Optimization method for structural geometric parameter design of air-cooled battery thermal management system

Abstract

The invention discloses an optimization method for structural geometric parameter design of an air-cooled battery thermal management system. The method comprises the following steps: generating initial flow and a temperature field by using a control equation; converting the control equation into a residual equation; performing forward operation to calculate flow distribution, a temperature field and a target function; reverse calculation is carried out, gradient information and a Hessian matrix are obtained, the optimal variable adjustment direction is determined based on the gradient information and the Hessian matrix, and the variable combination is iteratively updated in the direction; and judging convergence and verifying and calculating by a computational fluid dynamics (CFD) method. According to the method, mathematical expressions for evaluating, matching and designing three common problems are unified, solving can be carried out in the completely same mode, expandability is good, and the method can be suitable for various optimization system structure designs and a wider application range of system performance after evaluation and optimization; and the universality is good, and various types of objective functions of linear function relations can be processed.

Description

Technical Field

[0001] The present invention relates to the technical field of battery thermal management system optimization, and in particular to an optimization method for structural geometric parameter design of an air-cooled battery thermal management system. Background Art

[0002] As the global energy crisis intensifies, electric vehicles, with their low pollution and low energy consumption, are an effective way to achieve energy conservation and emission reduction. Lithium-ion batteries, core components of electric vehicles, have a temperature profile that directly determines the performance and safety of the entire vehicle. To effectively manage battery temperatures and temperature differentials, developing an efficient battery thermal management system is essential. Air cooling is an effective battery thermal management technology, and the system's structural design can effectively reduce hotspot temperatures (the highest temperature within a range) and minimize temperature differentials within the battery pack. Therefore, designing a thermal management system structure with optimal heat dissipation performance is crucial.

[0003] In recent years, researchers have extensively studied the structural design of air cooling systems to improve the performance of air-cooled battery thermal management systems. By adjusting the layout of the battery pack and the design of the airflow channels, the cooling efficiency of the system and the temperature uniformity of the battery can be significantly improved. In order to more effectively improve the heat dissipation performance of the air-cooling system, researchers have used a variety of optimization methods, such as agent models, genetic algorithms, and particle swarm optimization algorithms, to optimize key parameters such as the battery spacing and guide plate angle of the air-cooling system. These measures effectively reduced the hot spot temperature and the internal temperature difference of the battery pack, and improved the heat dissipation performance. However, previous studies mainly used combinatorial optimization algorithms and heuristic methods to design the structure of parallel air-cooled battery thermal management systems. These methods require evaluating the performance of a large number of systems with different parameters generated during the optimization process. In addition, when optimizing different variables and objectives, it is necessary to convert the evaluation functions that match these variables and objectives, which makes the optimization process complex to implement and has low computational efficiency. Summary of the Invention

[0004] In order to overcome the deficiencies of the above-mentioned prior art, the purpose of the present invention is to provide an optimization method for the structural geometric parameter design of an air-cooled battery thermal management system: the method of the present invention is based on automatic differentiation technology and develops a new optimization framework, which unifies the mathematical expressions of evaluation, matching and design problems, and can be solved in exactly the same way, and can simultaneously realize the functions of performance evaluation, parameter matching and structural optimization, with good versatility and high efficiency.

[0005] The technical solution of the present invention is specifically described as follows.

[0006] The present invention provides an optimization method for designing structural geometric parameters of an air-cooled battery thermal management system, comprising the following steps:

[0007] S1. Calculate the initial flow distribution in the air channel of the battery thermal management system and the initial temperature distribution of the battery cell using the flow conservation equation, the pressure drop consistency equation, and the energy conservation equation as control equations;

[0008] S2, transform the control equation into the residual equation;

[0009] S3. Construct a differentiable function F to unify the expression of evaluation, matching, and design problems. The evaluation problem is to calculate the flow rate in the flow channel based on given geometric parameters. The matching problem is to find the optimal geometric parameters to match the flow distribution with the specified pattern. The design problem is to optimize the geometric parameters with the goal of improving temperature uniformity or reducing power consumption.

