A method for modeling and parameter extraction of car lamp thermal network based on heat flow distribution decoupling

Abstract

This invention discloses a method for modeling and extracting parameters of automotive lighting thermal networks based on heat flow distribution decoupling, comprising the following steps: First, a steady-state temperature field and heat flow field are obtained through complete thermal simulation, and the heat flow pattern in the PCB plane and vertical plane is analyzed to identify the heat flow boundary formed by the encounter of heat flow lines between adjacent heat sources; then, multiple heat sources are decoupled according to physical phenomena, and the overall structure is divided into multiple single-heat-source sub-regions using this boundary as the dividing surface, with the dividing boundary set as an adiabatic and non-radiative surface; a physically meaningful continuous network thermal path model is constructed for each sub-region, and several thermal nodes are selected for each sub-region. Based on the geometric dimensions and material properties, key parameters such as diffusion thermal resistance, conduction thermal resistance, and heat capacity are directly calculated and extracted; finally, all sub-region models are integrated to form a high-fidelity system-level thermal network.

Description

Technical Field

[0001] This invention relates to the field of semiconductor device thermal management and computer-aided engineering (CAE) simulation technology, specifically to a method for modeling and extracting parameters of automotive lamp thermal networks based on heat flow distribution decoupling. Background Technology

[0002] Driven by global energy transition and lighting technology innovation, LED automotive lights, with their significant advantages such as energy saving, environmental friendliness, rapid response, and high compatibility with electric vehicle platforms, have gradually replaced traditional, large, and heavy halogen lamps, becoming the core solution for modern automotive lighting. In the design of automotive headlight modules, multiple high-power LED chips are often integrated onto the same PCB to meet lighting regulations and aesthetic requirements. Traditional incandescent light sources require high temperatures to produce sufficient visible light radiation. The biggest difference between LEDs, as semiconductor devices, and traditional light sources is that the luminous efficiency, lifespan, and reliability of LED devices are extremely sensitive to the junction temperature of the LED chip. Therefore, precise thermal management of LED automotive light modules is crucial.

[0003] While computational fluid dynamics (CFD) simulation is an accurate analysis tool, its time consumption and long response time cannot meet the needs of online thermal management systems and rapid design iterations. The thermal network method, which analogizes the heat transfer system to a circuit network, has the advantages of fast computation speed and easy embedding of control algorithms, making it an ideal choice for online thermal state estimation. Commonly used thermal network models include the continuous network thermal path model (Cauer model) and the local network thermal path model (Foster model). The Foster model equates all thermal resistances and capacities along the heat flow path to a first-order transfer function, which has no physical meaning. The corresponding model parameters can be obtained by fitting the temperature rise curve. The Cauer model, because its parameters, such as the selected hot nodes, thermal resistances, and heat capacities, correspond one-to-one with the actual physical structure, and each physical parameter comes from actual measurements or simulations, allows for direct modification of the thermal network parameters based on changes in the model's physical structure. The model exhibits transferability when changes occur, without the need for complex fitting processes. In summary, the Cauer thermal network model is highly interpretable and widely used in establishing LED thermal network models.

[0004] However, applying the Cauer thermal network model to tightly packed multi-heat-source LED modules faces two major challenges and bottlenecks: First, thermal coupling effects, where the heat flow from multiple heat sources diffuses and interferes with each other within the PCB plane, making it difficult to determine the independent heat transfer path of a single heat source, leading to a decrease in model accuracy; second, difficulty in parameter extraction, where it is difficult to directly delineate regions with clear physical meaning for each heat source in a strongly coupled and complex thermal field, thus making it impossible to directly calculate the diffusion thermal resistance, thermal capacity, and other parameters required for the Cauer thermal network model based on geometric and material properties. In existing research, a method for monitoring the junction temperature of IGBT modules in high-voltage frequency converters uses the physically meaningless Foster model (A method for monitoring the junction temperature of IGBT modules in high-voltage frequency converters [P]. China: CN121480326A, 2026-02-06.), which obtains the thermal capacity and thermal resistance parameters of each order of the IGBT thermal network by fitting transient thermal impedance curves. However, its interpretability at the level of physical principles is poor, and it does not delve into the thermal coupling phenomenon. When the structure of the device and the environment change, it needs to be completely rewritten. Researchers from Southwest Jiaotong University established a Foster model to form a Cauer thermal network model through Laplace transformation (A Thermal Network Modeling Method Applicable to Multi-Chip Thermal Coupling in Power Modules [P]. China: CN117852461A, 2024-04-09.). They achieved thermal coupling by injecting node heat loss into the equivalent chip using an enumeration method, achieving sufficient mathematical accuracy. However, the analysis of the thermal coupling principle has certain limitations. Therefore, there is an urgent need for a method that can effectively decouple multi-heat source effects based on the physical heat transfer mechanism and extract high-precision, highly interpretable Cauer thermal network parameters based on the clear physical structure after decoupling. Summary of the Invention

