ c_{batt} = \frac{\sum (\rho_i L_i) c_i}{\sum \rho_i L_i} $$
The prismatic LiFePO₄ battery used in this study has the following characteristic parameters:
| Parameter | Value |
|---|---|
| Dimensions (mm) | 135 (L) × 180 (H) × 30 (W) |
| Mass (g) | 1400 |
| Nominal voltage (V) | 3.2 |
| Nominal capacity (Ah) | 50 |
| Discharge cut-off voltage (V) | 2.5 |
| Charge cut-off voltage (V) | 3.65 |
The thermophysical parameters of the cell components are listed below:
| Component | Specific heat capacity (J/(kg·K)) | Thermal conductivity (W/(m·K)) |
|---|---|---|
| Cell core | 2173 | 18.3 : 18.3 : 1.1 (X:Y:Z) |
| Positive electrode | 2719 | 871 |
| Negative electrode | 8978 | 381 |
| Busbar | 8978 | 381 |
2.5 MSMD-NTGK Thermoelectric Coupling Model
The battery thermal management simulation was carried out with the CFD approach, which solves the governing conservation equations of mass, momentum, and energy. The continuity equation, momentum equation, and energy equation are expressed as:
$$ \frac{\partial \rho}{\partial t} + div(\rho \vec{u}) = 0 $$
$$ \frac{\partial(\rho u)}{\partial t} + div(\rho u \vec{u}) = div(\mu \, grad(u)) – \frac{\partial p}{\partial x} + S_u $$
$$ \frac{\partial(\rho v)}{\partial t} + div(\rho v \vec{u}) = div(\mu \, grad(v)) – \frac{\partial p}{\partial y} + S_v $$
$$ \frac{\partial(\rho w)}{\partial t} + div(\rho w \vec{u}) = div(\mu \, grad(w)) – \frac{\partial p}{\partial z} + S_w $$
$$ \frac{\partial(\rho T)}{\partial t} + div(\rho \vec{u} T) = div\left( \frac{\lambda}{c_p} grad(T) \right) + S_T $$
The MSMD (Multi-Scale and Multi-Domain) model is a powerful approach for simulating the coupled thermal and electrical behavior in lithium-ion batteries. The MSMD model partitions the battery into different calculation domains at particle scale, electrode scale, and cell scale, each with its own coordinate system. The NTGK sub-model is a semi-empirical electrochemical model that relates the voltage U and conductivity Y to the depth of discharge (DOD). The governing equations for the thermal and electric fields in the cell-scale CFD domain are:
$$ \frac{\partial \rho C_p T}{\partial t} – \nabla \cdot (k \nabla T) = \sigma_+ |\nabla \varphi_+|^2 + \sigma_- |\nabla \varphi_-|^2 + q_{ECh} + q_{short} + q_{abuse} $$
$$ \nabla \cdot (\sigma_+ \nabla \varphi_+) = -(j_{ECh} – j_{short}) $$
$$ \nabla \cdot (\sigma_- \nabla \varphi_-) = j_{ECh} – j_{short} $$
where \(\sigma_+\) and \(\sigma_-\) are the effective electrical conductivities of the positive and negative electrodes, \(\varphi_+\) and \(\varphi_-\) are the phase potentials, and \(j_{ECh}\) and \(q_{ECh}\) are the volumetric current transfer rate and electrochemical reaction heat, respectively. Under normal operating conditions, the short-circuit and abuse terms are zero. The volumetric current transfer rate and heat generation rate during electrochemical reaction are calculated as:
$$ j_{ECh} = \frac{Q_{nom}}{Q_{ref} V_{ol}} Y[U – V] $$
$$ q_{ECh} = j_{ECh} \left[ U – V – T \frac{dU}{dT} \right] $$
The DOD is calculated as follows:
$$ DOD = \frac{V_{ol}}{3600 Q_{nom}} \int_0^t j \, dt $$
The parameters U and Y are represented as fifth-degree polynomial fitting functions in terms of DOD:
$$ U = \sum_{n=0}^{5} a_n (DOD)^n $$
$$ Y = \sum_{n=0}^{5} b_n (DOD)^n $$
Based on the discharge characteristic experiments of the prismatic LiFePO₄ battery, the fitting parameters an and bn were adopted from the reference work. The battery in this study was connected in series with the proposed connection topology.
