Lithium-ion battery liquid cooling thermal management system and its multi-objective optimization design method

Abstract

This invention relates to the field of lithium-ion battery thermal management technology, and discloses a lithium-ion battery liquid-cooled thermal management system and its multi-objective optimization design method. The method includes: establishing and verifying an electro-thermal-flow coupled simulation model; determining design variables and sampling to obtain a performance index sample set; training a surrogate model to establish a fast mapping from variables to indices; and performing multi-objective optimization based on the surrogate model to obtain a Pareto solution set and determine the final solution. The system is an entity designed using this method, including a battery module, a bottom liquid-cooling plate with built-in liquid-cooling pipes, and independent liquid-cooling pipes surrounding the outside of the module. This invention solves the problem of coordinating the optimization of multiple objectives such as temperature uniformity, energy consumption, and structural compactness in liquid-cooling system design by integrating simulation, surrogate models, and multi-objective optimization. It achieves a leap from empirical trial and error to automatic global optimization, and the designed dual-loop cooling system significantly reduces energy consumption and material costs while ensuring heat dissipation safety.

Description

Technical Field

[0001] This invention relates to the field of lithium-ion battery thermal management technology, specifically to a lithium-ion battery liquid cooling thermal management system having an outer liquid cooling circuit and a bottom liquid cooling circuit, and a multi-objective optimization design method for the system. Background Technology

[0002] With the advancement of global energy structure transformation, lithium-ion batteries, due to their high energy density and long cycle life, have become the core energy carrier in electric vehicles and electrochemical energy storage systems. However, the performance, lifespan, and safety of lithium-ion batteries are highly sensitive to their operating temperature. Ideally, the battery pack temperature should be maintained within a narrow range of 15℃ to 40℃, and the maximum internal temperature difference should not exceed 5℃. If thermal management fails, the battery will operate under high, low, or uneven temperature conditions, which will not only cause irreversible aging such as a sudden decrease in capacity and a surge in internal resistance, but in severe cases, it may also lead to local overheating and trigger a chain of exothermic reactions, i.e., thermal runaway, which may then lead to combustion or even explosion. Therefore, an efficient and reliable thermal management system is an indispensable key technology for ensuring the safe, durable, and efficient operation of lithium-ion battery systems.

[0003] Among numerous thermal management technologies, liquid cooling, with its high specific heat capacity and excellent thermal conductivity, can efficiently remove the large amount of heat generated by batteries, making it the mainstream solution for addressing the heat dissipation needs of medium-to-high power density battery packs. Typical liquid cooling system design involves the coordination of multiple parameters, including cooling channel structure, coolant flow rate, and inlet temperature. This process heavily relies on numerical simulation methods such as computational fluid dynamics to predict the temperature field, flow field, and pressure drop. However, current design methods in engineering practice have significant limitations:

[0004] On the one hand, the design process still largely relies on engineers' experience-based trial and error or limited parameter scanning, which is inefficient and makes it difficult to find the global optimal solution in a complex multidimensional design space;

[0005] On the other hand, even with simulation assistance, traditional optimization approaches often focus on a single performance objective (such as pursuing only the lowest and highest temperatures). However, practical engineering applications must comprehensively balance multiple conflicting indicators such as temperature uniformity, system pumping energy consumption (manifested as pressure drop), manufacturing cost, and structural compactness (related to material volume). This disconnect between single-objective optimization and multi-dimensional practical needs often results in thermal management systems that excel in one indicator but sacrifice other key performance aspects, thus limiting further improvements in the overall performance of the battery system and the expansion of potential application scenarios.

[0006] Therefore, there is an urgent need for a design methodology that can systematically and efficiently optimize multiple conflicting objectives in a synergistic manner to guide the development of next-generation battery thermal management systems that achieve the best balance between safety, energy efficiency, and cost. Summary of the Invention

[0007] The present invention aims to solve the problem that existing lithium-ion battery liquid cooling thermal management systems are difficult to systematically weigh and optimize multiple key thermal management indicators under the condition of multi-cooling loop coordination during the design of structural and operational parameters.

[0008] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0009] On one hand, the present invention provides a multi-objective optimization design method for a lithium-ion battery liquid cooling thermal management system, comprising the following steps:

[0010] S1. Obtain the structural parameters, material thermal properties, and cell operating parameters of the system to be designed, and establish an electro-thermal-fluid coupled simulation model that includes an electrochemical heat generation sub-model, a fluid flow sub-model, and a heat transfer sub-model. The heat generation rate output by the electrochemical heat generation sub-model is used as the heat source input of the heat transfer sub-model and participates in the coupled solution together with the fluid flow sub-model.

