Charging module heat dissipation optimization method based on temperature field simulation

Abstract

The invention relates to the technical field of simulation optimization, and discloses a charging module heat dissipation optimization method based on temperature field simulation, and the method comprises the steps: firstly discretizing a heat dissipation space into voxel units, and collecting flow heat data; calculating the flow resistance sensitivity according to the relationship between the flow velocity vector and the main wind direction; calculating an effective convection potential energy density and a gradient vector thereof by combining a Reynolds number and a temperature difference; then, generating a chasing cold source driving vector based on the heat flux demand of the component; establishing a thermal wake flow cone projection model, and calculating a thermal wake flow rejection vector; generating a flow resistance avoidance vector in combination with the regional flow velocity direction; and finally, performing weighted synthesis on the three vectors to obtain optimized displacement, and updating coordinates of the component. According to the technical scheme, through multi-physics field coupling analysis, while a driving component finds a high cooling potential area, flow channel smoothness and thermal wake avoidance are effectively considered, and deterministic optimization of the charging module layout is realized.

Description

Technical Field

[0001] This invention relates to the field of simulation optimization technology, specifically to a method for optimizing the heat dissipation of a charging module based on temperature field simulation. Background Technology

[0002] As the core unit of energy conversion, the power density of the charging module has increased explosively, which has led to a sharp increase in the heat flux density inside the module. Heat dissipation design has become a key bottleneck restricting product performance and reliability. Existing heat dissipation layout design of charging modules usually relies on computational fluid dynamics (CFD) simulation technology. Through trial and error based on the experience of designers or random search based on general algorithms (such as genetic algorithms and particle swarm algorithms), a component arrangement scheme that meets the temperature index is found. However, existing optimization techniques have many limitations when facing the complex forced air cooling environment inside the charging module, failing to deeply reveal the physical coupling mechanism between the flow field and the temperature field. First, existing technologies often unilaterally pursue placing high-heat-generating devices in the region with the highest flow velocity, while ignoring the key physical quantity of "flow resistance sensitivity." This results in large-volume devices blocking key nodes of the core heat dissipation airflow, causing a surge in system air pressure and a decrease in total airflow, which in turn worsens the overall heat dissipation environment. Second, traditional temperature field analysis fails to accurately quantify the actual heat transfer potential of fluid micro-particles at different spatial locations. The difference in cooling capacity between low-temperature and high-temperature fluids cannot be distinguished solely by flow velocity, resulting in resource waste in high-cooling potential areas and a "mismatch" of devices with low heat dissipation requirements. Furthermore, in high-density layout scenarios, upstream heat-generating devices will generate a significant thermal wake effect on downstream devices. Existing simple constraints based on geometric distance cannot accurately describe this conical diffusion "thermal shielding" phenomenon, causing downstream devices to be enveloped by upstream hot airflow for a long time, forming a thermal cascade fault that is difficult to eliminate. In summary, there is an urgent need for a technical solution that can deeply analyze the microscopic thermal characteristics and achieve precise optimization of the layout of charging module components under multi-physics field constraints, in order to solve the technical problems that are common in current designs, such as low cooling efficiency, easy blockage of flow channels, and serious thermal interference. Summary of the Invention

[0003] This invention provides a method for optimizing the heat dissipation of a charging module based on temperature field simulation, which helps to solve the problems mentioned in the background art.

[0004] This invention provides the following technical solution: a method for optimizing the heat dissipation of a charging module based on temperature field simulation, comprising: The heat dissipation space of the charging module is discretized into voxel units, and the flow velocity and fluid temperature data of each unit are collected through simulation. Based on the angle between the unit velocity vector and the prevailing wind direction and the velocity modulus, the flow resistance sensitivity value of each unit is calculated. By combining local Reynolds number and fluid temperature difference data, the effective convective potential energy density characterizing the heat transfer potential is calculated. Perform three-dimensional spatial difference operations on the potential energy density field to obtain the potential energy gradient vector at each unit location; Based on the heat flux requirements of the components, the gradient vector is volume-integrated to generate a driving vector that chases the cold source. Establish a conical projection model of the thermal wake and calculate the thermal wake repulsion vector experienced by downstream components; By combining the average flow velocity direction in the component area with the cumulative flow resistance sensitivity, a flow resistance avoidance vector is generated; The optimized displacement is obtained by performing a weighted synthesis on the three vectors, and the component coordinates in the simulation model are updated.

[0005] Optionally, the step of discretizing the heat dissipation space of the charging module into voxel units and collecting flow velocity and fluid temperature data of each unit through simulation includes: A discretized model of the internal heat dissipation space of the charging module is obtained and divided into multiple regular hexahedral voxel units. The side length of the voxel unit is set according to the smallest critical feature size in the charging module. Perform a single steady-state computational fluid dynamics simulation to obtain the baseline state data for each voxel element; The baseline state data includes the air velocity vector at the center of the voxel, the local fluid temperature at the voxel, and the air fluid density.

[0006] Optionally, the calculation of the flow resistance sensitivity value of each unit based on the angle between the unit velocity vector and the prevailing wind direction and the velocity modulus includes: Based on the acquired flow velocity data, a flow resistance sensitivity scalar is calculated for each voxel; The calculation of the flow resistance sensitivity scalar is proportional to the square of the magnitude of the air fluid density and the velocity vector; The calculation process introduces a direction correction term, which is determined based on the dot product of the unit direction vector of the flow velocity and the preset axial unit vector of the main heat dissipation air duct, and uses a preset turbulence correction factor to adjust the repulsion weight for non-axial flow.