[0005] The technical problem to be solved by this invention is to overcome the problems of low accuracy, ambiguous physical meaning of parameters, and difficulty in extraction caused by thermal coupling in the existing technology of Cauer thermal network modeling of multi-heat source LED modules.

[0006] The present invention is achieved by at least one of the following technical solutions.

[0007] A method for modeling and extracting parameters of vehicle headlight thermal networks based on heat flux distribution decoupling includes the following steps: S1. Obtain the physical structure, material properties and working environment of the target LED module, and construct a geometric model including key heat dissipation structures such as LED light source, PCB board, thermal vias, prism lens, and plastic bracket. S2. Set up a dual thermal resistance model for each LED heat source, including the magnitude of the thermal resistance and the heat generation power of each LED. Mesh the dual thermal resistance model, set the material parameters, and verify the mesh independence after completing the steady-state thermal simulation to obtain the steady-state temperature field and heat flow field. S3. Obtain the heat flow vector distribution on the PCB surface and cross-section based on the steady-state field; S4. Construct a surface with its normal plane perpendicular to the PCB surface to separate the heat flow emitted by adjacent LEDs. Define the surface as the natural heat flow boundary, and the heat flow boundary does not cut any heat flow line. S5. Use all heat flow boundaries that meet the conditions as cutting surfaces, and divide the original overall model into multiple independent single heat source sub-region models through the cutting surfaces. S6. Establish a continuous network thermal path model for each single heat source sub-region, and extract the corresponding thermal network parameters based on the geometric and physical properties of each sub-region. S7. Integrate the continuous network thermal path models of all sub-regions to form a complete, parameterized system-level thermal network model.

[0008] Furthermore, in step S2, the method for determining the heat generation power of the LED is as follows: measure the input electrical power P and the total radiant flux W1 of the LED module when it is in steady state, and calculate the heat generation power W0 of the LED chip using the formula W0=P-W1.

[0009] Furthermore, the total radiative flux W1 is calculated by placing the LED module inside an integrating sphere and using spectral data measured by a spectrometer.

[0010] Further, in step S4, defining the natural heat flow boundary specifically includes: observing and adjusting the heat flow line display on the PCB surface in the post-processing interface so that the heat flow lines emitted by each LED are clearly distinguishable and do not overlap; drawing a separating curve at the contact point of the heat flow lines of every two adjacent LEDs, wherein the adjacent areas on both sides of the curve contain only the heat flow of a single LED; stretching the separating curve toward the PCB thickness direction to form a heat flow boundary surface.

[0011] Furthermore, the heat flow boundary surface satisfies the following physical characteristics: the heat flow boundary surface does not intersect with any heat flow lines, the heat flow vector at any infinitesimal point on both sides of the infinitesimal element is perpendicular to the PCB surface, and the surface does not intersect with any heat flow lines from the heat source, and the heat flow lines in the adjacent areas on both sides belong to different heat sources.

[0012] Furthermore, in step S5, the boundary conditions of the cutting surface of each sub-region formed after segmentation are set to an adiabatic surface, also known as an ideal surface. At the same time, in the setting of the radiative surface, the newly formed cutting surface of each sub-region is set to a non-radiative surface, so that these surfaces do not participate in any radiative heat transfer.