[0011] S2. The coupled simulation model is validated based on the experimentally measured discharge temperature data;

[0012] S3. Determine the set of design variables and their range of values. The set of design variables shall include at least the coolant inlet volumetric flow rate. Contact half angle between liquid cooling plate and aluminum sleeve Aspect ratio of liquid cooling tubes and aluminum sleeve thickness And generate a set of sample points within the range of values;

[0013] S4. Apply the coupled simulation model to each sample point to obtain the corresponding performance indicators. The performance indicators should include at least the highest temperature. Maximum temperature difference Differential pressure between the inlet and outlet of the liquid cooling pipe and aluminum volume ;

[0014] S5. Train the surrogate model based on the sample point set and the corresponding performance index to obtain the mapping relationship between design variables and performance indexes, and determine the target surrogate model for optimization based on the evaluation index.

[0015] S6. Perform multi-objective optimization based on the objective surrogate model to obtain the Pareto solution set;

[0016] S7. Based on the multi-objective decision rule, determine the final design scheme from the Pareto solution set, and substitute the final design scheme back into the coupled simulation model to obtain the performance index output corresponding to the final design scheme.

[0017] As an optional implementation, the sample point set generated in step S3 is generated using Latin hypercube sampling.

[0018] As an optional implementation, step S2 includes: collecting experimentally measured discharge temperature data under at least two discharge rate conditions, and comparing the temperature curve output by the coupled simulation model with the experimentally measured discharge temperature data to complete the verification of the coupled simulation model.

[0019] As an optional implementation, in step S5, the evaluation index includes at least the coefficient of determination R. 2 And / or error-related metrics, and compare the prediction accuracy of different surrogate models based on the evaluation metrics to determine the target surrogate model.

[0020] As an optional implementation, in step S5, the surrogate model includes at least one of the following: Gaussian process regression model (GPR), response surface model (RSM), radial basis function model (RBF), backpropagation neural network model (BPNN), or support vector regression model (SVR).

[0021] As an optional implementation, in step S6, the multi-objective optimization employs an evolutionary algorithm to obtain the Pareto solution set. The evolutionary algorithm includes at least the non-dominated sorting genetic algorithm NSGA-II and / or NSGA-III, and / or the multi-objective particle swarm optimization algorithm MOPSO, and / or the decomposition-type multi-objective evolutionary algorithm MOEA / D.

[0022] As an optional implementation, in step S7, the multi-objective decision rule includes: normalizing the objective indices in the Pareto solution set, and calculating the distance of each Pareto solution relative to the ideal solution based on the weights to form a ranking result, thereby selecting the final design scheme.

[0023] On the other hand, the present invention also provides a lithium-ion battery liquid cooling thermal management system designed using the above method, comprising: a battery module, a liquid cooling plate, a first liquid cooling pipe, a second liquid cooling pipe, an aluminum sleeve, and an aluminum block heat sink.

[0024] The battery module includes multiple cylindrical cells, and the aluminum sleeve is respectively sleeved on the outer periphery of each cylindrical cell;

[0025] The aluminum block heat sink is disposed between adjacent aluminum sleeves and is thermally connected to the adjacent aluminum sleeves.

[0026] Multiple sets of the second liquid cooling tubes are arranged along the height direction of the cylindrical cell, and all are arranged close to the outer circumference of the aluminum sleeve, forming a connecting transition section between adjacent aluminum sleeves, so that the second liquid cooling tubes form a continuous loop around the outside of the battery module, and the second liquid cooling tubes are provided with an outer flow inlet and an outer flow outlet.

[0027] The liquid cooling plate is disposed at the bottom of the battery module, the first liquid cooling pipe is independently arranged in the liquid cooling plate and forms a circuit, and the first liquid cooling pipe is provided with a bottom flow inlet and a bottom flow outlet.

[0028] As an optional implementation, the second liquid cooling pipe sequentially wraps around the outer periphery of each aluminum sleeve along the outside of the battery module, and the adjacent wrapping sections are connected by a bend transition section located between adjacent aluminum sleeves, thereby forming a wave-shaped wrapping path.

[0029] Furthermore, the outer flow inlet and the outer flow outlet are located on the same side of the battery module.

[0030] As an optional implementation, the first liquid cooling pipe is arranged in a U-shaped loop within the liquid cooling plate, the bottom flow inlet and the bottom flow outlet are located on the same side of the liquid cooling plate, and the bend of the U-shaped loop is located on the side away from the bottom flow inlet and the bottom flow outlet.