2.6 Evaluation Indices for Cooling Effectiveness
To comprehensively evaluate the performance of the battery pack cooling system, the following indices are employed:
(1) Maximum temperature Tmax: The highest temperature reached inside the battery pack under specific operating conditions. The normal operating Tmax of a lithium-ion battery should generally not exceed 40 ℃.
(2) Maximum temperature difference ΔT: The difference between the maximum and minimum temperatures inside the battery pack, representing the temperature uniformity:
$$ \Delta T = T_{max} – T_{min} $$
For safe operation, it is commonly desired that ΔT be less than 5 ℃.
(3) Mixing index IOM: Used to evaluate the mixing condition of hot and cold air inside the battery pack. It is defined as:
$$ IOM = \frac{T_{in-max} – T_{in-min}}{T_{out} – T_{in}} $$
A higher IOM value suggests poorer air mixing and a greater tendency to form local hot spots.
(4) Battery pack volumetric energy density ρenergy: The electrical energy stored per unit volume of the battery pack:
$$ \rho_{energy} = \frac{24 \times C \times U}{H \times L \times W} $$
where C is the battery capacity in Ah, U is the nominal voltage in V, and H, L, W are the height, length, and width of the battery pack in m.
2.7 Model Validation
To validate the MSMD-NTGK thermoelectric coupling model, I compared the simulated surface maximum temperature (Tmax) and minimum temperature (Tmin) of a single battery with the experimental data from the literature under discharge rates of 1C and 2C. The comparison results are summarized in the following table:
| Discharge rate | Parameter | Simulated value (℃) | Experimental value (℃) | Error (℃) |
|---|---|---|---|---|
| 1C | Tmax | 35.2 | 31.5 | 3.7 |
| 1C | Tmin | 30.8 | 30.1 | 0.7 |
| 2C | Tmax | 44.9 | 39.9 | 5.0 |
| 2C | Tmin | 36.4 | 35.4 | 1.0 |
Throughout the simulation, the error of Tmax was controlled within 5.16 ℃, and the error of Tmin did not exceed 1.05 ℃. The relative error of the temperature prediction was less than 15% for all operating conditions. This confirms that the established MSMD-NTGK model is capable of accurately predicting the temperature distribution characteristics of the battery under different discharge rates, providing a reliable simulation foundation for the subsequent thermal management research of the battery pack.
3. Air-Cooling Performance Analysis of the Energy Storage Battery Pack
3.1 Model Setup and Boundary Conditions
In this chapter, I adopted a battery pack composed of six prismatic LiFePO₄ batteries connected in series as the research object. The single cell dimensions were 135 mm×180 mm×30 mm (length × width × height). The spacing between adjacent cells was 5 mm, and the distance between the cells and the side walls of the case was 20 mm. The batteries were connected by copper busbars. The overall layout of the battery pack is schematically shown in the model description. The pack adopted a single-direction ventilation design. The cooling airflow entered from one side and exited from the opposite side. In the initial design, the inlet and outlet adopted a single rectangular structure with dimensions of 50 mm×50 mm.
The simulations were performed using the Fluent platform with the MSMD-NTGK thermoelectric coupling model. The physical property parameters of the cell were set according to the data in the previous chapter. The boundary conditions included a velocity inlet of 5 m/s at 25 ℃, pressure outlet at 101.325 kPa and 25 ℃, convective heat transfer coefficient of 5 W/(m²·K) on the external walls, and coupled thermal boundary conditions at internal interfaces. The turbulent flow was simulated with the realizable k-ε model. A transient solver was enabled with an energy equation. The discharge processes at 0.5C, 1C, 2C, and 3C were simulated. The time step was set to 1 s, and the total number of steps was 1440 to cover the complete discharge at 2C rate.