[0013] Furthermore, in step S6, the nodes of the thermal network in each sub-region include the LED chip, LED pin, average location of the end of diffused heat, top and bottom surfaces of the PCB, and the environment; The heat flow path of the continuous network thermal path model is as follows: the heat generated by the LED chip is connected to the PCB copper pad through solder, diffused through the pins to the substrate, and finally part of it is dissipated through convection radiation under the PCB board, and the other part is diffused through the copper layer to the upper surface of the PCB, and then dissipated through convection radiation on the upper surface.

[0014] Furthermore, in step S6, the extraction of the thermal capacity parameters of the thermal network includes: based on the volume, material density, and specific heat capacity of the sub-region PCB.

[0015] Furthermore, in step S6, the extraction of the thermal resistance parameters of the thermal network includes: based on the effective cross-sectional area A of the sub-region in the heat transfer direction. i The thermal conductivity K and characteristic length σ of the material are obtained through the formula R=σ / (K). A i Calculate the corresponding conduction thermal resistance R, and equate the diffusion thermal resistance to the thermal resistance of the cylindrical diffusion model. Iterate to obtain the diffusion thermal resistance of each sub-region.

[0016] Furthermore, in step S7, the continuous network thermal path model includes elements representing the internal thermal resistance of the LED particle, the thermal resistance of the solder layer, the thermal resistance of the PCB diffusion, the thermal resistance of the PCB conduction, the contact thermal resistance, the convection thermal resistance, and the radiation thermal resistance, as well as elements representing the thermal capacity of each part and each thermal resistance network node.

[0017] Compared with existing technologies, the beneficial effects of the present invention are as follows: 1. Achieve physical decoupling and improve model accuracy: By segmenting through "natural heat flow boundaries", coupling interference between heat sources is eliminated at the physical level, so that the sub-region model established for each heat source can more realistically reflect its independent heat transfer behavior, thereby significantly improving the prediction accuracy of the overall thermal network model.

[0018] 2. The parameters have clear physical meaning and strong interpretability: All thermal network parameters are calculated based on the real physical structure and materials of the decoupled sub-regions. Each thermal resistance and thermal capacity corresponds to a clear physical component and heat transfer process. The model has strong physical interpretability, which is convenient for guiding design improvement.

[0019] 3. Direct and reliable parameter extraction: It eliminates the reliance on complex mathematical fitting of the global transient response curve. The parameter extraction process is direct and stable, and the results have a clear physical basis, avoiding the uncertainty in the fitting process.

[0020] 4. Combining high fidelity and fast computation: The final system-level Cauer thermal network model inherits the high fidelity characteristics of physical simulation and has the advantage of millisecond-level computation speed of lumped parameter models, making it very suitable for embedding in vehicle ECUs for real-time junction temperature estimation and active thermal management control. Attached Figure Description

[0021] Figure 1 This is an overall flowchart of the LED vehicle lamp thermal network modeling and parameter extraction method based on heat flow distribution decoupling, which is an example.

[0022] Figure 2 This is a schematic diagram illustrating the distribution of heat flow lines and the identification of natural heat flow boundaries on the PCB surface of a multi-heat source LED module, as shown in the example.

[0023] Figure 3 This is a schematic diagram illustrating how a single heat source sub-region is formed by cutting a three-dimensional model based on a heat flow boundary surface, as an example.

[0024] Figure 4 This is a schematic diagram of a single heat source sub-region and its corresponding Cauer-type thermal network model for an example.

[0025] Figure 5 This is a schematic diagram illustrating the calculation principle of sub-region thermal network parameters (heat capacity and conduction thermal resistance) based on geometric dimensions.

[0026] Figure 6(a) is a comparison of the measured temperatures of the hybrid device provided in this embodiment.

[0027] Figure 6(b) is a comparison diagram of the thermal network model provided in this embodiment and the measured data. Detailed Implementation

[0028] 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.

