Adaptive Optimization Method and System for Thermal Coupling Parameters of Automotive Backlight Panels

Abstract

This application discloses an adaptive optimization method and system for thermal coupling parameters of automotive backlight panels, relating to the field of parameter simulation optimization technology. The specific implementation scheme is as follows: A full-envelope operating condition script is compiled; a thermal model and thermal coupling parameter set are imported; thermal simulations are automatically run in batches sequentially using the full-envelope operating condition script; raw datasets for each operating condition segment are collected; risky operating condition segments are screened; thermal coupling parameter perturbation groups are applied to each risky operating condition segment; high-risk operating condition segments are identified; parameter inversion is performed on each high-risk operating condition segment; and the thermal coupling parameter update strategy for each high-risk operating condition segment is determined, along with the optimal thermal coupling parameter version. This invention can improve the prediction accuracy of the thermal model under complex or extreme operating conditions, thereby avoiding problems such as underestimated junction temperature peaks, distorted hotspot locations, and drifting temperature rise rates. This provides a scientific and reliable data basis for the structural design, safety margin setting, and temperature control strategy optimization of automotive backlight panel radiators.

Description

Technical Field

[0001] This invention relates to the field of parameter simulation and optimization technology, and in particular to an adaptive optimization method and system for thermal coupling parameters of automotive backlight panels. Background Technology

[0002] With the widespread application of smart cockpits and high-brightness display technologies in the automotive field, the thermal management performance of automotive backlights has become a key factor affecting the stability and lifespan of the entire vehicle display system. A typical automotive backlight generally consists of LED modules, a PCB substrate, a metal heat sink, a light guide plate, and various interface materials. Due to the differences in material properties and structural layout of each component, complex multi-channel heat flow paths are formed between the modules. To accurately describe the heat transfer capabilities between these modules, materials, and interfaces, the thermal management field commonly uses the thermal coupling coefficient as a core parameter. The thermal coupling coefficient comprehensively reflects the influence of local structural characteristics, interface material properties, and other factors, and is the basis for predicting the temperature rise distribution of the backlight and designing thermal safety margins.

[0003] Currently, the industry largely employs simulation-based methods for thermal management optimization. A common process involves establishing a compact thermal model early in development by combining structural, material, and boundary condition information, either through one-time offline calibration or by leveraging historical experience databases, to set the thermal coupling coefficient between the LED module and the heat dissipation path. This compact thermal model can predict the temperature field distribution and potential hotspot locations of each module under typical operating conditions, aiding in heat dissipation structure design and temperature control strategy formulation. Specifically, by simulating the temperature distribution of automotive backlight panels under different typical operating conditions (such as maximum brightness, high temperature and humidity, or low temperature cold start), the compact thermal model can intuitively obtain the junction temperature peaks and hotspot locations of key components such as LED chips, PCB substrates, light guides, and heat sinks. Design engineers can then assess the heat load distribution of each module and rationally select the heat sink area, material type, and structural layout. For example, if the model predicts an abnormal temperature rise in a certain area, targeted measures such as adding heat sinks, optimizing airflow, or adjusting interface material parameters can be taken to reduce the risk of localized overheating. In addition, the compact thermal model can provide temperature rise envelope curves under different power, ambient temperature and wind speed conditions through batch simulation of operating conditions, providing a data basis for the whole vehicle thermal safety margin reservation and temperature control threshold setting, and supporting the accurate formulation of power current limiting curves and intelligent protection strategies.

[0004] However, as the actual operating conditions of automotive backlight panels become increasingly complex, real-world operation involves not only multi-level step switching of LED power but also extreme conditions such as drastic fluctuations in ambient temperature, changes in cooling airflow, and long-term aging. Under these dynamic conditions, the actual thermal coupling coefficient will drift significantly across multiple timescales, from seconds to days, due to factors such as power disturbances and interface material fatigue. For example, during accelerated aging cycles, the compressibility of the interface thermal conductive adhesive continuously decreases, leading to a gradual increase in contact thermal resistance. Consequently, the previously calibrated thermal coupling coefficient becomes invalid, causing the model output to deviate from the actual state. This drift effect is particularly pronounced in pure electric vehicles, where the display module shares power and cooling circuits with the vehicle's electrical system, resulting in frequent LED power fluctuations, concentrated heat loads, and faster aging of interface materials. However, due to limitations in existing compact thermal models, the thermal coupling coefficient is usually assumed to be constant, lacking a real-time correction mechanism for the aforementioned dynamic changes. This assumption leads to problems such as systematic underestimation of junction temperature peaks, distorted prediction of hotspot locations, and drift in temperature rise rates during batch simulations. It fails to provide a reliable basis for scientifically setting safety margins and temperature control strategies in full-coverage thermal design. Ultimately, radiator specifications, power current limiting, or protection schemes derived from distorted models often require repeated adjustments during vehicle prototyping and verification, resulting in extended development cycles and increased development costs. Summary of the Invention