[0010] For the evaluation and matching problems, the function F is a combination of the flow residual and the pressure drop residual;

[0011] For the design problem, the function F is a combination of the flow residual, the pressure drop residual, the energy residual, and the objective function, which includes the battery cell temperature deviation objective function and the driving airflow power consumption objective function;

[0012] S4. Use the hyperbolic tangent function as a transformation technique to convert the constrained minimization problem into an unconstrained problem; S5. Search for the minimum value of the function F through the credible region Newton conjugate gradient algorithm to find the combination of geometric parameter variables that minimizes the function F.

[0013] In the present invention, in step S1, the battery thermal management system is a parallel air-cooled battery thermal management system; the flow resistance network model and the simplified heat transfer model are used to describe the relationship between the flow distribution, battery temperature and geometric parameter variables of the parallel air-cooled battery thermal management system; the geometric parameter variables include divergent channel width, parallel channel width and convergent channel width.

[0014] In the present invention, in step S2, the flow distribution of the air channel is obtained by the flow conservation residual and the pressure drop consistency residual, and the battery temperature distribution is obtained by the battery energy conservation residual and the parallel channel energy conservation residual.

[0015] In the present invention, in step S3, Formula 1, Formula 2, and Formula 3 are expressions of the function F to be minimized in the evaluation, matching, and design problems, respectively:

[0016]

[0017] Among them, R flow,i 、R pressure,i 、R heat,b,i and R heat,pc,iThey represent the flow conservation residual, pressure drop consistency residual, battery cell energy conservation residual and parallel channel (PC) energy conservation residual, respectively; i represents the sequence number of each battery cell; f ref 、p ref and Q ref is a reference value used to scale the residuals to a similar order of magnitude; T and O P are the objective functions of battery cell temperature deviation and driving airflow power consumption, β T and β P is the weighting coefficient, f in and T0 are the inlet gas flow rate and temperature, respectively.

[0018] In the present invention, in step S5, the method for searching for the minimum value of the function F by the credible region Newton conjugate gradient algorithm is specifically as follows:

[0019] S51. Run the fluid motion and numerical heat conduction iterations several times in the forward direction, calculate the flow rate, temperature field and objective function according to the residual equation, and record the calculation results;

[0020] S52, run in reverse, use the chain rule to calculate the first-order partial derivative of F with respect to the temperature field and flow rate, and obtain the gradient of F with respect to the geometric parameter variable;

[0021] S53, run in reverse again, calculate the second-order partial derivatives of F with respect to the temperature field and flow rate through the same back-propagation mechanism, and construct the Hessian matrix;

[0022] S54, based on gradient and Hessian matrix analysis, determine the optimal variable adjustment direction, provide a search direction for the trusted region Newton conjugate gradient algorithm, and iteratively update the variable combination along this direction;

[0023] S55. When the F value converges to the expected threshold, the process is terminated; otherwise, the process returns to step S51 and recalculates.

[0024] In the present invention, in step S51 and step S53, the flow distribution of the air channel is obtained by the flow conservation residual and the pressure drop consistency residual, and the battery temperature distribution is obtained by the battery energy conservation residual and the parallel channel energy conservation residual.

[0025] The present invention further includes step S6: updating the optimized structural parameters, using computational fluid dynamics (CFD) methods to calculate the flow and temperature field distribution characteristics of the optimized system, and verifying the effectiveness of the optimization method.

[0026] Compared with the prior art, the present invention has the following beneficial effects:

[0027] 1. Accurate gradient: The present invention uses a simplified numerical model to simulate fluid motion and heat transfer processes, iterates these operations in a cyclic manner, and uses an automatic differentiation algorithm to perform accurate gradient calculation and optimization without manually specifying the optimization criteria.

[0028] 2. Good scalability: The present invention can use the same processing method to achieve system performance evaluation, parameter matching, and structural optimization. For air-cooled battery thermal management systems with different flow channel structures (such as Z-type, J-type, and U-type parallel air-cooling structures, serial air-cooling structures, or longitudinal air-cooling structures), after calculating the system flow and temperature field distribution according to their corresponding control equations, the same method of converting them into the F minimization problem can be used to optimize different structures and evaluate system performance in a wider range of applications.

[0029] 3. Good versatility: The present invention can process objective functions of various types of linear function relationships. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a flow chart of the method of the present invention.

[0031] Figure 2 It is a structural diagram of the structural geometric parameter design of the parallel air-cooled battery thermal management system according to an embodiment of the present invention.