3.2 Grid Independence Verification
To ensure the accuracy of the simulation results while reducing the computational cost, grid independence verification was performed. Five different mesh sizes were evaluated: 290,000; 470,000; 800,000; 910,000; and 1,050,000 cells. The maximum temperature (Tmax) and maximum temperature difference (ΔT) were monitored as the evaluation parameters. When the grid number increased from 290,000 to 800,000, both Tmax and ΔT changed significantly, and an obvious convergence inflection point was observed at 800,000 cells. When the grid was further refined to 910,000 cells, the changes were extremely small (Tmax increased by 0.008 ℃ and ΔT increased by 0.037 ℃). With an additional increase to 1,050,000 cells, the variations became negligible. Therefore, the mesh with approximately 800,000 cells was selected as the optimal configuration, balancing accuracy and computational efficiency.
3.3 Effect of Discharge Rate
The discharge rate determines the current density and therefore directly affects the heat generation rate. In this section, the battery pack was arranged in a linear manner with inlet and outlet along the X axis direction. The results for different discharge rates are summarized below:
| Discharge rate | Tmax (℃) | Tmin (℃) | ΔT (℃) |
|---|---|---|---|
| 0.5C | 25.8 | 25.3 | 0.5 |
| 1C | 27.1 | 25.8 | 1.3 |
| 1.5C | 29.5 | 26.7 | 2.8 |
| 2C | 33.4 | 28.7 | 4.7 |
| 3C | 34.8 | 29.6 | 5.2 |
With the increase in discharge rate from 0.5C to 3C, the maximum temperature rose from 25.8 ℃ to 34.8 ℃, and the maximum temperature difference rose from 0.5 ℃ to 5.2 ℃. The greater the discharge rate, the faster the increase rate of Tmax. The temperature distribution of the battery pack showed a characteristic of high temperature in the middle and low temperature at both sides. The cells close to the inlet window had the lowest temperature because of the large contact area with the cooling airflow. The middle cells suffered from heat accumulation due to adjacent cell heat conduction and uneven flow field distribution. The electrodes and busbars also exhibited considerably higher temperatures due to higher current concentration and ohmic Joule heating. This indicates that the discharge rate is a decisive factor for the cooling performance of the energy storage battery pack.
3.4 Effect of Airflow Organization
The airflow organization method strongly affects the cooling efficiency. I compared two configurations: airflow along the X axis direction (battery width direction) and airflow along the Y axis direction (battery length direction). The simulation results at z=120 mm are presented below:
| Airflow direction | Tmax (℃) | ΔT (℃) | Flow characteristics |
|---|---|---|---|
| X direction | 29.8 | 3.8 | Large vortex region at outlet, significant stagnation |
| Y direction | 28.1 | 2.5 | Fewer vortices, more uniform flow |
When the airflow was along the X direction, a large-area vortex was observed near the outlet plane, which seriously hindered heat dissipation. In contrast, the Y-direction airflow showed better cooling performance. The Tmax was reduced by 1.7 ℃ and the ΔT was reduced by 1.3 ℃. The short-flow-channel and multi-channel design effectively reduced the flow distance and increased the number of flow passages, thereby improving the heat exchange between air and batteries. I therefore adopted the Y-direction airflow as the baseline for subsequent optimizations.
3.5 Effect of Inlet Air Velocity
The inlet air velocity is an important operating parameter of the air-cooling system. I carried out simulations for inlet velocities of 1, 3, 5, and 7 m/s while maintaining the discharge rate at 2C. The results are shown below:
| Inlet velocity (m/s) | Tmax (℃) | Tmin (℃) | ΔT (℃) | Pressure drop (Pa) |
|---|---|---|---|---|
| 1 | 30.0 | 26.6 | 3.4 | 21.7 |
| 3 | 28.9 | 26.2 | 2.7 | 38.5 |
| 5 | 28.2 | 25.7 | 2.5 | 56.9 |
| 7 | 27.8 | 25.4 | 2.3 | 79.6 |
As the inlet velocity increased, the Tmax and ΔT both decreased, indicating improved cooling performance. However, the higher velocity also led to greater pressure drop and thus higher energy consumption and noise. A comprehensive analysis of the cooling effect and power consumption showed that 5 m/s provided the best balance between cooling effectiveness and energy consumption. The cooling performance satisfied the design requirements with Tmax=28.2 ℃ and ΔT=2.5 ℃.