[0005] This invention provides an adaptive optimization method and system for thermal coupling parameters of automotive backlight panels. This method improves the prediction accuracy of thermal models under complex or extreme conditions, thereby avoiding problems such as underestimated junction temperature peaks, distorted hotspot locations, and drift in temperature rise rates. Furthermore, it provides scientific and reliable data for the structural design, safety margin setting, and temperature control strategy optimization of automotive backlight panel radiators. The technical solution provided by this application is as follows: According to a first aspect of this application, an adaptive optimization method for thermal coupling parameters of an automotive backlight panel is provided, the method comprising: Obtain the target automotive backlight panel information, compile the full-envelope operating condition script, divide the time axis into continuous operating condition segments, and import the thermal model and thermal coupling parameter set.

[0006] The thermal simulation is automatically run in batches using the full-envelope operating condition script. During the simulation of each operating condition segment, the original dataset of each operating condition segment is collected. Based on the original dataset of each operating condition segment, the temperature dataset of each operating condition segment is extracted, the mean temperature residual of each operating condition segment is calculated, and each risky operating condition segment is screened.

[0007] Thermally coupled parameter disturbance groups are applied to each risky operating condition segment, and the degree of disturbance fluctuation in each risky operating condition segment is analyzed to determine each high-risk operating condition segment.

[0008] For each high-risk operating condition segment, parameter inversion is performed, and the optimal thermal coupling parameter solution, confidence level, and consistency score of each high-risk operating condition segment are output. This determines the thermal coupling parameter update strategy for each high-risk operating condition segment and the optimal thermal coupling parameter version.

[0009] According to a second aspect of this application, an adaptive optimization system for thermal coupling parameters of automotive backlight panels is provided. The system includes: a full-envelope operating condition script compilation module, a risk operating condition segment screening module, a high-risk operating condition segment determination module, and an optimal thermal coupling parameter version determination module.

[0010] The full-envelope working condition script compilation module is used to obtain the target automotive backlight panel information, compile the full-envelope working condition script, divide the time axis into continuous working condition segments, and import the thermal model and thermal coupling parameter set.

[0011] The risk operating condition segment screening module is used to automatically run thermal simulations in batches sequentially using full-envelope operating condition scripts. During the simulation of each operating condition segment, it collects the original dataset of each operating condition segment, extracts the temperature dataset of each operating condition segment based on the original dataset of each operating condition segment, calculates the mean residual temperature of each operating condition segment, and screens each risk operating condition segment.

[0012] The high-risk operating condition segment determination module is used to apply thermally coupled parameter perturbation groups to each high-risk operating condition segment, analyze the degree of fluctuation of the perturbation application in each high-risk operating condition segment, and then determine each high-risk operating condition segment.

[0013] The optimal thermal coupling parameter version determination module is used to perform parameter inversion for each high-risk operating condition segment, output the optimal thermal coupling parameter solution, confidence level and consistency score for each high-risk operating condition segment, thereby determining the thermal coupling parameter update strategy for each high-risk operating condition segment and determining the optimal thermal coupling parameter version.

[0014] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following: 1. This invention achieves full-coverage modeling of the entire operating cycle, extreme environments, and typical aging scenarios through full-envelope operating condition scripts and multi-scenario segmented simulations. It automatically extracts and identifies risks in the mean temperature residuals of each operating condition segment, enabling timely identification of the thermal model's adaptability shortcomings under high-risk conditions and improving its ability to detect abnormal drift. Multiple sets of thermal coupling parameter perturbations are applied to high-risk operating conditions to assess the sensitivity of each parameter, providing data for subsequent parameter inversion to lock in the controlling factors and improving the targeting and efficiency of thermal model optimization. Using a parameter inversion algorithm with constrained optimization, the thermal coupling parameters are progressively approximated and converged within the physical boundaries, ensuring the engineering feasibility and physical rationality of parameter updates. Through confidence and consistency scoring and a tiered parameter release strategy, only highly reliable and stable parameters are automatically fixed, reducing the risk of incorrect thermal model correction and runaway from the source. Finally, through full-condition simulation verification and automatic closed-loop backtracking, the local inaccuracies caused by single-condition optimization are resolved, achieving high consistency and globally optimal adaptability of the thermal model throughout its entire lifecycle.