[0031] This invention fundamentally changes the traditional design model by constructing an integrated framework that combines high-precision electro-thermal-fluid coupling simulation, surrogate models, and multi-objective optimization algorithms. It successfully achieves collaborative automatic optimization of multiple conflicting objectives such as battery maximum temperature, maximum temperature difference, system voltage drop, and material volume. By replacing time-consuming simulations with surrogate models, it achieves millisecond-level performance prediction and uses multi-objective algorithms for global optimization to obtain Pareto optimal solutions. In theory, this breaks through the limitations of single-objective optimization, and in practice, it significantly improves design efficiency and the overall performance of the solution.

[0032] Furthermore, the liquid-cooled thermal management system designed based on this optimization method is a novel dual-loop cooling architecture that combines a bottom liquid cooling plate with an internal flow channel and an external independent liquid cooling pipe. This unique structure is the physical manifestation of the optimization algorithm after global optimization of structural parameters such as contact half-angle and flow channel width-to-height ratio. Its advantages are directly translated into engineering effectiveness: it can effectively control the flow resistance (energy consumption) of the cooling system while ensuring excellent temperature uniformity of the battery module and reducing the amount of aluminum used to achieve lightweighting and cost control. Attached Figure Description

[0033] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below.

[0034] Figure 1 This is a flowchart illustrating the multi-objective optimization design method for a lithium-ion battery liquid cooling and thermal management system provided in an embodiment of the present invention.

[0035] Figure 2This is a characteristic curve showing the relationship between terminal voltage and state of charge (SOC) during the discharge process of a single battery in an embodiment of the present invention;

[0036] Figure 3 This is a graph showing the variation of ohmic internal resistance, polarization internal resistance and total internal resistance of a single cell under different states of charge in an embodiment of the present invention.

[0037] Figure 4 The entropy-thermal coefficient of a single cell under different states of charge in the embodiments of the present invention is ( The curve showing the change of )

[0038] Figure 5 This is an isometric schematic diagram of the overall structure of the lithium-ion battery liquid cooling and thermal management system provided in an embodiment of the present invention;

[0039] Figure 6 This is a schematic diagram of the front cross-sectional structure of the lithium-ion battery liquid cooling and thermal management system provided in an embodiment of the present invention;

[0040] Figure 7 This is a top-view cross-sectional view of the lithium-ion battery liquid cooling thermal management system provided in an embodiment of the present invention, mainly showing the bypass path of the second liquid cooling pipe;

[0041] Figure 8 This is a perspective view of the flow channel layout of the first liquid cooling pipe inside the liquid cooling plate provided in an embodiment of the present invention;

[0042] Figure 9 This is a schematic diagram of the mesh generation of the simulation model established for the liquid cooling and thermal management system of a lithium-ion battery in an embodiment of the present invention;

[0043] Figure 10 This is a comparison and verification curve of the temperature predicted by the simulation model and the experimentally measured temperature under different discharge rates in the embodiments of the present invention.

[0044] In the diagram: 1. Liquid cooling plate; 2. First liquid cooling pipe; 21. Bottom flow inlet; 22. Bottom flow outlet; 3. Second liquid cooling pipe; 31. Outer flow inlet; 32. Outer flow outlet; 4. Aluminum sleeve; 5. Aluminum block heat sink; 6. Battery cell. Detailed Implementation

[0045] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0046] Example 1:

[0047] This embodiment provides a multi-objective optimization design method for a lithium-ion battery liquid cooling thermal management system. This method aims to solve the technical problems of excessive reliance on empirical trial and error in existing design processes, and the inability of single-objective optimization to simultaneously address multiple conflicting indicators such as temperature uniformity, energy consumption, and structural compactness. By constructing an integrated framework of "simulation-proxy-optimization," global optimization of design variables is achieved. Figure 1 As shown, the method mainly includes the following steps:

[0048] First, step S1 is executed to obtain the basic parameters of the system to be designed and establish a coupled model. Specifically, the designer first needs to clarify the geometric dimensions of the lithium-ion battery module to be designed, the electrochemical parameters of the selected battery cells, and the thermophysical parameters of the relevant materials. In this embodiment, we take a 2-parallel 3-series module composed of a 37.1Ah ternary 4695 large cylindrical battery for automotive use as an example. The specific battery technical parameters are shown in Table 1 below.