3.6 Effect of the Number of Outlets
The number of air outlets is a key structural parameter affecting the airflow distribution and energy consumption. I compared configurations with one to five outlets (nout=1 to 5) while keeping the total inlet area constant. The results are summarized below:
| Number of outlets | Tmax (℃) | ΔT (℃) | Pressure drop (Pa) |
|---|---|---|---|
| 1 | 28.0 | 2.5 | 610.8 |
| 2 | 28.0 | 2.5 | 190.2 |
| 3 | 28.1 | 2.6 | 78.9 |
| 4 | 28.1 | 2.6 | 57.4 |
| 5 | 28.4 | 2.7 | 41.4 |
The increase in the outlet number significantly reduced the pressure drop from 610.8 Pa to 41.4 Pa, demonstrating the advantage of multi-outlet design in reducing energy consumption. However, the Tmax and ΔT increased slightly as the number of outlets increased. This is caused by the appearance of vortices in some regions when multiple outlets are adopted, resulting in local heat accumulation. Nevertheless, all configurations satisfied the design constraints. Among all configurations, the four-outlet scheme provided the best trade-off between cooling effectiveness and energy consumption; therefore, it was adopted as the reference configuration for further optimization.
4. Optimization of the Air-Cooling System for the Energy Storage Battery Pack
4.1 Linear Arrangement Optimization
4.1.1 Effect of Inlet and Outlet Positions
In this section, I studied the impact of different inlet and outlet positions on the cooling performance. Six different configurations were considered, including Z-middle to Z-middle (case a), Z-middle to Z-upper (case b), Z-middle to Z-lower (case c), Z-lower to Y+ (case d), Z-lower to Y-middle (case e), and Z-lower to Y- (case f). The simulation results for these configurations are presented below:
| Case | Inlet position | Outlet position | Tmax (℃) | ΔT (℃) | Flow characteristics |
|---|---|---|---|---|---|
| a | Z-middle | Z-middle | 28.8 | 3.1 | Uniform airflow, no dead zone |
| b | Z-middle | Z-upper | 30.5 | 4.2 | Heat accumulation at upper part |
| c | Z-middle | Z-lower | 31.2 | 4.6 | Heat accumulation at lower part |
| d | Z-lower | Y+ | 29.6 | 3.8 | Vortex formation |
| e | Z-lower | Y-middle | 30.1 | 4.0 | Poor airflow distribution |
| f | Z-lower | Y- | 32.4 | 5.2 | Cooling dead zone near outlet |
It can be clearly seen that the outlet axis direction and the relative position between the inlet and outlet significantly affect the cooling performance. When the inlet and outlet were located on the same axis, the airflow was more uniform and heat dissipation was more efficient. Case (a), in which the inlet was at Z-middle and the outlet at Z-middle, exhibited the best cooling performance, with the lowest Tmax of 28.8 ℃ and ΔT of 3.1 ℃. Meanwhile, case (f) presented the worst performance because the outlet near the inlet created a cooling dead zone at the remote end of the battery pack.
4.1.2 Effect of Constant-Temperature Radiation Plate Temperature Tr
Through the temperature field analysis of case (a), I found that high-temperature regions were mainly distributed in the upper part of the battery pack. To improve the temperature uniformity and reduce the cell temperature difference, I proposed adding a constant-temperature radiation plate at the top of the battery pack. The plate enhances radiation cooling at the top region, effectively balancing the overall temperature distribution. Under an ambient temperature of 25 ℃, the effect of the radiation plate temperature Tr (20 ℃, 22 ℃, 24 ℃, 26 ℃) was systematically investigated, and the results are listed below:
| Tr (℃) | Tmax (℃) | ΔT (℃) | IOM |
|---|---|---|---|
| Without plate | 30.4 | 3.8 | 1.04 |
| 20 | 29.5 | 3.2 | 6.36 |
| 22 | 29.7 | 3.4 | 2.12 |
| 24 | 30.2 | 3.6 | 1.47 |
| 26 | 30.6 | 3.9 | 0.95 |
When the radiation plate temperature was below the ambient temperature, the cooling effect was enhanced. At Tr=20 ℃, Tmax was reduced by 3.0% and ΔT by 15.8% compared with the case without the radiation plate. However, the IOM value increased to 6.36, indicating poor mixing of hot and cold air and a higher risk of local hot spots. At Tr=24 ℃, the cooling effect was still improved (Tmax reduced by 0.7%, ΔT reduced by 3.6%), while the IOM was 1.47, which is acceptable. Therefore, I chose Tr=24 ℃ as the optimal setting for the constant-temperature radiation plate.