[0007] Optionally, the calculation of the effective convective potential energy density characterizing the heat transfer potential, by combining local Reynolds number and fluid temperature difference data, includes: Based on the air fluid density, the air velocity vector at the voxel center, the preset characteristic length constant, and the fluid dynamic viscosity, the local Reynolds number of each voxel is calculated. Calculate the effective convective potential energy density of each voxel. The value of this density is positively correlated with the power function of the local Reynolds number and the power function of the air Prandtl number. The calculation of the effective convective potential energy density also includes a temperature difference driving term, which is based on the difference between the upper limit temperature that the component can withstand and the local fluid temperature, and is normalized by comparing the difference between the upper limit temperature that the component can withstand and the inlet temperature of the cooling medium.

[0008] Optionally, performing a three-dimensional spatial difference operation on the potential energy density field to obtain the potential energy gradient vector at each unit location includes: The effective convective potential energy density field is processed using the three-dimensional discrete central difference method. Calculate the potential energy density difference between adjacent voxels in the three coordinate axes for the current voxel; Divide the difference by the voxel side length in the corresponding coordinate axis direction to obtain the voxel's components in the three directions, which are then combined to form a gradient vector pointing in the direction of the fastest increase in cooling capacity.

[0009] Optionally, the step of volume-integrating the gradient vector based on the heat flux requirements of the components to generate a driving vector for chasing the cold source includes: The heat flux demand intensity of the component is calculated, which is determined by the heat loss power density of the component, and nonlinear gain correction is performed based on the ratio of the deviation between the highest surface temperature of the component and the design target temperature. Determine the set of voxels occupied by the components in discrete space; The average gradient vector is obtained by summing and averaging the gradient vectors of all voxels in the set. The heat flux demand intensity is multiplied by the average gradient vector to generate the chasing cold source driving vector of the component.

[0010] Optionally, establishing the conical projection model of the thermal wake and calculating the thermal wake repulsion vector experienced by downstream components includes: Define a wake cone region extending from the upstream components along the flow velocity direction, which is defined by a preset half-cone angle; Determine whether the center of the downstream component is located within the wake cone region of the upstream component; If it is located within the region, the magnitude of the repulsion vector is calculated. This magnitude is directly proportional to the power product of the upstream and downstream components and inversely proportional to the square of the distance between the geometric centers of the two components. The calculation of the repulsion vector also includes a radial attenuation term, which is exponentially attenuated based on the vertical distance from the center of the downstream component to the wake axis of the upstream component. The direction of the repulsion vector is set to be perpendicular to the local streamline and pointing away from the wake centerline.

[0011] Optionally, the step of generating a flow resistance avoidance vector by combining the average flow velocity direction of the component region with the cumulative flow resistance sensitivity includes: The volume average velocity vector is obtained by calculating the arithmetic mean of the velocity vectors of all voxel units occupied by the statistical components. The cumulative flow resistance sensitivity is obtained by summing the flow resistance sensitivity values ​​of all voxel units occupied by the component. A flow resistance avoidance vector is generated, the direction of which is opposite to the direction of the volume average flow velocity vector, and its magnitude is proportional to the cumulative flow resistance sensitivity.

[0012] Optionally, the step of performing a weighted synthesis of the three vectors to obtain the optimized displacement and updating the component coordinates in the simulation model includes: Set the weighting coefficients for heat dissipation optimization, flow resistance avoidance, and thermal wake avoidance; Set a relaxation step size, which is determined based on the minimum side length of the heat dissipation space of the charging module and a preset step size coefficient; The cold source chasing drive vector, the flow resistance avoidance vector, and the thermal wake repulsion vector applied by all upstream components are multiplied by their respective weighting coefficients. The weighted vectors are combined, and the combined result is multiplied by the relaxation step size to obtain the total optimized displacement vector. The total optimized displacement vector is superimposed onto the geometric center coordinates of the component before the update to obtain the updated three-dimensional coordinates.