[0049] Table 1: Battery Technical Parameters

[0050]

[0051] Among them, the rated capacity of the battery It is 37.1Ah, rated voltage It is 3.59V, cutoff voltage It is 2.50V, and the battery height is... The battery radius is 96.6mm. It is 46mm.

[0052] Simultaneously, the physical properties of each component material in the model also need to be obtained, as shown in Table 2 below. The density of the battery cell is 2640 kg / m³. 3 The heat capacity is 895 J / (kg·K), and the thermal conductivity is 1.65 W / (m·K) in the translayer direction and 19.7 W / (m·K) in the interlayer direction, exhibiting anisotropic thermal conductivity characteristics; the density of the coolant (water) is 997.56 kg / m³. 3 Its heat capacity is 4180 J / (kg·K), its thermal conductivity is 0.62 W / (m·K), and its dynamic viscosity is 8.89 × 10⁻⁶. -4 Pa·s; The density of aluminum (used in liquid cooling plates, sleeves, and heat sinks) is 2700 kg / m³. 3 Its heat capacity is 900 J / (kg·K) and its thermal conductivity is 238 W / (m·K).

[0053] Table 2: Model Material Parameters

[0054]

[0055] Furthermore, the dynamic characteristic parameters of the battery are obtained through experimental measurements, such as... Figure 2 The figure shows the curve of terminal voltage as a function of state of charge (SOC). Figure 3 The curves showing the changes in ohmic internal resistance, polarization internal resistance, and total internal resistance with SOC are presented. Figure 4 It shows the entropy heat coefficient ( The curve of the change with SOC is shown. Based on the above parameters, an electro-thermal-fluid coupled simulation model is established in multiphysics simulation software (such as COMSOL), which includes an electrochemical heat generation sub-model, a fluid flow sub-model, and a system heat transfer sub-model. In this coupled model, the electrochemical heat generation sub-model calculates the heat generation rate based on the real-time operating state of the battery (such as current, SOC, and temperature), and this heat generation rate is used as a heat source input into the heat transfer sub-model; the fluid flow sub-model calculates the velocity field and pressure distribution of the cooling medium in the flow channel and couples it to the heat transfer sub-model to calculate the convective heat transfer coefficient; the heat transfer sub-model finally calculates the transient temperature field distribution in the system. Figure 9 The diagram shows the mesh generation of the coupled simulation model, demonstrating how a reasonable mesh generation ensures a balance between computational accuracy and efficiency.

[0056] Subsequently, step S2 is executed to experimentally verify the coupled simulation model. To ensure that the simulation model accurately reflects the thermal behavior of the physical system, it must be validated using experimental data. In this embodiment, data on the change of battery surface temperature over time are collected under three different discharge rate conditions: 1C, 1.5C, and 2C. Figure 10 As shown in the figure, the temperature curve output by the coupled simulation model is compared with the experimentally measured discharge temperature data. It can be seen from the figure that the simulated curve and the experimental point are in extremely high agreement; the temperature rise trend and the final temperature value are very close, and the error is within the acceptable range for engineering applications. This fully demonstrates that the established electro-thermal-flow coupled model has extremely high accuracy and reliability, and can serve as a basic tool for subsequent optimization design.

[0057] Next, step S3 is executed to determine the set of design variables and their value ranges, and sampling is performed. To optimize the system, parameters that significantly impact system performance need to be selected as design variables. In this embodiment, four key design variables are selected: coolant inlet volumetric flow rate... Contact half angle between liquid cooling plate and aluminum sleeve Aspect ratio of liquid cooling tubes (i.e., the width-to-height ratio) and the thickness of the aluminum sleeve. Based on engineering experience and installation space limitations, the value ranges for each variable were determined, as shown in Table 3 below.

[0058] Table 3: Range of Design Variables

[0059]

[0060] Among them, the coolant inlet volumetric flow rate The value ranges from 1.6 to 2.2 cm. 3 / s, for example, you can choose 1.6cm. 3 / s, 1.9cm 3 / s or 2.2cm 3 / s; contact half angle The value range is from 55° to 75°, for example, 55°, 65° or 75° can be selected; the aspect ratio of the liquid cooling pipe. The value range is from 0.6 to 1.4, for example, 0.6, 1.0, or 1.4 can be selected; aluminum sleeve thickness The value range is 1.5 to 3.5 mm, for example, 1.5 mm, 2.5 mm, or 3.5 mm can be selected. To obtain representative sample points in the multidimensional design space, this embodiment employs the Latin hypercube sampling method. Latin hypercube sampling divides the value range of each variable into several equally probable intervals, randomly selects a sample point within each interval, and finally randomly combines the samples from each dimension, thereby ensuring a uniform distribution of sample points throughout the entire design space. This method generated a set of 200 initial sample points.