4.1.3 Effect of Ambient Temperature Te
The ambient temperature affects both the battery pack cooling efficiency and the radiation performance of the constant-temperature plate. I set Tr=24 ℃ and varied the ambient temperature Te from 20 ℃ to 24 ℃. The results are given as follows:
| Te (℃) | ΔTr-e (℃) | Tmax (℃) | ΔT (℃) | Total heat dissipated E (W) | Radiation heat Er (W) |
|---|---|---|---|---|---|
| 20 | 4 | 26.15 | 4.50 | 106.8 | 2.4E-3 |
| 21 | 3 | 27.05 | 4.37 | 84.3 | 1.6E-3 |
| 22 | 2 | 27.82 | 4.25 | 61.5 | 0.8E-3 |
| 23 | 1 | 28.20 | 4.12 | 39.4 | 0.3E-3 |
| 24 | 0 | 28.56 | 4.02 | 18.0 | -1.3E-4 |
As ΔTr-e decreased, Tmax increased while ΔT decreased. The total heat dissipated and the radiation heat both monotonically decreased. When ΔTr-e was 0, the radiation plate changed from a cooling element to a heat feedback element, resulting in a continuous increase in Tmax. This indicates that the temperature difference between the radiation plate and the environment is a critical parameter for the effective operation of the radiation cooling system.
4.2 Staggered Arrangement Optimization
4.2.1 Effect of Staggered Distance d1
The linear arrangement of batteries may cause non-uniform airflow distribution and limited heat transfer area. To improve this, I proposed a staggered arrangement. The staggered distance d1 is defined as the relative offset in the Y direction between adjacent rows of batteries. In order to investigate the effect of d1, I fixed the X-direction spacing between batteries at 5 mm and simulated cases with d1 = 0, 3, 5, 7, 10, 15, and 20 mm. The results are presented below:
| d1 (mm) | Tmax (℃) | Tmin (℃) | ΔT (℃) |
|---|---|---|---|
| 0 | 28.2 | 25.7 | 2.5 |
| 3 | 28.2 | 25.6 | 2.6 |
| 5 | 28.2 | 25.6 | 2.6 |
| 7 | 28.1 | 25.5 | 2.6 |
| 10 | 28.1 | 25.8 | 2.3 |
| 15 | 28.1 | 25.5 | 2.6 |
| 20 | 28.1 | 25.4 | 2.7 |
The analysis shows that the ΔT exhibits a camel-back trend with respect to d1, meaning that increasing d1 does not always improve the cooling effect. There exists an optimal value that maximizes the cooling performance. When d1=10 mm, the ΔT reached the minimum value of 2.3 ℃. The streamlines show that with the increase of d1, the high-velocity region inside the battery pack expanded, allowing the cold air to carry away more heat. However, when d1 exceeded 10 mm, vortices began to appear between the electrodes, the electrodes and busbars, weakening the heat removal effect. Therefore, the optimal staggered distance was determined to be 10 mm.