[0013] The present invention has the following beneficial effects: 1. For the specific environment of charging modules, characterized by high power density and limited heat dissipation space, this solution proposes a layout optimization method based on the flow-thermal coupling mechanism. This method discretizes the internal space and calculates three virtual driving forces guiding the movement of components: a "cold source chasing force" based on heat transfer potential, a "flow resistance avoidance force" based on flow resistance sensitivity, and a "wake-avoidance force" based on conical projection. The reason for constructing these three forces is that in complex forced air cooling environments, simply seeking low-temperature zones can easily lead to components blocking the core airflow channel, or due to series... The current arrangement causes downstream devices to draw in hot exhaust gas from upstream. This solution uses a weighted synthesis of these three vectors to drive high-heat-generating devices to actively seek areas with high heat dissipation efficiency, while forcing them to avoid high-speed turbulent flow areas that can cause a surge in wind pressure. It also forces them to be laterally offset to avoid the thermal wake of upstream devices. This solution effectively resolves the contradiction between "local heat dissipation needs" and "unobstructed global flow channels" in a confined space. It prevents the core airflow from being reduced due to flow channel blockage and eliminates the cumulative effect of thermal cascading between devices, thereby achieving a dual optimization of heat dissipation performance and aerodynamic characteristics. 2. By combining the angle between the unit velocity vector and the main wind direction and the velocity modulus, the flow resistance sensitivity value is calculated; the dot product operation of the unit velocity direction vector and the axial unit vector of the main heat dissipation air duct is introduced; it can accurately identify non-axial flow regions with high kinetic energy and chaotic flow direction in the flow field; this enables the algorithm to quantify the degree of hydrodynamic repulsion of each point in space against the intrusion of the entity, so as to intelligently mark the high flow resistance risk area where it is not suitable to place flow obstruction devices in subsequent optimization, effectively preventing the core air duct blockage and the surge in system air pressure caused by improper component layout, and ensuring the stability of the overall heat dissipation air volume; 3. The effective convective potential energy density is calculated by comprehensively considering local Reynolds number, fluid temperature difference data, and air fluid properties. It abandons the traditional perspective of judging heat dissipation capacity solely based on flow velocity or temperature. It constructs a comprehensive physical quantity that characterizes the heat transfer potential of space. This index not only reflects the flow intensity of fluid micro-particles but also combines the fluid's own heat capacity margin. It can accurately distinguish between pseudo-high-efficiency zones where the flow velocity is high but the temperature is already high, resulting in saturated cooling capacity, and true high-efficiency zones where the flow velocity is moderate but the temperature is extremely low and has great heat absorption potential. This provides a navigation map that conforms to thermodynamic principles for finding the best heat dissipation location for components. 4. The potential energy gradient vector is obtained by performing a three-dimensional spatial difference operation on the potential energy density field; the potential energy density difference between adjacent voxels is processed using the three-dimensional discrete central difference method; the scalar form of the heat dissipation potential field is transformed into a vector field with a clear directionality; this allows components to move without blindly searching randomly, but to make deterministic displacements along the direction of the mathematical gradient with the fastest growth of cooling capacity; this greatly improves the directness and effectiveness of the optimization path, ensuring that each step of movement is physically directed to a position with a better heat dissipation environment, and improving the convergence speed of the algorithm; 5. The heat flux demand intensity is calculated based on the deviation between the component's heat loss power and surface temperature and the design target; and a chasing cold source driving vector is generated by combining the volume average gradient; a nonlinear gain correction mechanism is introduced, which enables high-heat-generating devices on the edge of overheating to obtain greater movement weight; this mechanism simulates the physical thermophoresis effect, ensuring that key devices with high heat flux density can preferentially seize high cooling potential energy resources; at the same time, volume integration avoids local deviations caused by single-point sampling, ensuring that the physical device as a whole smoothly migrates to the optimal heat dissipation area, and realizing the on-demand allocation of heat dissipation resources; 6. By establishing a conical projection model of the thermal wake and calculating the thermal wake repulsion vector, and using the half-cone angle to define the diffusion angle of the thermal wake propagating downstream, this model can accurately simulate the conical thermal shielding area formed by upstream heating devices in the flow field. This model overcomes the shortcomings of traditional models based on simple Euclidean distance constraints that cannot reflect flow direction thermal interference. It forces downstream devices under the wake of upstream devices to produce lateral avoidance movements. It effectively breaks the situation of high-heat-generating devices arranged in series, fundamentally eliminates the thermal cascade accumulation effect, and significantly improves the inlet air temperature environment of downstream devices. 7. A flow resistance avoidance vector is generated by calculating the volume average flow velocity direction and cumulative flow resistance sensitivity of the component region; the component is regarded as an entity with overall resistance characteristics rather than a point mass; so that the generated repulsive force can accurately reflect the overall degree of obstruction caused by the entity to the current flow field; this vector drives the component to actively withdraw from those flow resistance sensitive areas with extremely high flow velocities and chaotic directions, even if the heat dissipation potential energy of the area is high; this mechanism plays a negative feedback protection role in physics, preventing short-sighted behavior of sacrificing the smoothness of the global flow field for local heat dissipation, and maintaining the health of the system flow field; 8. By performing weighted synthesis on three vectors—cold source chasing, flow resistance avoidance, and thermal wake repulsion—a comprehensive driving mechanism with multi-physics coupling was constructed. Different weighting coefficients were used to flexibly adjust the balance between heat dissipation requirements, aerodynamic characteristics, and topological constraints, enabling the optimization process to find the best compromise solution among conflicting physical objectives. With the setting of relaxation step size, overshoot or oscillation of components during the optimization process was prevented. Ultimately, a deterministic layout scheme that satisfies the independent heat dissipation requirements of each component and conforms to the overall system's optimal hydrodynamic principle was achieved. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the basic process of the present invention. Detailed Implementation

[0015] 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. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] Example 1, refer to Figure 1 A method for optimizing the heat dissipation of a charging module based on temperature field simulation includes: The heat dissipation space of the charging module is discretized into voxel units, and the flow velocity and fluid temperature data of each unit are collected through simulation. Based on the angle between the unit velocity vector and the prevailing wind direction and the velocity modulus, the flow resistance sensitivity value of each unit is calculated. By combining local Reynolds number and fluid temperature difference data, the effective convective potential energy density characterizing the heat transfer potential is calculated. Perform three-dimensional spatial difference operations on the potential energy density field to obtain the potential energy gradient vector at each unit location; Based on the heat flux requirements of the components, the gradient vector is volume-integrated to generate a driving vector that chases the cold source. Establish a conical projection model of the thermal wake and calculate the thermal wake repulsion vector experienced by downstream components; By combining the average flow velocity direction in the component area with the cumulative flow resistance sensitivity, a flow resistance avoidance vector is generated; The optimized displacement is obtained by performing a weighted synthesis on the three vectors, and the component coordinates in the simulation model are updated.