[0029] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0030] like Figure 1 As shown in this embodiment, a method for modeling and extracting parameters of the thermal network of LED automotive lights based on heat flux distribution decoupling belongs to the field of semiconductor device thermal management technology. This method addresses the thermal coupling problem of multi-heat-source LED modules. First, the steady-state temperature field and heat flow field are obtained through complete thermal simulation. The heat flow pattern in the PCB plane and vertical plane is analyzed, and the heat flow boundary formed by the encounter of heat flow lines between adjacent heat sources is identified. Then, based on physical phenomena, multiple heat sources are decoupled. Using this boundary as the dividing surface, the overall structure is divided into multiple single heat source sub-regions. The dividing boundary is set as an adiabatic and non-radiative surface. Next, a physically meaningful continuous network thermal path model is constructed for each sub-region. Several thermal nodes are selected for each sub-region, and key parameters such as diffusion thermal resistance, conduction thermal resistance, and heat capacity are directly calculated and extracted based on geometric dimensions and material properties. Finally, the continuous network thermal path models of all sub-regions are integrated to form a high-fidelity system-level thermal network.

[0031] This invention solves the modeling errors and parameter extraction difficulties caused by the coupling of multiple heat sources from the perspective of physical phenomena. The model built has the advantages of high accuracy, strong physical interpretability and fast calculation, providing an efficient tool for real-time thermal state estimation and thermal management of LED vehicle lights.

[0032] Taking a certain type of automotive LED headlight module (containing 10 LEDs) as an example, the thermal network model of it is established using the method of this invention.

[0033] S1. Obtain the physical structure, material properties, and working environment of the target LED module, and construct a geometric model that includes key heat dissipation structures such as LED light source, PCB board, thermal vias, prism lens, and plastic bracket. Use the 3D CAD software Catia to establish a detailed geometric model of the module, including additional cuboid modeling of the metallized thermal vias in the PCB to accurately simulate the higher axial thermal conductivity at the thermal vias.

[0034] S2. Set up a dual thermal resistance model for each LED heat source, including the magnitude of the thermal resistance and the heat generation power of each LED. Mesh the model and set material parameters. In FloEFD software, assign material parameters to each component, such as the thermal conductivity and density of the aluminum 6061 heat sink, aluminum substrate, and thermal grease. After completing the steady-state thermal simulation, verify the mesh independence and obtain the steady-state temperature field and heat flow field.

[0035] Preferably, the method for determining the heat generation power of the LED is as follows: The heat generation power is determined experimentally. The module is placed in an integrating sphere and lit up to a thermal steady state. The input electrical power of the LED module during steady-state operation is measured as P = 34.0W. At the same time, the total radiant flux W1 = 10.5W is calculated from the data of the spectrometer. Therefore, the total heat generation power W0 = P - W1 = 23.5W is calculated by the formula, and the average heat generation of each LED is about 2.35W.

[0036] S3. Based on the steady-state field obtained in step S2, obtain the heat flow vector distribution on the PCB surface and cross-section for observation of the PCB surface. The characteristics of the heat flow are: on the surface heat flow map, the heat flows from adjacent LED heat sources meet within the PCB plane and then change direction to continue diffusing outwards from the PCB. On the cross-sectional heat flow map, the direction of diffusion after the heat flows meet is perpendicular to the PCB surface.

[0037] S4. Construct a surface whose normal plane is perpendicular to the PCB surface. This surface separates the heat flow emitted by adjacent LEDs. There is only one heat flow line emitted by a single LED on both sides of the heat flow boundary. Define this surface as the natural heat flow boundary. The heat flow boundary does not cut any heat flow line.

[0038] The specific operations for defining the natural heat flow boundary include: such as Figure 2 As shown, the thermal flow lines of the PCB are displayed in CFD post-processing. By adjusting the arrow density and size, the radial outward diffusion of heat flow emanating from beneath each LED pad can be clearly seen. Observing the area between adjacent LEDs (such as LED1 and LED2), the thermal flow lines originate from two directions and converge at a certain point in the middle. In this convergence area, a curve Γ1 is defined manually or with algorithmic assistance, positioned precisely at the interface between the two thermal flows, with only a single direction of heat flow on both sides of the curve. This process is performed on all adjacent LED pairs, resulting in a series of curves Γ1...Γ9. These curves are stretched along the PCB thickness direction to obtain three-dimensional surfaces F1...F9, which are the "thermal flow boundary surfaces".

[0039] As a further embodiment, the heat flow boundary surface satisfies the following physical characteristics: the surface does not intersect with any heat flow lines, the heat flow vector at any micro-element on both sides of the micro-element is perpendicular to the PCB surface, and the surface does not intersect with any heat flow lines from the heat source, and the heat flow lines in the adjacent areas on both sides belong to different heat sources.