4.2.2 Effect of Battery Spacing d2
With the staggered distance fixed at 10 mm, I further varied the battery spacing d2 in the X direction (0, 3, 5, 7, and 10 mm) to investigate its influence. The results are as follows:
| d2 (mm) | Tmax (℃) | Tmin (℃) | ΔT (℃) |
|---|---|---|---|
| 0 | 29.9 | 26.3 | 3.6 |
| 3 | 28.8 | 26.0 | 2.8 |
| 5 | 28.1 | 25.8 | 2.3 |
| 7 | 28.3 | 25.7 | 2.6 |
| 10 | 28.7 | 25.6 | 3.1 |
As d2 increased from 0 mm to 5 mm, both Tmax and ΔT decreased. However, when d2 exceeded 5 mm, Tmax and ΔT increased again. The reason is that increasing d2 expands the battery pack length, causing less cold air to flow toward both sides of the battery pack, which reduces the average internal airflow velocity and thus weakens the heat removal. Therefore, the optimal battery spacing was 5 mm for the staggered arrangement.
4.3 Aligned Arrangement Optimization
4.3.1 Cooling Effect and Airflow Organization
To further optimize the air-cooling system, I also proposed an aligned arrangement of the batteries. In this configuration, the six batteries were equally divided into two groups, and each group was arranged in a line. The aligned arrangement effectively reduced the number of cells per row, weakening the thermal coupling effect, and shortened the longitudinal dimension of the battery pack, improving the airflow distribution. The comparison of the three arrangements (linear, staggered, and aligned) under their respective optimal or reference conditions is listed below:
| Arrangement | d1 (mm) | d2 (mm) | Tmax (℃) | ΔT (℃) |
|---|---|---|---|---|
| Linear | 0 | 5 | 28.2 | 2.6 |
| Staggered | 10 | 5 | 28.1 | 2.5 |
| Aligned | 0 | 5 | 28.2 | 2.1 |
Although the Tmax values of the three arrangements were similar and all below the safety limit, the temperature uniformity was noticeably different. The aligned arrangement exhibited the best temperature uniformity with ΔT=2.1 ℃, which was 19.2% lower than the linear arrangement and 16% lower than the staggered arrangement. The aligned arrangement shortened the heat transfer path between the battery cells and improved the uniformity of the airflow distribution, reducing the flow short-circuit phenomenon.
4.3.2 Effect of Guide Plate Height
In the aligned arrangement, when the airflow is along the X direction, the temperature distribution is more uniform than in the Y direction, but the maximum temperature is slightly higher due to insufficient cold air inflow into the battery gaps. To solve this problem, I introduced a guide plate at the top of the battery pack. The guide plate changes the flow path and increases the airflow velocity in the gaps. However, the guide plate also creates a vortex on its leeward side, and the size of the vortex increases with the guide plate height. I therefore optimized the guide plate height l by considering four values: 0, 20, 25, and 30 mm. The results are shown below:
| l (mm) | Tmax (℃) | ΔT (℃) |
|---|---|---|
| 0 | 28.2 | 2.2 |
| 20 | 28.2 | 2.1 |
| 25 | 28.2 | 2.0 |
| 30 | 28.2 | 2.2 |
At l=25 mm, the maximum temperature difference reached its minimum of 2.0 ℃, which was 0.5% lower than the case without a guide plate in terms of Tmax and 9.0% lower in terms of ΔT. Therefore, the optimal guide plate height for the aligned arrangement was 25 mm. The guide plate increased the airflow penetration into the gaps, enhancing the heat exchange efficiency, while simultaneously confining the vortex to an acceptable level.
4.4 Comparison of Energy Density of Different Arrangements
When optimizing the air-cooling system, the volumetric energy density ρenergy of the battery pack should also be considered. Increasing the staggered distance d1 or the battery spacing d2 reduces the energy density and thus shortens the endurance capability. The energy densities of the battery packs under different arrangement conditions are compared in the following table:
| Arrangement | d1 (mm) | d2 (mm) | ρenergy (kWh·m-3) | Ratio to single cell ρenergy |
|---|---|---|---|---|
| Linear | 0 | 5 | 360479.81 | 6.84% |
| Staggered | 3 | 5 | 354537.84 | 6.73% |
| Staggered | 5 | 5 | 350684.16 | 6.66% |
| Staggered | 7 | 5 | 346913.37 | 6.59% |
| Staggered | 10 | 5 | 341406.80 | 6.48% |
| Staggered | 15 | 5 | 332607.66 | 6.31% |
| Staggered | 20 | 5 | 324250.68 | 6.16% |
| Staggered | 10 | 3 | 355691.61 | 6.75% |
| Staggered | 10 | 5 | 341406.80 | 6.48% |
| Staggered | 10 | 7 | 328225.07 | 6.23% |
| Staggered | 10 | 10 | 310256.55 | 5.89% |
| Aligned | 0 | 5 | 322008.69 | 6.11% |
The energy density gradually decreased with the increase of d1 and d2. Among the three arrangements, the linear arrangement had the highest energy density, followed by the staggered arrangement, and the aligned arrangement had the lowest energy density. Therefore, for energy storage battery packs with lower discharge rates, adopting a staggered arrangement with optimized staggered distance and battery spacing can improve the cooling effect while maintaining a reasonable energy density.