[0017] The process of discretizing the heat dissipation space of the charging module into voxel units and collecting flow velocity and fluid temperature data of each unit through simulation includes: A discretized model of the internal heat dissipation space of the charging module is obtained and divided into multiple regular hexahedral voxel units. The side length of the voxel unit is set according to the smallest critical feature size in the charging module. Perform a single steady-state computational fluid dynamics simulation to obtain the baseline state data for each voxel element; The baseline state data includes the air velocity vector at the center of the voxel, the local fluid temperature at the voxel, and the air fluid density.

[0018] The calculation of the flow resistance sensitivity value for each unit is based on the angle between the unit's velocity vector and the prevailing wind direction, and the velocity modulus, including: Based on the acquired flow velocity data, a flow resistance sensitivity scalar is calculated for each voxel; The calculation of the flow resistance sensitivity scalar is proportional to the square of the magnitude of the air fluid density and the velocity vector; The calculation process introduces a direction correction term, which is determined based on the dot product of the unit direction vector of the flow velocity and the preset axial unit vector of the main heat dissipation air duct, and uses a preset turbulence correction factor to adjust the repulsion weight for non-axial flow.

[0019] The calculation of the effective convective potential energy density, characterizing the heat transfer potential, by combining local Reynolds number and fluid temperature difference data includes: Based on the air fluid density, the air velocity vector at the voxel center, the preset characteristic length constant, and the fluid dynamic viscosity, the local Reynolds number of each voxel is calculated. Calculate the effective convective potential energy density of each voxel. The value of this density is positively correlated with the power function of the local Reynolds number and the power function of the air Prandtl number. The calculation of the effective convective potential energy density also includes a temperature difference driving term, which is based on the difference between the upper limit temperature that the component can withstand and the local fluid temperature, and is normalized by comparing the difference between the upper limit temperature that the component can withstand and the inlet temperature of the cooling medium.

[0020] The step of performing a three-dimensional spatial difference operation on the potential energy density field to obtain the potential energy gradient vector at each unit location includes: The effective convective potential energy density field is processed using the three-dimensional discrete central difference method. Calculate the potential energy density difference between adjacent voxels in the three coordinate axes for the current voxel; Divide the difference by the voxel side length in the corresponding coordinate axis direction to obtain the voxel's components in the three directions, which are then combined to form a gradient vector pointing in the direction of the fastest increase in cooling capacity.

[0021] The step of integrating the gradient vector based on the heat flux requirements of the components to generate a driving vector for chasing the cold source includes: The heat flux demand intensity of the component is calculated, which is determined by the heat loss power density of the component, and nonlinear gain correction is performed based on the ratio of the deviation between the highest surface temperature of the component and the design target temperature. Determine the set of voxels occupied by the components in discrete space; The average gradient vector is obtained by summing and averaging the gradient vectors of all voxels in the set. The heat flux demand intensity is multiplied by the average gradient vector to generate the chasing cold source driving vector of the component.

[0022] The establishment of the conical projection model of the thermal wake and the calculation of the thermal wake repulsion vector experienced by downstream components include: Define a wake cone region extending from the upstream components along the flow velocity direction, which is defined by a preset half-cone angle; Determine whether the center of the downstream component is located within the wake cone region of the upstream component; If it is located within the region, the magnitude of the repulsion vector is calculated. This magnitude is directly proportional to the power product of the upstream and downstream components and inversely proportional to the square of the distance between the geometric centers of the two components. The calculation of the repulsion vector also includes a radial attenuation term, which is exponentially attenuated based on the vertical distance from the center of the downstream component to the wake axis of the upstream component. The direction of the repulsion vector is set to be perpendicular to the local streamline and pointing away from the wake centerline.

[0023] The generation of a flow resistance avoidance vector by combining the average flow velocity direction of the component area with the cumulative flow resistance sensitivity includes: The volume average velocity vector is obtained by calculating the arithmetic mean of the velocity vectors of all voxel units occupied by the statistical components. The cumulative flow resistance sensitivity is obtained by summing the flow resistance sensitivity values ​​of all voxel units occupied by the component. A flow resistance avoidance vector is generated, the direction of which is opposite to the direction of the volume average flow velocity vector, and its magnitude is proportional to the cumulative flow resistance sensitivity.