[0061] Then, step S4 is executed to calculate the performance indicators of the sample points. The design variable values ​​corresponding to the 200 sample points generated in step S3 are sequentially input into the coupled simulation model verified in step S2 for calculation. For each sample point, the key performance indicators of the system under 2C discharge conditions are extracted, including the maximum temperature of the battery module. Maximum temperature difference Differential pressure between the inlet and outlet of the liquid cooling pipe and aluminum volume .in, and It characterizes heat dissipation performance and temperature uniformity, which are directly related to battery safety and lifespan; It characterizes the flow resistance and is directly related to pumping power consumption (energy consumption). This characterizes the system's lightweight nature and material costs. Through simulation calculations, a dataset containing 200 sets of "input variable-output indicators" was obtained.

[0062] Next, step S5 is executed to train and select the optimal surrogate model. Since directly calling the CFD coupled model for multi-objective optimization is computationally extremely expensive and time-consuming, this embodiment introduces surrogate model technology to replace costly simulation calculations. The dataset obtained in step S4 is randomly divided into a training set (e.g., 80% of the data) and a test set (e.g., 20% of the data). In this embodiment, five mainstream surrogate models are constructed and trained: Gaussian Process Regression (GPR), Response Surface Model (RSM), Radial Basis Function Model (RBF), Backpropagation Neural Network (BPNN), and Support Vector Regression (SVR). The training set is used to fit the parameters of these five models, establishing a mapping relationship between design variables and performance indicators. Subsequently, the accuracy of the trained model is evaluated using the test set, mainly referring to the coefficient of determination (R²). 2 Indicators such as ) are used. Table 4 below shows the comparison of the coefficients of determination of the five surrogate models on the prediction set data. As can be seen from the data in Table 4, the GPR model has better prediction performance. , , and R on four indicators 2 The values ​​were 98.09%, 97.69%, 98.85%, and 99.99993%, respectively, and its overall prediction accuracy was superior to that of the RSM, RBF, BPNN, and SVR models. Therefore, this embodiment ultimately selected the GPR model as the target surrogate model for subsequent multi-objective optimization. This surrogate model can accurately predict the system performance under given design variables within milliseconds, greatly improving optimization efficiency.

[0063] Table 4: Coefficients of determination (R²) of the prediction results for the five surrogate models on the prediction set data 2 )

[0064]

[0065] Next, step S6 is executed, performing multi-objective optimization based on the target surrogate model. The GPR surrogate model is used as the fitness function, and a multi-objective evolutionary algorithm is employed to perform global optimization within the design variable space. This embodiment compares four advanced multi-objective optimization algorithms: Non-Dominated Sorting Genetic Algorithm II (NSGA-II), Non-Dominated Sorting Genetic Algorithm III (NSGA-III), Multi-Objective Particle Swarm Optimization (MOPSO), and Decomposition-based Multi-Objective Evolutionary Algorithm (MOEA / D). The optimization objective is to simultaneously minimize... , , and After iterative calculations, each algorithm obtained a set of Pareto optimal solutions. Table 5 below lists the performance comparison of the optimal solution sets obtained by each algorithm after decision processing. Among them, the NSGA-II algorithm performs excellently in terms of solution distribution and convergence, and its obtained optimal solution can... Controlled at 38.629℃, The temperature was reduced to 3.3926℃ while maintaining a low pressure drop of 46.607 Pa and a small aluminum volume of 0.001903 m³. 3 .

[0066] Table 5: Results of Multi-Objective Optimization

[0067]

[0068] Note: In Table 5, "-" indicates that multi-objective optimization was not used, and the corresponding design variables are... , , , They are 1.7cm respectively. 3 / s, 60°, 1, 3mm.

[0069] Finally, step S7 is executed to determine the final design scheme. Since the Pareto solution set contains multiple non-dominated solutions, a decision needs to be made based on actual requirements. This embodiment uses an entropy-weighted approximation ideal solution ranking method to rank the Pareto solution set obtained by NSGA-II. The entropy-weighted method can objectively determine the weight of each indicator based on the dispersion of the data itself, avoiding the subjectivity of manually assigning weights. By calculating the Euclidean distance between each solution and the ideal solution (the hypothetical solution where all indicators are optimal) and the negative ideal solution, the relative proximity is calculated and ranked. Finally, the solution with the highest comprehensive score is selected as the final design scheme.