5. Optimization of a Novel Combined Cooling System
5.1 Optimization of the Guide Plate
For this chapter, I built a battery pack consisting of 12 prismatic LiFePO₄ batteries connected in series, arranged in a linear configuration to preserve a high energy density. The discharge rate was set to 3C to reflect high-rate heat generation in practical energy storage applications. The inlet was a 100 mm × 100 mm square structure centrally located, and the outlet system consisted of four 50 mm × 50 mm square outlets uniformly distributed on the outlet surface. This setup was used as the research platform for the novel combined cooling system.
5.1.1 Effect of Guide Plate Angle
The mounting angle α of the guide plate is a critical parameter that influences the airflow path and velocity distribution. I simulated six different angles: 15°, 30°, 45°, 60°, 75°, and 90°, together with the baseline case without a guide plate. The results are summarized below:
| α (°) | Tmax (℃) | Taver (℃) | ΔT (℃) | IOM | ΔP (Pa) |
|---|---|---|---|---|---|
| Without guide plate | 33.2 | 31.5 | 5.63 | 1.08 | 79.7 |
| 15 | 33.1 | 31.3 | 5.42 | 0.74 | 78.9 |
| 30 | 33.0 | 31.2 | 5.38 | 0.66 | 78.6 |
| 45 | 33.1 | 31.4 | 5.53 | 0.85 | 79.0 |
| 60 | 33.1 | 31.4 | 5.57 | 0.88 | 79.3 |
| 75 | 33.2 | 31.5 | 5.60 | 0.93 | 79.5 |
| 90 | 33.2 | 31.5 | 5.63 | 1.02 | 79.8 |
It is clear that all configurations with the guide plate achieved a lower ΔT and IOM compared with the baseline. Among them, α=30° gave both the lowest maximum temperature difference (5.38 ℃) and the lowest IOM (0.66), indicating that the airflow distribution was effectively homogenized and local overheating was avoided. Thus, an angle of 30° was determined to be optimal.
5.1.2 Effect of Guide Plate Thickness
With the angle fixed at 30°, I then optimized the thickness d of the guide plate. The values of 2 mm, 4 mm, 6 mm, 8 mm, and 10 mm were considered. The evaluation indices are presented below:
| d (mm) | ΔT (℃) | IOM |
|---|---|---|
| 2 | 5.42 | 0.68 |
| 4 | 5.40 | 0.68 |
| 6 | 5.39 | 0.67 |
| 8 | 5.36 | 0.66 |
| 10 | 5.34 | 0.65 |
As the thickness increased, the flow became more organized, the vortices were reduced, and the heat exchange efficiency was enhanced. The best performance was observed at d=10 mm, where ΔT reached the minimum value of 5.34 ℃ and the IOM was 0.65. The thicker guide plate redirected more airflow into the battery gaps and improved the uniformity of the velocity distribution.
5.1.3 Effect of Guide Plate Height
Finally, with α=30° and d=10 mm, the guide plate height l was varied from 20 mm to 35 mm. The evaluation results are listed below:
| l (mm) | ΔT (℃) | IOM |
|---|---|---|
| 20 | 5.40 | 1.22 |
| 25 | 5.30 | 1.10 |
| 30 | 5.36 | 1.16 |
| 35 | 5.61 | 1.30 |
The ΔT first decreased and then increased with the guide plate height. The lowest ΔT of 5.30 ℃ was obtained at l=25 mm, where the IOM was also relatively low. Therefore, l=25 mm was selected as the optimal guide plate height for the novel cooling system.