[0024] The process of performing a weighted synthesis of the three vectors to obtain the optimized displacement and updating the component coordinates in the simulation model includes: Set the weighting coefficients for heat dissipation optimization, flow resistance avoidance, and thermal wake avoidance; Set a relaxation step size, which is determined based on the minimum side length of the heat dissipation space of the charging module and a preset step size coefficient; The cold source chasing drive vector, the flow resistance avoidance vector, and the thermal wake repulsion vector applied by all upstream components are multiplied by their respective weighting coefficients. The weighted vectors are combined, and the combined result is multiplied by the relaxation step size to obtain the total optimized displacement vector. The total optimized displacement vector is superimposed onto the geometric center coordinates of the component before the update to obtain the updated three-dimensional coordinates. For the specific environment of charging modules, which have high power density and limited heat dissipation space, this solution proposes a layout optimization method based on the flow-thermal coupling mechanism. This solution discretizes the internal space and calculates three virtual driving forces to guide the movement of components: "chasing the cold source force" based on heat transfer potential, "avoiding flow resistance" based on flow resistance sensitivity, and "avoiding the wake force" based on conical projection. The reason for constructing these three forces is that in the complex forced air cooling environment, simply seeking low-temperature areas can easily lead to components blocking the core airflow channel, or downstream components drawing in hot exhaust gas from upstream due to tandem arrangement. This solution uses a weighted synthesis of these three vectors to drive high-heat-generating components to actively seek high heat dissipation efficiency areas while forcing them to avoid high-speed turbulent flow areas that can cause a surge in air pressure, and to force them to be laterally misaligned to avoid the thermal wake shielding of upstream components. It effectively solves the contradiction between "local heat dissipation needs" and "global flow channel unobstructed" in a confined space, preventing the core airflow from decreasing due to flow channel blockage and eliminating the cumulative effect of thermal cascading between components, thereby achieving dual optimization of heat dissipation performance and aerodynamic characteristics. Example 2: A method for optimizing the heat dissipation of a charging module based on temperature field simulation, further comprising: The process of discretizing the heat dissipation space of the charging module into voxel units and collecting flow velocity and fluid temperature data of each unit through simulation includes: Obtain a discretized model of the internal heat dissipation space of the charging module, denoted as the heat dissipation domain. Divide it into The volume is The regular hexahedral voxel unit; in which The volume of a single voxel unit. The side length is based on the smallest critical feature dimension in the charging module, such as the fin spacing of the heat sink or the pin spacing of the power device. If the value is too large, the mesh will be too coarse and the number of voxels will be too high. If the value is too small, it will lead to "numerical dissipation due to insufficient resolution," where subtle flow channel features (such as local turbulence) will be averaged, making it impossible for the optimization results to identify local hotspots; if the value is too small, the mesh will be too fine, resulting in an insufficient number of voxels. If the value is too large, it will cause the computational memory consumption to increase cubically, and the gradient calculation noise during the optimization process will increase (over-capturing tiny fluctuations), which may cause the components to oscillate repeatedly within a tiny range and fail to converge. The first steady-state computational fluid dynamics simulation was used to obtain the... Reference state vector of individual elements: ; in: For the first The air velocity vector at the center of the individual element, in units of ; For the first Local fluid temperature at individual units, in units of ; Air fluid density, unit .