[0015] 2. This invention can rapidly identify critical scenarios with significantly amplified thermal model errors in full-envelope, multi-condition batch simulations, avoiding the limitations of traditional manual judgment that may miss abnormal conditions. It locates the operating condition intervals most sensitive to thermal coupling coefficient drift, such as typical high-stress thermal fields like high-power step switching, low wind speed and high ambient temperature, and significant local aging, thereby improving the timeliness and accuracy of thermal model error diagnosis and abnormal drift capture. By directly marking operating condition segments where the temperature mean residual exceeds the temperature mean residual threshold as risk operating condition segments, the complex multidimensional temperature distribution information in large-scale simulation data is condensed into a risk set usable for decision-making, reducing the complexity of thermal model operation and maintenance. It also provides a more focused target area for subsequent parameter perturbation analysis and inversion solutions, improving overall computational efficiency and optimization convergence speed. Furthermore, the identification of risk operating condition segments can avoid unnecessary parameter adjustments under normal operating conditions with small thermal model errors, maintaining model structural stability and enhancing the overall engineering credibility of the thermal model.

[0016] 3. This invention improves the accuracy of parameter inversion and the efficiency of computational resource utilization by applying parameter perturbations to high-risk operating conditions and screening high-risk operating conditions based on the degree of fluctuation. From these high-risk operating conditions, key operating conditions that are more sensitive to thermally coupled parameters and can better expose structural defects in the thermal model are further identified, thereby improving the accuracy of parameter inversion and the efficiency of computational resource utilization. By analyzing the response amplitude of the mean temperature residual to parameter perturbations, it determines which high-risk operating conditions are most sensitive to changes in thermally coupled parameters, i.e., which operating conditions most realistically reflect the actual physical effects of thermal path aging, interface thermal resistance deterioration, or heat dissipation deterioration. These operating conditions are designated as high-risk operating conditions, ensuring that the data used for parameter inversion is more representative and that the inversion process focuses on the parameters that have the most significant impact on the thermal model, rather than being interfered with by low-response noise in weakly coupled operating conditions. This improves the stability and interpretability of the inversion results.

[0017] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0018] The accompanying drawings are provided for a better understanding of this solution and do not constitute a limitation of this application. Wherein: Figure 1 This is a flowchart of the adaptive optimization method for thermal coupling parameters of automotive backlight panels; Figure 2 This is a system framework diagram for adaptive optimization of thermal coupling parameters of automotive backlight panels; Figure 3 This is a flowchart for confirming high-risk operating conditions; Figure 4 This is a flowchart for determining the optimal thermal coupling parameter version; Figure 5 It is a time-series simulation curve of the average temperature during the operating condition. Detailed Implementation

[0019] The following description, in conjunction with the accompanying drawings, illustrates exemplary embodiments of this application, including various details to aid understanding. These should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this application. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0020] This invention provides an adaptive optimization method and system for thermal coupling parameters of automotive backlight panels. For example... Figure 1 The flowchart shown is for the adaptive optimization method of thermal coupling parameters for automotive backlight panels. It includes: obtaining information about the target automotive backlight panel, compiling a full-envelope working condition script, dividing the time axis into continuous working condition segments, and importing the thermal model and thermal coupling parameter set.

[0021] Specifically, the process involves obtaining information about the target automotive backlight panel and creating a full-coverage circuit operating script, as follows: Obtain information about the target automotive backlight panel, including its structural dimensions, partition layout, material type, physical performance parameters, typical operating conditions, and application environment parameters.

[0022] Set the initial values ​​of the thermal coupling parameters and the allowable disturbance range.

[0023] In specific embodiments, the initial values ​​of thermal coupling parameters are typically set based on the structural design, material properties, manufacturing process, historical experience database, and relevant experimental data of the target automotive backlight panel. For example, based on the type and thickness of the interface material between the LED module and the heat sink, the initial values ​​of parameters such as equivalent thermal resistance, interface thermal resistance, and module thermal conductivity are set by consulting the thermal conductivity parameter manual of the material supplier, actual operating condition test results, or existing simulation model historical parameters. The allowable disturbance range is set according to the physical reasonable range of the parameters, process tolerances, aging drift limits, etc., and is usually between ±10% and ±30%, and can be dynamically adjusted in conjunction with sensitivity analysis results. For example, the allowable disturbance range for interface thermal resistance can be set to [0.15, 0.45] K·cm² / W, and the disturbance range for module thermal conductivity can be set to [150, 240] W / (m·K).