5.2 Coupled Optimization of the Guide Plate and the Constant-Temperature Radiation Plate
After obtaining the optimal guide plate parameters (α=30°, d=10 mm, l=25 mm), I further coupled the guide plate with a constant-temperature radiation plate set at 24 ℃. This combination integrates forced convection with radiation heat transfer, resulting in a new type of combined cooling system. The performance of the combined system was evaluated and compared with the baseline case without either the guide plate or the radiation plate. The comparative results are presented in the table below:
| Configuration | Tmax (℃) | Tmin (℃) | Taver (℃) | ΔT (℃) | ΔP (Pa) |
|---|---|---|---|---|---|
| Without guide plate and radiation plate | 33.2 | 27.3 | 31.5 | 5.63 | 79.65 |
| With guide plate and 24 ℃ radiation plate | 32.5 | 27.6 | 31.3 | 4.95 | 70.54 |
The combined cooling system reduced Tmax by 2.1%, while ΔT was reduced from 5.63 ℃ to 4.95 ℃, a decrease of 12.1%. The temperature distribution became more uniform, which is beneficial for the electrochemical performance and lifetime of the battery. The pressure drop ΔP was also reduced from 79.65 Pa to 70.54 Pa, a decrease of 11.3%, indicating that the combined system reduced the energy consumption of the air-cooling system while improving its cooling performance. This confirms the effectiveness of the coupled utilization of the guide plate and the constant-temperature radiation plate in energy storage battery thermal management.
6. Conclusions and Outlook
In this dissertation, I carried out a systematic numerical study on the air-cooling system optimization for energy storage battery packs based on a novel MSMD-NTGK thermoelectric coupling model. The main conclusions are as follows:
(1) The MSMD-NTGK thermoelectric coupling model was validated by comparing the simulated temperature with the experimental data of a single prismatic LiFePO₄ battery under 1C and 2C discharge rates. The maximum error of Tmax was 5.16 ℃, and the error of Tmin was within 1.05 ℃, confirming the reliability of the model.
(2) The heat generation of the energy storage battery increases significantly with the discharge rate. At 3C discharge, the maximum temperature reached 34.8 ℃ and ΔT reached 5.2 ℃. The airflow organization has a significant effect on the cooling performance; the short-flow-channel and multi-channel design greatly improves the cooling effect.
(3) Increasing the inlet air velocity improves the cooling effect but also increases energy consumption and noise. An inlet velocity of 5 m/s was determined to be the optimal balance. The number of outlets also plays a significant role; the four-outlet configuration reduced the pressure drop by 90.6% while maintaining the temperature within the required range.
(4) The constant-temperature radiation plate at the top of the battery pack effectively improved the temperature uniformity, but its cooling efficiency depends on the temperature difference between the plate and the environment. The optimal temperature setting should be slightly below the ambient temperature.
(5) For the staggered arrangement, the optimal staggered distance and battery spacing were found to be 10 mm and 5 mm, respectively. Under this condition, Tmax=28.1 ℃ and ΔT=2.5 ℃, which were 0.5% and 4.5% lower than those of the linear arrangement with the same spacing.
(6) The aligned arrangement showed the best cooling performance among the three arrangements, with ΔT=2.1 ℃, but it had the lowest volumetric energy density. The installation of a guide plate in the aligned arrangement further improved the cooling effect; the optimal guide plate height was 25 mm.
(7) For the 12-cell battery pack under 3C discharge, the optimal guide plate parameters were determined as α=30°, d=10 mm, and l=25 mm. Coupling the optimized guide plate with a 24 ℃ constant-temperature radiation plate reduced the maximum temperature difference by 12.1% and the pressure drop by 11.3% compared with the baseline without these components.
Future work should focus on building a dedicated experimental platform to further validate and refine the MSMD-NTGK model parameters, incorporating the effect of battery aging on heat generation, and integrating electrochemical impedance spectroscopy (EIS) with the thermal management system for real-time monitoring of the internal state of the energy storage battery pack.