[0040] S5. Using all heat flow boundaries that meet the conditions as cutting surfaces, the original overall model is divided into multiple independent single heat source sub-region models.

[0041] like Figure 3 As shown, curved surfaces F1...F9 are used as cutting tools to cut the original complete PCB and its attached heat dissipation structure into 10 independent sub-regions. Boundary conditions are set for the cut surfaces of each sub-region, designating them as adiabatic surfaces (also known as ideal surfaces). Simultaneously, in the radiating surface settings, the newly formed cut surfaces of each sub-region are set as non-radiative surfaces, ensuring these surfaces do not participate in any radiative heat transfer. Subsequently, steady-state thermal simulation is performed on each sub-region individually (only the LEDs within that sub-region are activated), and its temperature field is compared with the temperature field of the corresponding region in the original complete model. The results verify that, after applying the correct boundary conditions, the simulation results of the sub-regions are highly consistent with the corresponding regions of the original model, achieving a simulation accuracy of less than 1% in this embodiment, proving the correctness of the decoupling and boundary condition settings.

[0042] S6. Establish a continuous network thermal path model (hereinafter referred to as Cauer thermal network model) for each single heat source sub-region, and extract its thermal network parameters based on the geometric and physical properties of each sub-region.

[0043] The thermal network for each sub-region uses the following nodes: LED chip, LED pins, average location of diffused heat termination, top and bottom surfaces of the PCB, and environment. The heat flow path is as follows: heat generated by the LED chip is connected to the PCB copper pads through solder, diffused through the pins to the substrate, and finally dissipated through convection radiation from the bottom of the PCB. Another part diffuses through the copper layer to the top surface of the PCB, and then dissipates through convection radiation from the top surface. There are two directions of conductive thermal resistance that need to be identified.

[0044] The extraction of the thermal capacity parameters of the thermal network includes: based on the volume V of the sub-region PCB. i Material density ρ and specific heat capacity c are obtained through the formula C i =ρ V i c calculates the thermal capacity corresponding to the sub-region PCB. i Taking one of the sub-regions as an example ( Figure 4 Its physical structure is clear. The LED chip is connected to the 1.6mm thick PCB copper pads via solder. Heat is diffused and conducted to the FR4 substrate through the pins, and finally dissipated through convection radiation. Another part is diffused through the copper layer to the upper surface of the PCB, and then dissipated through convection radiation on the upper surface, thus constructing its structure. Figure 4 The Cauer network is shown. The heat capacity C1 of the first PCB sub-region is calculated as follows: The PCB volume V1 corresponding to this sub-region is directly measured using CAD software and is 985.13 mm³. Given that the density of the aluminum substrate ρ = 2785.3 kg / m³ and the specific heat c = 879.1 J / (kg·K), then C1 = ρ V1 c=2785.3 (985.13e-9) 879.1 = 2.41 J / K.

[0045] Extraction of the thermal resistance parameter of the heat network: based on the effective cross-sectional area A of the sub-region in the heat transfer direction. i Thermal conductivity of materials k And the characteristic length σ, obtained through the formula R=σ / ( k A i Calculate the corresponding thermal resistance R, the downward thermal resistance R. z12 The calculation is as follows: The equivalent conductive cross-sectional area A1 of the PCB in this sub-region along the thickness direction is 598.82 mm², and the total thickness is... l The longitudinal thermal conductivity of the aluminum substrate is 1.6mm. =37.9 W / (m·K). The downward thermal resistance is... Similarly, the upward thermal resistance is , This represents the characteristic length of the longitudinal thermal resistance of the PCB in the heat flow path from the LED pin to the bottom surface of the PCB.