[0032] Figure 3 This is a flow and temperature field diagram after the structural geometric parameter design is optimized for the parallel air-cooled battery thermal management system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0033] The present invention will be further described below with reference to the accompanying drawings and examples, but the embodiments of the present invention are not limited to the examples.

[0034] An optimization method for the design of structural geometric parameters of air-cooled battery thermal management systems, such as Figure 1 As shown, it includes using the control equation to generate the initial flow and temperature field; converting the control equation into a residual equation; forward running to calculate the flow distribution, temperature field and objective function; reverse calculation to obtain gradient information and Hessian matrix, determine the optimal variable adjustment direction based on the gradient information and Hessian matrix, and iteratively update the variable combination along this direction; convergence judgment and computational fluid dynamics method to verify the calculation steps.

[0035] This paper unifies the performance evaluation, parameter matching, and structural optimization of an air-cooled battery thermal management system into a single minimization problem: finding a set of geometric parameters that minimizes a function F, selected from a combination of flow residual and pressure drop residual, or a combination of flow residual, pressure drop residual, energy residual, and an objective function (battery cell temperature deviation and driving airflow power consumption). A flow resistance network model and a simplified heat transfer model are used to simulate the system's flow and temperature distributions. Automatic differentiation techniques and the trusted region Newton conjugate gradient algorithm are combined to calculate the geometric parameters that minimize F. Computational fluid dynamics (CFD) methods are then used to verify the effectiveness of the performance evaluation, parameter matching, and structural optimization.

[0036] like Figure 2 As shown, in this embodiment, the Figure 2 The air-cooled battery thermal management system shown in the figure features a Z-shaped parallel flow channel structure. The solid area of ​​the system consists of a 12×2 prismatic battery pack array. Heat generated by the battery pack is dissipated through forced air convection. During system operation, cooling air enters the diverging channel (DD) through the Z-shaped lower inlet and is then distributed to the parallel channels (PC) to cool the batteries. Finally, the hot air converges in the converging channel (CD) and flows out through the outlet, completing the heat exchange process.

[0037] The geometric parameter variables include but are not limited to the diverging channel width, the parallel channel width and the converging channel width. In this embodiment, when optimizing the structure of the system, the geometric parameter variable to be optimized is the width of the parallel channel, which is denoted as w PC , the flow rate of each parallel channel is recorded as f PC The temperature field uses the average temperature of a single battery cell, denoted as T b In the evaluation problem, the geometric parameters of the system are known, and the goal is to calculate the flow rate of each airflow channel; in the matching problem, each f PC The value of is predetermined, and the goal is to find w that satisfies its distribution PC In the design problem, the goal is to search for the optimal geometric parameters that minimize the battery cell temperature deviation or the power consumption of driving the airflow while satisfying the residual equation. Therefore, the function F for the performance evaluation and parameter matching problem is a combination of the flow residual and the pressure drop residual, while the function F selected for the structural optimization problem is a combination of the flow residual, the pressure drop residual, the energy residual, and the objective function (battery cell temperature deviation and the power consumption of driving the airflow).

[0038] The inlet air flow rate of the system is 0.015m 3 / s, the initial temperature is 298.15K, and the heat source of the battery pack is 137562W / m 3 The width of the converging and diverging channels is 20 mm; there are 13 parallel channels, each with a width of PCThe value of is 3mm. To keep the system volume constant, w PC The sum is fixed at 39mm. The specific calculation steps are as follows:

[0039] S1. Using a flow resistance network model and a simplified heat transfer model, the flow conservation equation, the pressure drop consistency equation, and the energy conservation equation control the flow distribution in the air channel and the temperature of the battery cell. The governing equations of the model are as follows:

[0040] f DD,i -f PC,i+1 -f DD,i+1 =0 (4)

[0041] f CD,i +f PC,i+1 -f CD,i+1 =0 (5)

[0042]

[0043] h CD,i ·S CD ΔT up,i +h DD,i ·S DD ΔT down,i +h PC,i ·S PC ΔT left,i +h PC,i+1 ·S PC ΔT right,i -φ·V=0 (7)

[0044] ρ·C P ·U PC,i ·(T CD,i -T0)·A PC,i -h PC,i ·S PC ΔT right,i-1 -h PC,i ·S PC ΔT left,i =0(8)

[0045] Equations (4) and (5) are flow conservation equations, Equation (6) is the pressure drop consistency equation, and Equations (7) and (8) are energy conservation equations. Where f, P, h, S, and ΔT represent flow, pressure, convection heat transfer coefficient between battery and air, area, and temperature difference, respectively; Φ and V represent the heat generation power per unit volume of the battery and the volume of a single battery cell; ρ and C p , U and T0 represent the density, specific heat capacity, velocity and inlet gas temperature of air respectively.