[0070] In the verification case of this embodiment, the final design parameters after the above process optimization are: coolant inlet volumetric flow rate. It is 2.0522cm 3 / s, contact half angle The angle is 67.963°, and the aspect ratio of the liquid cooling pipe is [missing information]. The thickness of the aluminum sleeve is 1.2731. The thickness is 2.5380 mm. Compared to the unoptimized initial design, the optimized design achieves better performance at the highest temperature. The temperature decreased by 0.973% (approximately 0.36℃), and more importantly, it decreased at the maximum temperature difference. The volume of aluminum material decreased significantly by 7.917%. It decreased by 3.230%, while the pressure drop... The pressure increased only slightly by about 14 Pa, which is well within the redundancy range of the system's pumping capacity. This indicates that the method has successfully found an excellent balance between improving heat dissipation uniformity, reducing weight, and controlling energy consumption.

[0071] In summary, the method provided in this embodiment, by integrating high-precision coupled simulation, machine learning surrogate models, and evolutionary algorithms, breaks through the limitations of traditional single-objective design. It can quickly and accurately provide the optimal design parameters for the overall performance of lithium-ion battery liquid cooling systems, significantly shortening the R&D cycle and improving product performance.

[0072] Example 2:

[0073] This embodiment also provides a lithium-ion battery liquid cooling thermal management system designed and optimized based on the method described in Embodiment 1 above. This system employs a unique dual-loop cooling architecture, combining a bottom liquid cooling plate with independent side liquid cooling pipes to achieve efficient heat dissipation for the large cylindrical battery cells.

[0074] like Figure 5 As shown, the lithium-ion battery liquid cooling thermal management system mainly includes: a liquid cooling plate 1, a first liquid cooling pipe 2, a second liquid cooling pipe 3, an aluminum sleeve 4, an aluminum block heat sink 5, and a battery module as the object to be cooled. The battery module consists of multiple cylindrical cells 6 (in this embodiment, there are 6 46950 cells arranged in 2 parallel and 3 series).

[0075] Each cylindrical battery cell 6 is tightly fitted with an aluminum sleeve 4 around its outer periphery. The aluminum sleeve 4 serves two purposes: firstly, it acts as a heat-conducting medium between the battery cell 6 and the external heat dissipation structure; secondly, it provides structural support and protection. The thickness of the aluminum sleeve 4... This is one of the key parameters optimized by the method described in Example 1. In this example, it is preferably 2.5380 mm. This thickness ensures sufficient heat capacity and thermal conductivity while avoiding excessive material waste. An aluminum block heat sink 5 is provided between adjacent aluminum sleeves 4. The aluminum block heat sink 5 enhances heat conduction within the module by filling the gaps between the battery cells and making thermally conductive connections with adjacent aluminum sleeves 4 (e.g., through welding or integral molding). This helps to homogenize temperature differences between the battery cells and prevents the accumulation of localized hot spots.

[0076] To achieve efficient cooling of the battery, this system is designed with two independent liquid cooling circuits.

[0077] The first circuit is located at the bottom of the module. For example... Figure 6 and Figure 8 As shown, the liquid cooling plate 1 is horizontally positioned at the bottom of the battery module, maintaining good thermal contact with the bottom surface of each cylindrical cell 6. The liquid cooling plate 1 has internal flow channels, within which the first liquid cooling pipe 2 is arranged. Figure 8As clearly shown, the first liquid cooling pipe 2 is arranged in a U-shaped loop within the liquid cooling plate 1. This U-shaped loop has a bottom flow inlet 21 and a bottom flow outlet 22, both of which are located on the same side of the liquid cooling plate 1 (e.g., the left side in the figure). The first liquid cooling pipe 2 extends from the bottom flow inlet 21 into the plate, extends straight to the far end, then turns around through a semi-circular bend (i.e., a return bend), and then extends straight back to the bottom flow outlet 22. This U-shaped design is simple in structure, has low flow resistance, and can effectively remove the heat generated at the bottom of the battery, preventing heat buildup at the bottom.