[0046] The thermal resistance of the heat network R k The resistance to horizontal heat diffusion from the LED pads to the surrounding PCB copper and aluminum layers can be characterized by sub-region model-assisted simulation or calculation based on theoretical formulas for copper layer thickness and shape. The method employed in this invention is as follows: Figure 5 As shown, the thermal resistance expression of the diffusion thermal resistance equivalent to the inner and outer walls of a cylinder is obtained iteratively to obtain the diffusion thermal resistance of each sub-region. The outer diameter of the diffusion model of the inner and outer walls of the cylinder is... The inner diameter is The upper diffusion thermal resistance is defined as The lower layer thermal resistance is defined as It was observed that letters can be introduced. If the common part is represented in the two thermal resistance formulas, then... , Based on observation, only a small amount of heat flowed out from the top surface of the PCB, and the PCB thickness was 1.6mm. The length is defined as 1.40-1.59 mm, and the size of Z1 is limited to 3-10 iterations. The iteration result is... =1.45 mm, Z1=4.41.

[0047] The LED thermal network adopts the internationally accepted dual thermal resistance network model, and the heat capacity of the LED does not need to be considered because it is too small to have a significant impact. Other parameters, such as convective thermal resistance, can be obtained through simulation in the third step.

[0048] S7. Integrate the Cauer thermal network models of all sub-regions to form a complete, parameterized system-level thermal network model.

[0049] The Cauer-type thermal network model includes components representing the internal thermal resistance of LED particles, solder layer thermal resistance, PCB diffusion thermal resistance, PCB conduction thermal resistance, contact thermal resistance, convection thermal resistance, and radiation thermal resistance, as well as components representing the heat capacity of each part and various thermal resistance network nodes. For all 10 sub-regions, only half (five) of the symmetrical structure are actually needed. The Cauer thermal network model is established and parameters are extracted. Then, the output ports of all sub-models are connected together, connected to the thermal resistance network of the heat sink, and then connected to the "ambient end," thus forming a complete system-level thermal network model. Importing this parameterized model into Simulink, defining the heat source as the heating power changing with autotemperature and the ambient temperature, allows for the prediction of the junction temperature of each LED within milliseconds. Comparison with test data from the actual device in a room temperature chamber: Figure 6(a) shows the comparison results of the simulated and experimental steady-state temperatures at various measurement points on the PCB. The average absolute error between the two is less than 0.9°C, and the average relative error is controlled within 1.3%, verifying the correctness of the topology of the established thermal network model and the accuracy of the extracted thermal resistance parameters. Figure 6(b) further illustrates the transient temperature rise curve of a measurement point near a pin on the front of the PCB, comparing the thermal network simulation results with experimental data. Throughout the entire dynamic process from the initial temperature rise to a steady state, the simulation curve and the experimental curve show good agreement, indicating that the extraction of the heat capacity parameter has high accuracy. The model's prediction error at each measurement point across the entire operating range is less than ±1.5°C, fully meeting the requirements for formulating active thermal management strategies such as real-time thermal monitoring and fan speed regulation.

[0050] The above embodiments fully demonstrate the effectiveness of the present invention. The method provided by the present invention systematically solves the pain points of constructing multi-heat-source Cauer thermal network models, and provides a powerful tool for the thermal design and online management of automotive LED lights and other high-density electronic devices.

[0051] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, enabling those skilled in the art to better understand and utilize the invention.

Claims

1. A method for modeling and extracting parameters of vehicle lamp thermal networks based on heat flux distribution decoupling, characterized in that, Includes the following steps: S1. Obtain the physical structure, material properties and working environment of the target LED module, and construct a geometric model including key heat dissipation structures such as LED light source, PCB board, thermal vias, prism lens, and plastic bracket. S2. Set up a dual thermal resistance model for each LED heat source, including the magnitude of the thermal resistance and the heat generation power of each LED. Mesh the dual thermal resistance model, set the material parameters, and verify the mesh independence after completing the steady-state thermal simulation to obtain the steady-state temperature field and heat flow field. S3. Obtain the heat flow vector distribution on the PCB surface and cross-section based on the steady-state field; S4. Construct a surface with its normal plane perpendicular to the PCB surface to separate the heat flow emitted by adjacent LEDs. Define the surface as the natural heat flow boundary, and the heat flow boundary does not cut any heat flow line. S5. Use all heat flow boundaries that meet the conditions as cutting surfaces, and divide the original overall model into multiple independent single heat source sub-region models through the cutting surfaces. S6. Establish a continuous network thermal path model for each single heat source sub-region, and extract the corresponding thermal network parameters based on the geometric and physical properties of each sub-region. S7. Integrate the continuous network thermal path models of all sub-regions to form a complete, parameterized system-level thermal network model.