[0024] Design a full-envelope line operating condition parameter script covering the entire life cycle and extreme conditions, divide the simulation time axis into multiple continuous operating condition segments, and output the full-envelope line operating condition script and operating condition segment division table.

[0025] The full-coverage line operating condition parameter script design is based on actual application scenarios of the entire vehicle and backlight panel, covering the entire life cycle of the product from new manufacturing to aging, different geographical environments, extreme temperatures and humidity, and various electrical loads. Specifically, it includes: power variation scenarios, such as minimum brightness, maximum brightness, frequent switching scenarios, and different grayscale display conditions; environmental variation scenarios, such as ambient temperature -40℃ to 85℃, humidity 20% to 95%, wind speed 0 to 3m / s, etc.; extreme and aging scenarios, such as high temperature and high humidity, high and low temperature cycling, accelerated aging (parameter drift after 1000 hours of continuous illumination), vibration and shock, etc. The script content includes the duration of each operating condition, parameter change curves, and operating condition switching logic, ensuring that all actual and boundary conditions are covered.

[0026] It should be noted that the script content for the full-envelope line operating condition parameter script mentioned above is only an example parameter for this embodiment. In actual applications, specific settings can be made according to the specific implementation scenario and simulation requirements. This application does not impose any limitations.

[0027] A full-envelope operating condition script is a script file that lists and codes all typical, extreme, and transitional operating condition parameters that need to be simulated, according to the above design principles. The script includes the operating condition type, parameter values ​​(such as power, ambient temperature, and wind speed), duration, and switching sequence. This script serves as the automatic input for batch thermal simulations.

[0028] The work condition segment division table divides the entire work condition script across the entire line into several consecutive, non-overlapping work condition segments along the timeline. Each work condition segment corresponds to a unique combination of work condition parameters. The table details the start and end times, number, corresponding parameters, and remarks for each work condition segment, facilitating subsequent data statistics and comparison.

[0029] A thermal model refers to a mathematical and physical modeling entity that reflects the structure, material properties, geometric boundaries, power distribution, and heat flow paths of a target automotive backlight module. It is typically a compact thermal model within finite element simulation software (such as ANSYS (selected in this embodiment), FloTHERM, COMSOL, etc.) or a custom simulation platform, including the definitions of each module node, thermal resistance / conduction paths, and related boundary conditions. The thermal coupling parameter set refers to the set of parameters in the aforementioned thermal model that describe the equivalent heat transfer capabilities between modules, materials, and interfaces, including module thermal conductivity, interface thermal resistance, and contact thermal resistance. Based on the input specifications of the actual simulation platform, the aforementioned model structure and parameter set are loaded in batches through model configuration files, parameter tables, database interfaces, or APIs. Common methods include: batch importing thermal model files and parameter CSV tables through scripts provided by the simulation software; and inputting the operating condition script, initial parameter values, and allowable ranges into the system all at once through a data interface or parameter configuration interface to automatically generate simulation input tasks.

[0030] It should be noted that compact thermal models are a modeling technique that abstracts complex three-dimensional heat conduction, convection, and interface effects into a network composed of nodes (representing devices / submodules) and equivalent thermal resistance, thermal capacity, heat sources, etc., to approximately describe the multi-channel, multi-material heat transfer laws within electronic devices, modules, or systems. Typically based on equivalent thermal resistance networks (RC-networks) and distributed parameter models, they are widely used as standard model types in simulation platforms (such as ANSYS Icepak, Mentor FloTHERM, Simcenter, COMSOL, etc.) and dedicated thermal analysis software.

[0031] Taking a 10-inch automotive backlight panel as an example (including an LED array, a flexible printed circuit board (FPC), a graphite heat sink, and an aluminum heat sink), we first obtain its structural dimensions (200mm long, 80mm wide, and 6mm thick), zoning layout (LED arrangement, light guide plate area), main material types (flexible printed circuit board (FPC), thermal grease, aluminum heat sink), physical performance parameters (thermal conductivity of each layer, interface thermal resistance), typical operating conditions (maximum brightness, summer exposure, winter cold start), and application environment (temperature -30℃ to 85℃, humidity 20% to 95%, wind speed 0~2m / s). Referring to the material handbook and historical simulations, we initially set the thermal conductivity of the flexible printed circuit board (FPC) to 180W / (m·K) and the interface thermal resistance to 0.23K·cm² / W, and set the allowable perturbation range to ±20%. Design a full-coverage operating script, including low-temperature cold start, maximum brightness, ambient temperature step change from -30℃ to 85℃, different wind speeds, and accelerated aging (simulating 1000 hours of operation), forming 10 typical operating condition segments and switching sequences. Import the pre-built backlight thermal network model (including all structures and boundary conditions) into the simulation platform (ANSYS Icepak), and batch load the above initial values ​​and disturbance ranges through the parameter table.