[0078] The second circuit is located on the side of the module. For example... Figure 5 , Figure 6 and Figure 7 As shown, multiple second liquid cooling tubes 3 are arranged in layers along the height direction of the cylindrical battery cell 6. In this embodiment, multiple sets (e.g., 7 sets) of parallel second liquid cooling tubes 3 are arranged along the height direction, each set being arranged close to the outer circumference of the aluminum sleeve 4. Figure 7 As shown, the second liquid cooling pipe 3 is not a simple straight pipe, but is designed as a single continuous wavy pipe. This pipe sequentially wraps around the outer periphery of each aluminum sleeve 4 along the outside of the battery module, allowing the coolant to cover a large area of ​​the battery side. To achieve continuous wrapping, the second liquid cooling pipe 3 is connected between adjacent aluminum sleeves 4 by bending transition sections, thus forming a wavy, surrounding path that closely follows the contour of the module. This path maximizes the contact area between the cooling pipe and the aluminum sleeve 4, i.e., the contact half-angle. Optimized contact half-angle The angle is 67.963°, which means that most of the circumference of each cell is covered by liquid cooling pipes, significantly improving heat exchange efficiency. The second liquid cooling pipe 3 has an outer flow inlet 31 and an outer flow outlet 32, and the outer flow inlet 31 and the outer flow outlet 32 ​​are also located on the same side of the battery module, which facilitates pipe integration.

[0079] Furthermore, the cross-sectional shape of the liquid cooling pipe is also a key factor affecting heat transfer and flow resistance. In this embodiment, the aspect ratio of the liquid cooling pipe (especially the second liquid cooling pipe 3) is... (Right now Figure 6 and Figure 7 The width of the liquid cooling pipe is marked in the middle. With height The ratio (total flow rate) has been precisely optimized and set to 1.2731. This slightly flattened cross-sectional shape, compared to a square or circular cross-section, provides a larger contact heat transfer area for the same flow area, while the increase in flow resistance remains within a controllable range. Coolant inlet volumetric flow rate It was set to 2.0522cm 3 / s, this flow rate ensures that the cooling medium can remove heat without causing excessive pumping pressure loss.

[0080] The system works as follows: When the battery module generates heat, the heat is first transferred to the aluminum sleeve 4 and the bottom liquid cooling plate 1 through the cylindrical battery cell 6. Due to the high thermal conductivity of aluminum, the heat dissipates rapidly. At this time, coolant enters the first liquid cooling pipe 2 and the second liquid cooling pipe 3 from the bottom inlet 21 and the outer inlet 31, respectively. The coolant at the bottom flows through a U-shaped loop, carrying away the heat from the bottom surface; the coolant on the side flows through a wave-shaped loop, efficiently absorbing the heat from the side. The aluminum block heat sink 5 helps to balance the temperature difference between the battery cells. Finally, the coolant carrying heat flows out from the bottom outlet 22 and the outer outlet 32, completing the heat exchange cycle.

[0081] The lithium-ion battery liquid cooling thermal management system in this embodiment employs structural parameters optimized by a multi-objective genetic algorithm, achieving a compact layout. The combination of corrugated side tubes and U-shaped bottom tubes constructs a comprehensive three-dimensional cooling network. Experimental and simulation results both demonstrate that under 2C high-rate discharge, this system can control the battery's maximum temperature to approximately 38.6℃, with a maximum temperature difference within 3.4℃, far superior to traditional designs, while also using less material and consuming less energy. This system is not only suitable for automotive power battery modules but also for energy storage battery modules with extremely high heat dissipation requirements, possessing broad industrial application value.

Claims

1. A multi-objective optimization design method for a lithium-ion battery liquid cooling thermal management system, characterized in that, Includes the following steps: S1. Obtain the structural parameters, material thermal properties parameters and cell (6) operating parameters of the system to be designed, and establish an electro-thermal-fluid coupled simulation model including an electrochemical heat generation sub-model, a fluid flow sub-model and a heat transfer sub-model. The heat generation rate output by the electrochemical heat generation sub-model is used as the heat source input of the heat transfer sub-model and participates in the coupled solution together with the fluid flow sub-model. S2. The coupled simulation model is validated based on the experimentally measured discharge temperature data; S3. Determine the set of design variables and their range of values. The set of design variables shall include at least the coolant inlet volumetric flow rate. The contact half angle between the liquid cooling plate (1) and the aluminum sleeve (4) Aspect ratio of liquid cooling tubes And the thickness of the aluminum sleeve (4) And generate a set of sample points within the range of values; S4. Apply the coupled simulation model to each sample point to obtain the corresponding performance indicators. The performance indicators should include at least the highest temperature. Maximum temperature difference Differential pressure between the inlet and outlet of the liquid cooling pipe and aluminum volume ; S5. Train the surrogate model based on the sample point set and the corresponding performance index to obtain the mapping relationship between design variables and performance indexes, and determine the target surrogate model for optimization based on the evaluation index. S6. Perform multi-objective optimization based on the objective surrogate model to obtain the Pareto solution set; S7. Based on the multi-objective decision rule, determine the final design scheme from the Pareto solution set, and substitute the final design scheme back into the coupled simulation model to obtain the performance index output corresponding to the final design scheme.