2. The method for modeling and extracting parameters of vehicle lamp thermal networks based on heat flux distribution decoupling according to claim 1, characterized in that, In step S2, the method for determining the heat generation power of the LED is as follows: measure the input electrical power P and the total radiant flux W1 of the LED module when it is in steady state, and calculate the heat generation power W0 of the LED chip using the formula W0=P-W1.

3. The method for modeling and extracting parameters of vehicle lamp thermal networks based on heat flux distribution decoupling according to claim 2, characterized in that, The total radiative flux W1 is calculated by placing the LED module inside an integrating sphere and using spectral data measured by a spectrometer.

4. The method for modeling and extracting parameters of vehicle lamp thermal networks based on heat flux distribution decoupling according to claim 1, characterized in that, In step S4, defining the natural heat flow boundary specifically includes: observing and adjusting the heat flow line display on the PCB surface in the post-processing interface so that the heat flow lines emitted by each LED are clearly distinguishable and do not overlap; drawing a separating curve at the contact point of the heat flow lines of every two adjacent LEDs, wherein the adjacent areas on both sides of the curve contain only the heat flow of a single LED; stretching the separating curve toward the PCB thickness direction to form a heat flow boundary surface.

5. The method for modeling and extracting parameters of vehicle lamp thermal networks based on heat flux distribution decoupling according to claim 1 or 4, characterized in that, The heat flux boundary surface satisfies the following physical characteristics: The heat flow boundary does not intersect with any heat flow lines. The heat flow vector at any point on either side of the micro-element is perpendicular to the PCB surface, and the surface does not intersect with any heat flow lines from the heat source. The heat flow lines in the adjacent areas on both sides belong to different heat sources.

6. The method for modeling and extracting parameters of vehicle lamp thermal networks based on heat flux distribution decoupling according to claim 1, characterized in that, In step S5, the boundary conditions of the cutting surface of each sub-region formed after segmentation are set to an adiabatic surface, also known as an ideal surface. At the same time, in the setting of the radiative surface, the newly formed cutting surface of each sub-region is set to a non-radiative surface, so that these surfaces do not participate in any radiative heat transfer.

7. The method for modeling and extracting parameters of vehicle lamp thermal networks based on heat flux distribution decoupling according to claim 1, characterized in that, In step S6, the nodes of the thermal network in each sub-region include LED chip, LED pin, average location of diffused heat termination, top and bottom surfaces of PCB, and environment; The heat flow path of the continuous network thermal path model is as follows: the heat generated by the LED chip is connected to the PCB copper pad through solder, diffused through the pins to the substrate, and finally part of it is dissipated through convection radiation under the PCB board, and the other part is diffused through the copper layer to the upper surface of the PCB, and then dissipated through convection radiation on the upper surface.

8. The method for modeling and extracting parameters of vehicle lamp thermal networks based on heat flux distribution decoupling according to claim 1, characterized in that, In step S6, the extraction of the thermal network's thermal capacity parameters includes: based on the volume, material density, and specific heat capacity of the sub-region PCB.

9. The method for modeling and extracting parameters of vehicle lamp thermal networks based on heat flux distribution decoupling according to claim 1 or 7, characterized in that, In step S6, the extraction of the thermal resistance parameters of the thermal network includes: based on the effective cross-sectional area A of the sub-region in the heat transfer direction. i The thermal conductivity K and characteristic length σ of the material are obtained through the formula R=σ / (K). A i Calculate the corresponding conduction thermal resistance R, and equate the diffusion thermal resistance to the thermal resistance of the cylindrical diffusion model. Iterate to obtain the diffusion thermal resistance of each sub-region.

10. The method for modeling and extracting parameters of vehicle lamp thermal networks based on heat flux distribution decoupling according to claim 1, characterized in that, In step S7, the continuous network thermal path model includes components representing the internal thermal resistance of LED particles, solder layer thermal resistance, PCB diffusion thermal resistance, PCB conduction thermal resistance, contact thermal resistance, convection thermal resistance and radiation thermal resistance, as well as components representing the thermal capacity of each part, and each thermal resistance network node.

Images