[0025] The calculation of the flow resistance sensitivity value for each unit is based on the angle between the unit's velocity vector and the prevailing wind direction, and the velocity modulus, including: Based on the acquired flow velocity data, calculate the first... scalar of flow resistance sensitivity of individual elements This is used to quantify the sensitivity of the location to flow retardation. ; in: The unit direction vector of the flow velocity. ; This is the preset unit vector along the axial direction of the main heat dissipation air duct; Air fluid density, unit ; The preset turbulence correction factor has a value of [value missing]. This value is used to adjust the repulsion sensitivity to non-axial flow. A larger value makes the algorithm "extremely picky," with very strict criteria for judging high flow resistance regions. It only allows components to exist in regions with very straight streamlines, which may lead to excessive compression of available layout space and components being squeezed into low-speed regions at the edges. A smaller value makes the algorithm "ignore" turbulence and lateral flow, and the flow resistance sensitivity degenerates to be only related to kinetic energy. This may result in the optimized layout avoiding high-speed regions but blocking key turning air passages, increasing the overall pressure drop of the system. By combining the angle between the unit velocity vector and the main wind direction and the velocity modulus, the flow resistance sensitivity value is calculated; the dot product operation of the unit velocity direction vector and the axial unit vector of the main heat dissipation air duct is introduced; it can accurately identify non-axial flow regions with high kinetic energy and chaotic flow direction in the flow field; this enables the algorithm to quantify the degree of hydrodynamic repulsion of each point in space against the intrusion of the entity, so as to intelligently mark the high flow resistance risk area where it is not suitable to place flow obstruction devices in subsequent optimization, effectively preventing the core air duct blockage and the surge in system air pressure caused by improper component layout, and ensuring the stability of the overall heat dissipation air volume; The calculation of the effective convective potential energy density, characterizing the heat transfer potential, by combining local Reynolds number and fluid temperature difference data includes: Statistical local Reynolds number: ; in: The density of the air fluid; For the first The air velocity vector at the center of the individual element; : is a preset characteristic length constant, which is taken as the hydraulic diameter of the main heat dissipation channel of the charging module, in order to eliminate the influence of the mesh size on the Reynolds number calculation; : Dynamic viscosity of air fluid; Calculate the first Effective convective potential energy density of individual elements: ; in, For air Prandtl number; To ensure the components can withstand the upper limit of temperature, This refers to the inlet temperature of the cooling medium. This is a normalization constant used for numerical scaling, with a value of [value missing]. Its function is to map the results of complex physical unit calculations to a numerical range suitable for optimization algorithms. The larger the value, the greater the potential energy field. The numerical values ​​and their gradients A sharp increase in the value is equivalent to increasing the optimization step size, which may cause components to move too quickly, resulting in problems such as "flying out of bounds" or "significant oscillations near extreme points." Conversely, a smaller value results in a very small potential energy gradient, insufficient driving force, slow component movement, and low optimization efficiency. This paper calculates the effective convective potential energy density by comprehensively considering local Reynolds number, fluid temperature difference data, and air fluid properties. It abandons the traditional perspective of judging heat dissipation capacity solely based on flow velocity or temperature, and constructs a comprehensive physical quantity characterizing the heat transfer potential of space. This index not only reflects the flow intensity of fluid micro-particles but also incorporates the fluid's own heat capacity margin. It can accurately distinguish between pseudo-high-efficiency zones where the flow velocity is high but the temperature is already high, leading to saturation of cooling capacity, and true high-efficiency zones where the flow velocity is moderate but the temperature is extremely low, possessing great heat absorption potential. This provides a navigation map that conforms to thermodynamic principles for finding the optimal heat dissipation location for components. The step of performing a three-dimensional spatial difference operation on the potential energy density field to obtain the potential energy gradient vector at each unit location includes: Calculate the scalar field using the three-dimensional discrete central difference method In the gradient vector at individual units Identify the direction in which cooling capacity will increase the fastest: ; in: , They are respectively in Along the axis, with the current voxel The effective convective potential energy density of two adjacent voxels; , , These represent the side lengths of the voxel unit along the three coordinate axes. The potential energy gradient vector is obtained by performing a three-dimensional spatial difference operation on the potential energy density field; the potential energy density difference between adjacent voxels is processed using the three-dimensional discrete central difference method; the scalar form of the heat dissipation potential field is transformed into a vector field with a clear directionality. This allows components to move deterministically along the mathematical gradient direction where cooling capacity increases the fastest, without blindly searching randomly. This significantly improves the directness and effectiveness of the optimization path, ensuring that each movement physically points to a position with better heat dissipation, thus improving the algorithm's convergence speed. The step of integrating the gradient vector based on the heat flux requirements of the components to generate a driving vector for chasing the cold source includes: Calculate the first The heat flux demand intensity of individual components : ; in: For the first The heat loss power of components is obtained from the component specifications. For the first The physical volume of the components; The first obtained through computational fluid dynamics simulation The highest temperature value on the surface of the component; The target design temperature for the components; For the first voxel set occupied by components The gradient vector within the volume is volume-averaged to generate the main driving vector. : ; in: For the first The total number of voxels occupied by the components; For the first The set of voxel indices occupied by the components; For the set of The gradient vector of individual elements. The heat flux demand intensity is calculated based on the deviation of the component's heat loss power and surface temperature from the design target; and a chasing cold source driving vector is generated by combining the volume average gradient; a nonlinear gain correction mechanism is introduced, so that high heat-generating devices on the edge of overheating can obtain greater movement weight; this mechanism simulates the physical thermophoresis effect, ensuring that key devices with high heat flux density can preferentially seize high cooling potential energy resources; at the same time, volume integration avoids the local deviation caused by single-point sampling, ensuring that the entire physical device smoothly migrates to the optimal heat dissipation area, realizing the on-demand allocation of heat dissipation resources; The establishment of the conical projection model of the thermal wake and the calculation of the thermal wake repulsion vector experienced by downstream components include: For any upstream component and downstream components ,like lie in The semi-cone angle is Calculate the repulsion vector within the wake cone: ; in: , upstream devices and downstream devices The power; The Euclidean distance between the geometric centers of the two devices; For downstream devices From the center point to the upstream device The perpendicular distance from the flow direction axis; Indicates the wake at distance The characteristic radius at the location; the semi-cone angle The diffusion angle of the thermal wake as it propagates downstream is defined, and is set according to the wake expansion law of bluff body flow in fluid mechanics, with a value of [value missing]. The larger the value, the wider the identified wake region, resulting in downstream devices experiencing a large-scale virtual repulsion force. This leads to overly dispersed component layouts, potentially reducing space utilization or even preventing components from being placed on the PCB due to mutual repulsion. Conversely, the smaller the value, the narrower the identified wake region, only able to identify obstructions directly behind the device. If the streamlines undergo slight bends, the algorithm may fail to identify thermal interference from the "oblique rear," resulting in poor actual heat dissipation. The repulsion vector direction is perpendicular to the local streamlines and points away from the wake center, forcing downstream devices to laterally avoid the impact of the thermal wake from upstream devices. By establishing a conical projection model of the thermal wake and calculating the thermal wake repulsion vector, and using the half-cone angle to define the diffusion angle of the thermal wake propagating downstream, this model can accurately simulate the conical thermal shielding area formed by upstream heat-generating devices in the flow field. This model overcomes the shortcomings of traditional models based on simple Euclidean distance constraints that cannot reflect flow direction thermal interference. It forces downstream devices under the wake of upstream devices to produce lateral avoidance movements, effectively breaking the situation of high-heat-generating devices arranged in series, fundamentally eliminating the thermal cascade accumulation effect, and significantly improving the inlet air temperature environment of downstream devices. The generation of a flow resistance avoidance vector by combining the average flow velocity direction of the component area with the cumulative flow resistance sensitivity includes: Calculate the first The average volumetric flow rate of the region where each component is located : ; in: For the first The air velocity vector at the center of the voxel; For the first The total number of voxels occupied by the components; For the first The set of voxel indices occupied by the components; Calculate the flow resistance repulsion force that pushes the device away from the high flow resistance sensitive area: ; in: For the internal components The flow resistance sensitivity of voxels. A flow resistance avoidance vector is generated by calculating the volume average velocity direction and cumulative flow resistance sensitivity of the component region; the component is regarded as an entity with overall resistance characteristics rather than a point mass; so that the generated repulsive force can accurately reflect the overall degree of obstruction caused by the entity to the current flow field; this vector drives the component to actively withdraw from those flow resistance sensitive areas with extremely high flow velocities and chaotic directions, even if the heat dissipation potential energy of the area is high; this mechanism plays a negative feedback protection role in physics, preventing the short-sighted behavior of sacrificing the smoothness of the global flow field for local heat dissipation, and maintaining the health of the system flow field; The process of performing a weighted synthesis of the three vectors to obtain the optimized displacement and updating the component coordinates in the simulation model includes: Based on preset weights With relaxation step length Synthesize the overall optimized displacement vector: ; in: The weighting coefficient for heat dissipation optimization is set to 1, which is used to define the priority ratio of the "chasing the cold source driving vector" in the total resultant force. The flow resistance avoidance weighting coefficient is set to a value of [value missing]. This is used to adjust the contribution of the "flow resistance avoidance vector". If the value is too large, the algorithm will become conservative and the components will be pushed to the edge of the box or the low flow velocity area (where the flow resistance cost is small). This will cause the core airflow channel to be unobstructed, but the components will overheat due to lack of cooling airflow. If the value is too small, the flow resistance penalty will be ignored. The optimized layout may form a "screen"-like blockage in the center of the flow channel. Although theoretically the heat dissipation potential energy is high at this position, in practice it will cause a sharp drop in the total flow of the system, resulting in the failure of the overall simulation. To avoid weighting factors for the thermal wake, a value of [value missing] is set. This is a force used to control the "misaligned arrangement" between components. If the value is too large, the repulsive force will be too strong, and the components will move away from each other, resulting in an extremely sparse layout that may exceed the physical boundary of the PCB board, making it impossible to route. If the value is too small, the serial layout cannot be broken, and high-heat-generating components will still be arranged in a row. Downstream components will continue to draw in hot air from upstream, causing "thermal cascading" to fail. To relax the step size, , The shortest side length of the heat dissipation space for the charging module. This is the step size coefficient, with a value of [value missing]. This indicates that a single-step movement does not exceed 5%-10% of the feature size. If the value is too large, an "overshoot" phenomenon may occur, where the component may skip the optimal position or repeatedly jump between two grids, failing to converge stably. If the value is too small, the convergence will be extremely slow, requiring hundreds or thousands of iterations to move the component to the target position, which will greatly increase the computation time cost. For the first Components chase the cold source driving vector; For the first The flow resistance avoidance vector experienced by the component; For upstream components For downstream components The thermal wake repulsion vector; Update the coordinates of the components in the simulation model based on this displacement: ; in: Before the update, the first The three-dimensional coordinates of the geometric center of the component; For the simulation update The proposed three-dimensional coordinates of the components are presented. By performing weighted synthesis on three vectors—cold source chasing, flow resistance avoidance, and thermal wake repulsion—a comprehensive driving mechanism with multi-physics coupling is constructed. Different weighting coefficients are used to flexibly adjust the balance between heat dissipation requirements, aerodynamic characteristics, and topological constraints, enabling the optimization process to find the optimal compromise solution among conflicting physical objectives. With the setting of relaxation step size, overshoot or oscillation of components during the optimization process is prevented. Ultimately, a deterministic layout scheme is achieved that satisfies the independent heat dissipation requirements of each component while conforming to the overall system's optimal hydrodynamic principle.