2. The multi-objective optimization design method for the lithium-ion battery liquid cooling thermal management system according to claim 1, characterized in that: In step S3, the sample point set is generated using Latin hypercube sampling.

3. The multi-objective optimization design method for the lithium-ion battery liquid cooling thermal management system according to claim 1, characterized in that, Step S2 includes: collecting experimentally measured discharge temperature data under at least two discharge rate conditions, and comparing the temperature curve output by the coupled simulation model with the experimentally measured discharge temperature data to complete the verification of the coupled simulation model.

4. The multi-objective optimization design method for the lithium-ion battery liquid cooling thermal management system according to claim 1, characterized in that: In step S5, the evaluation index at least includes a determination coefficient R 2 and / or an error type index, and the prediction accuracy of different agent models is compared based on the evaluation index to determine the target agent model.

5. The multi-objective optimization design method for the lithium-ion battery liquid cooling thermal management system according to claim 1, characterized in that, In step S5, the surrogate model includes at least one of the following: Gaussian process regression model (GPR), response surface model (RSM), radial basis function model (RBF), backpropagation neural network model (BPNN), or support vector regression model (SVR).

6. The multi-objective optimization design method for the lithium-ion battery liquid cooling thermal management system according to claim 1, characterized in that: In step S6, the multi-objective optimization employs an evolutionary algorithm to obtain the Pareto solution set. The evolutionary algorithm includes at least the non-dominated sorting genetic algorithm NSGA-II and / or NSGA-III, and / or the multi-objective particle swarm optimization algorithm MOPSO, and / or the decomposition-type multi-objective evolutionary algorithm MOEA / D.

7. The multi-objective optimization design method for the lithium-ion battery liquid cooling thermal management system according to claim 1, characterized in that, In step S7, the multi-objective decision-making rule includes: normalizing the objective indices in the Pareto solution set, and calculating the distance of each Pareto solution relative to the ideal solution based on the weights to form a ranking result, thereby selecting the final design scheme.

8. A lithium-ion battery liquid cooling and thermal management system designed using the method described in any one of claims 1-7, characterized in that, include: Battery module, liquid cooling plate (1), first liquid cooling pipe (2), second liquid cooling pipe (3), aluminum sleeve (4) and aluminum block heat sink (5); The battery module includes multiple cylindrical cells (6), and the aluminum sleeve (4) is respectively sleeved on the outer periphery of each cylindrical cell (6); The aluminum block heat sink (5) is disposed between adjacent aluminum sleeves (4) and is thermally connected to the adjacent aluminum sleeves (4); Multiple sets of the second liquid cooling pipes (3) are arranged along the height direction of the cylindrical cell (6), and are all arranged close to the outer circumference of the aluminum sleeve (4), forming a connecting transition section between adjacent aluminum sleeves (4), so that the second liquid cooling pipes (3) can form a continuous loop around the outside of the battery module, and the second liquid cooling pipes (3) are provided with an outer flow inlet (31) and an outer flow outlet (32). The liquid cooling plate (1) is disposed at the bottom of the battery module, the first liquid cooling pipe (2) is independently arranged in the liquid cooling plate (1) and forms a circuit, and the first liquid cooling pipe (2) is provided with a bottom flow inlet (21) and a bottom flow outlet (22).

9. The lithium-ion battery liquid cooling and thermal management system according to claim 8, characterized in that: The second liquid cooling pipe (3) runs around the outer periphery of each aluminum sleeve (4) along the outside of the battery module, and the adjacent running sections are connected by a bend transition section located between adjacent aluminum sleeves (4), thereby forming a wave-shaped loop path. Furthermore, the outer flow inlet (31) and the outer flow outlet (32) are located on the same side of the battery module.

10. The lithium-ion battery liquid cooling and thermal management system according to claim 8, characterized in that: The first liquid cooling pipe (2) is arranged in a U-shaped loop inside the liquid cooling plate (1). The bottom flow inlet (21) and the bottom flow outlet (22) are located on the same side of the liquid cooling plate (1), and the bend of the U-shaped loop is located on the side away from the bottom flow inlet (21) and the bottom flow outlet (22).

Images