[0046] S2. Convert the flow conservation equation, pressure drop consistency equation and energy conservation equation into residual form;

[0047] S3. Construct a function F to unify the expression of evaluation, matching, and design problems. Equations 1, 2, and 3 are the expressions of the function F to be minimized in the evaluation, matching, and design problems, respectively:

[0048]

[0049] Among them, R flow,i 、R pressure,i 、R heat,b,i and R heat,pc,i They represent the flow conservation residual equation, pressure drop consistency residual equation, battery cell energy conservation residual equation and parallel channel (PC) energy conservation residual equation respectively; f ref 、p ref and Q ref is a reference value used to scale the residuals to a similar order of magnitude; T and O P are the objective functions of battery cell temperature deviation and driving airflow power consumption, β T and β P is the weighting coefficient, f in and T0 are the inlet gas flow rate and temperature, respectively.

[0050] S4. Using the hyperbolic tangent function as a transformation technique, the constrained minimization problem is converted into an unconstrained problem, which not only solves the variable constraints but also satisfies the constant volume of the parallel air-cooling system.

[0051] S5. Run the fluid motion and numerical heat conduction iterations in the forward direction for several times, calculate the flow rate, temperature field and objective function according to the conservation equations, and record the calculation results;

[0052] S6. Run in reverse and use the chain rule to calculate the first-order partial derivatives of F with respect to the temperature field and flow rate to obtain the gradient of F with respect to all independent variables;

[0053] S7, run in reverse again, calculate the second-order partial derivatives of F with respect to the temperature field and flow through the same back-propagation mechanism, and construct the Hessian matrix;

[0054] S8. Determine the optimal variable adjustment direction based on gradient and Hessian matrix analysis, provide a search direction for the trusted region Newton conjugate gradient algorithm, and iteratively update the variable combination along this direction;

[0055] S9. If the F value continues to decrease, return to step S8 for iterative optimization;

[0056] S10. When the F value converges to the expected threshold, the process is terminated; otherwise, the process returns to step S5 and recalculates.

[0057] S11. Update the optimized structural parameters and use the computational fluid dynamics (CFD) method to calculate the flow and temperature field distribution characteristics of the optimized system to verify the effectiveness of this optimization method.

[0058] In the evaluation and matching problem, F is the combination of flow residual and pressure drop residual. Based on the CFD calculation results, the average relative error of the flow distribution calculated by this method is 3.8%; based on the geometric parameters obtained by solving the inverse calculation model, the average absolute error is 0.032mm. In the design problem, F is the combination of flow residual, pressure drop residual, energy residual and objective function. The standard deviation of the optimized battery cell temperature is reduced from 16.52K to 0K. The optimized flow distribution and temperature field are as follows Figure 3 shown.

[0059] As described above, the flow resistance network model and the simplified heat transfer model are introduced in the embodiment of the present invention to describe the relationship between the flow distribution, battery temperature and geometric parameters of the Z-type parallel air-cooled system; F is taken as a function of the residual equation and the objective function, and the residual equation and the objective function are taken as functions of the geometric parameter variables, flow and temperature; in this way, the solution process of any combination of known and unknown quantities is unified into the problem of minimizing F; the automatic differentiation technology and the Newton conjugate gradient method are used to solve the unified minimization problem, and the problems of evaluation, matching and design of the parallel air-cooled battery thermal management system are uniformly solved.

[0060] The above embodiments are typical implementations of the present invention, but the implementation of the present invention is not limited to the embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the principles and essence of the present invention should be considered equivalent replacement methods and are included in the scope of protection of the present invention.