[0026] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0027] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for optimizing the heat dissipation of a charging module based on temperature field simulation, characterized in that, include: The heat dissipation space of the charging module is discretized into voxel units, and the flow velocity and fluid temperature data of each unit are collected through simulation. Based on the angle between the unit velocity vector and the prevailing wind direction and the velocity modulus, the flow resistance sensitivity value of each unit is calculated. By combining local Reynolds number and fluid temperature difference data, the effective convective potential energy density characterizing the heat transfer potential is calculated. Perform three-dimensional spatial difference operations on the potential energy density field to obtain the potential energy gradient vector at each unit location; Based on the heat flux requirements of the components, the gradient vector is volume-integrated to generate a driving vector that chases the cold source. Establish a conical projection model of the thermal wake and calculate the thermal wake repulsion vector experienced by downstream components; By combining the average flow velocity direction in the component area with the cumulative flow resistance sensitivity, a flow resistance avoidance vector is generated; The optimized displacement is obtained by performing a weighted synthesis on the three vectors, and the component coordinates in the simulation model are updated.

2. The method for optimizing heat dissipation of a charging module based on temperature field simulation according to claim 1, characterized in that, The process of discretizing the heat dissipation space of the charging module into voxel units and collecting flow velocity and fluid temperature data of each unit through simulation includes: A discretized model of the internal heat dissipation space of the charging module is obtained and divided into multiple regular hexahedral voxel units. The side length of the voxel unit is set according to the smallest critical feature size in the charging module. Perform a single steady-state computational fluid dynamics simulation to obtain the baseline state data for each voxel element; The baseline state data includes the air velocity vector at the center of the voxel, the local fluid temperature at the voxel, and the air fluid density.

3. The method for optimizing heat dissipation of a charging module based on temperature field simulation according to claim 1, characterized in that, The calculation of the flow resistance sensitivity value for each unit is based on the angle between the unit's velocity vector and the prevailing wind direction, and the velocity modulus, including: Based on the acquired flow velocity data, a flow resistance sensitivity scalar is calculated for each voxel; The calculation of the flow resistance sensitivity scalar is proportional to the square of the magnitude of the air fluid density and the velocity vector; The calculation process introduces a direction correction term, which is determined based on the dot product of the unit direction vector of the flow velocity and the preset axial unit vector of the main heat dissipation air duct, and uses a preset turbulence correction factor to adjust the repulsion weight for non-axial flow.

4. The method for optimizing heat dissipation of a charging module based on temperature field simulation according to claim 1, characterized in that, The calculation of the effective convective potential energy density, characterizing the heat transfer potential, by combining local Reynolds number and fluid temperature difference data includes: Based on the air fluid density, the air velocity vector at the voxel center, the preset characteristic length constant, and the fluid dynamic viscosity, the local Reynolds number of each voxel is calculated. Calculate the effective convective potential energy density of each voxel. The value of this density is positively correlated with the power function of the local Reynolds number and the power function of the air Prandtl number. The calculation of the effective convective potential energy density also includes a temperature difference driving term, which is based on the difference between the upper limit temperature that the component can withstand and the local fluid temperature, and is normalized by comparing the difference between the upper limit temperature that the component can withstand and the inlet temperature of the cooling medium.

5. The method for optimizing heat dissipation of a charging module based on temperature field simulation according to claim 1, characterized in that, The step of performing a three-dimensional spatial difference operation on the potential energy density field to obtain the potential energy gradient vector at each unit location includes: The effective convective potential energy density field is processed using the three-dimensional discrete central difference method. Calculate the potential energy density difference between adjacent voxels in the three coordinate axes for the current voxel; Divide the difference by the voxel side length in the corresponding coordinate axis direction to obtain the voxel's components in the three directions, which are then combined to form a gradient vector pointing in the direction of the fastest increase in cooling capacity.

6. The method for optimizing heat dissipation of a charging module based on temperature field simulation according to claim 1, characterized in that, The step of integrating the gradient vector based on the heat flux requirements of the components to generate a driving vector for chasing the cold source includes: The heat flux demand intensity of the component is calculated, which is determined by the heat loss power density of the component, and nonlinear gain correction is performed based on the ratio of the deviation between the highest surface temperature of the component and the design target temperature. Determine the set of voxels occupied by the components in discrete space; The average gradient vector is obtained by summing and averaging the gradient vectors of all voxels in the set. The heat flux demand intensity is multiplied by the average gradient vector to generate the chasing cold source driving vector of the component.

7. The method for optimizing heat dissipation of a charging module based on temperature field simulation according to claim 1, characterized in that, The establishment of the conical projection model of the thermal wake and the calculation of the thermal wake repulsion vector experienced by downstream components include: Define a wake cone region extending from the upstream components along the flow velocity direction, which is defined by a preset half-cone angle; Determine whether the center of the downstream component is located within the wake cone region of the upstream component; If it is located within the region, the magnitude of the repulsion vector is calculated. This magnitude is directly proportional to the power product of the upstream and downstream components and inversely proportional to the square of the distance between the geometric centers of the two components. The calculation of the repulsion vector also includes a radial attenuation term, which is exponentially attenuated based on the vertical distance from the center of the downstream component to the wake axis of the upstream component. The direction of the repulsion vector is set to be perpendicular to the local streamline and pointing away from the wake centerline.

8. The method for optimizing heat dissipation of a charging module based on temperature field simulation according to claim 1, characterized in that, The generation of a flow resistance avoidance vector by combining the average flow velocity direction of the component area with the cumulative flow resistance sensitivity includes: The volume average velocity vector is obtained by calculating the arithmetic mean of the velocity vectors of all voxel units occupied by the statistical components. The cumulative flow resistance sensitivity is obtained by summing the flow resistance sensitivity values ​​of all voxel units occupied by the component. A flow resistance avoidance vector is generated, the direction of which is opposite to the direction of the volume average flow velocity vector, and its magnitude is proportional to the cumulative flow resistance sensitivity.

9. The method for optimizing heat dissipation of a charging module based on temperature field simulation according to claim 1, characterized in that, The process of performing a weighted synthesis of the three vectors to obtain the optimized displacement and updating the component coordinates in the simulation model includes: Set the weighting coefficients for heat dissipation optimization, flow resistance avoidance, and thermal wake avoidance; Set a relaxation step size, which is determined based on the minimum side length of the heat dissipation space of the charging module and a preset step size coefficient; The cold source chasing drive vector, the flow resistance avoidance vector, and the thermal wake repulsion vector applied by all upstream components are multiplied by their respective weighting coefficients. The weighted vectors are combined, and the combined result is multiplied by the relaxation step size to obtain the total optimized displacement vector. The total optimized displacement vector is superimposed onto the geometric center coordinates of the component before the update to obtain the updated three-dimensional coordinates.

Images