Claims

1. An optimization method for the design of structural geometric parameters of an air-cooled battery thermal management system, characterized in that: The following steps are involved: S1. Calculate the initial flow distribution in the air channel of the battery thermal management system and the initial temperature distribution of the battery cell using the flow conservation equation, the pressure drop consistency equation, and the energy conservation equation as control equations; S2, transform the control equation into the residual equation; S3. Construct a differentiable function F to unify the expression of evaluation, matching, and design problems. The evaluation problem is to calculate the flow rate in the flow channel based on given geometric parameters. The matching problem is to find the optimal geometric parameters to match the flow distribution with the specified pattern. The design problem is to optimize the geometric parameters with the goal of improving temperature uniformity or reducing power consumption. For the evaluation and matching problems, the function F is a combination of the flow residual and the pressure drop residual; For the design problem, the function F is a combination of the flow residual, the pressure drop residual, the energy residual, and the objective function, which includes the battery cell temperature deviation objective function and the driving airflow power consumption objective function; S4. Using the hyperbolic tangent function as a transformation technique, the constrained minimization problem is converted into an unconstrained problem. S5. Search for the minimum value of function F through the credible region Newton conjugate gradient algorithm to find the combination of geometric parameter variables that minimizes function F.

2. The optimization method for structural geometric parameter design of an air-cooled battery thermal management system according to claim 1, characterized in that: In step S1, the battery thermal management system is a parallel air-cooled battery thermal management system; the flow resistance network model and the simplified heat transfer model are used to describe the relationship between the flow distribution, battery temperature and geometric parameter variables of the parallel air-cooled battery thermal management system; the geometric parameter variables include divergent channel width, parallel channel width and convergent channel width.

3. The optimization method for structural geometric parameter design of an air-cooled battery thermal management system according to claim 1, characterized in that: In step S2, the flow distribution of the air channel is obtained by calculating the flow conservation residual and the pressure drop consistency residual, and the battery temperature distribution is obtained by calculating the battery energy conservation residual and the parallel channel energy conservation residual.

4. The optimization method for structural geometric parameter design of an air-cooled battery thermal management system according to claim 1, characterized in that: In step S3, Equations 1, 2, and 3 are expressions of the function F to be minimized in the evaluation, matching, and design problems, respectively: Among them, R flow,i 、R pressure,i 、R heat,b,i and R heat,pc,i They represent the flow conservation residual, pressure drop consistency residual, battery cell energy conservation residual and parallel channel (PC) energy conservation residual, respectively; i represents the sequence number of each battery cell; f ref 、p ref and Q ref is a reference value used to scale the residuals to a similar order of magnitude; T and O P are the objective functions of battery cell temperature deviation and driving airflow power consumption, β T and β P is the weighting coefficient, f in and T0 are the inlet gas flow rate and temperature, respectively.

5. The optimization method for structural geometric parameter design of an air-cooled battery thermal management system according to claim 1, characterized in that: In step S5, the method for searching for the minimum value of the function F by the credible region Newton conjugate gradient algorithm is as follows: S51. Run the fluid motion and numerical heat conduction iterations several times in the forward direction, calculate the flow rate, temperature field and objective function according to the residual equation, and record the calculation results; S52, run in reverse, use the chain rule to calculate the first-order partial derivative of F with respect to the temperature field and flow rate, and obtain the gradient of F with respect to the geometric parameter variable; S53, run in reverse again, calculate the second-order partial derivatives of F with respect to the temperature field and flow rate through the same back-propagation mechanism, and construct the Hessian matrix; S54, based on gradient and Hessian matrix analysis, determine the optimal variable adjustment direction, provide a search direction for the trusted region Newton conjugate gradient algorithm, and iteratively update the variable combination along this direction; S55. When the F value converges to the expected threshold, the process is terminated; otherwise, the process returns to step S51 and recalculates.

6. The optimization method for structural geometric parameter design of an air-cooled battery thermal management system according to claim 5, characterized in that: In step S51 and step S53, the flow distribution of the air channel is obtained by the flow conservation residual and the pressure drop consistency residual, and the battery temperature distribution is obtained by the battery energy conservation residual and the parallel channel energy conservation residual.

7. The optimization method for structural geometric parameter design of an air-cooled battery thermal management system according to claim 1, characterized in that: The method also includes step S6: updating the optimized structural parameters, and using the computational fluid dynamics (CFD) method to calculate the flow and temperature field distribution characteristics of the optimized air-cooled battery thermal management system to verify the effectiveness of this optimization method.

Images