[0032] The thermal simulation is automatically run in batches using the full-envelope operating condition script. During the simulation of each operating condition segment, the original dataset of each operating condition segment is collected. Based on the original dataset of each operating condition segment, the temperature dataset of each operating condition segment is extracted, the mean temperature residual of each operating condition segment is calculated, and each risky operating condition segment is screened.

[0033] Specifically, thermal simulations are automatically run in batches sequentially using a full-circuit operating condition script, and the original datasets for each operating condition are collected during the simulation process. The specific process is as follows: Call the full-envelope operating condition script to load the thermal model and thermal coupling parameter set.

[0034] The simulation tasks for each operating condition segment are executed sequentially according to the time sequence based on the full-coverage line operating condition script.

[0035] During the simulation of each operating condition segment, the collected simulation output dataset is associated with and recorded as the operating condition segment number to which the simulation output dataset belongs.

[0036] The collected simulation output data is tagged and managed according to the number of the working condition section to form the original dataset of each working condition section.

[0037] The original datasets for each operating condition include temperature datasets, power datasets, and environmental variable datasets for each operating condition.

[0038] The temperature dataset for each operating condition refers to the collection of time-series temperature data output at various sampling points on the backlight panel during the simulation run of each operating condition, according to a set sampling frequency. Sampling points typically include LED module nodes, different areas of the FPC substrate, light guide plates, heat sink interfaces, and other representative thermal nodes. The temperature dataset reflects the dynamic temperature rise of each measuring point under that operating condition.

[0039] A power dataset refers to the time-series collection of input / output power values ​​for each heat-generating module (such as LED array, driver IC, etc.) of the backlight panel at each moment within the same operating condition. Collecting power datasets can accurately reflect the power consumption and load characteristics of each module under different operating conditions. This ensures the accuracy of simulation source terms and helps to trace the fluctuations in heat load under various operating conditions.

[0040] The environmental variable dataset refers to the time-series set of environmental boundary conditions covering various operating conditions during the simulation process, including external conditions such as ambient temperature, relative humidity, incoming airflow speed, and wind direction. Environmental variables are key external driving factors affecting the thermal management performance of the backlight panel. Ensuring that the simulation boundary conditions match the design conditions and supporting multiple operating conditions is crucial.

[0041] Continuing with the above embodiments, taking the batch simulation of a 10-inch automotive backlight panel as an example, the system calls the full-envelope operating condition script and automatically loads parameters such as different ambient temperatures (e.g., 25℃, 45℃, 75℃), LED power (e.g., 2W, 4W, 8W), and wind speeds (e.g., 0, 1.5, 3m / s) in the order specified in the script. It then executes simulation tasks for 10 operating condition segments sequentially, numbering each segment from 1 to 10. During the simulation of each operating condition segment, the system presets 6 sampling points (directly below the LED chip, center of the FPC, edge of the FPC, light guide plate interface, and inlet and outlet of the heat sink) to automatically collect temperature data once per second, forming a temperature dataset. For example, the temperature dataset for operating condition segment 3 is a full-time temperature matrix of all sampling points in that segment. Simultaneously, the power output curves of the LED driver module and other heat sources under that operating condition segment are recorded, forming a power dataset. The temporal changes in the simulation environment temperature, humidity, and wind speed are recorded, forming an environmental variable dataset. The system automatically labels and archives the above three types of datasets according to the operating condition segment number, naming them Operating Condition Segment 3 Temperature Dataset, Operating Condition Segment 3 Power Dataset, and Operating Condition Segment 3 Environmental Variable Dataset, respectively, to obtain the original dataset of Operating Condition Segment 3, providing a standardized data foundation for subsequent parameter optimization, model verification, and anomaly analysis.

[0042] like Figure 3The flowchart shown above illustrates the process for identifying high-risk operating conditions. First, temperature datasets for each operating condition are extracted, and the mean temperature for each condition is calculated. Then, the residual temperature mean is calculated. The process checks if the residual temperature mean is greater than or equal to a threshold. If this condition is met, the condition is marked as a high-risk operating condition; otherwise, it is marked as a normal operating condition. For the marked high-risk operating conditions, a thermal coupling parameter perturbation group is applied, and the rate and absolute change of the perturbed temperature mean residual are calculated to analyze the response differences after the perturbation. Then, it is determined whether a first risk criterion or a second risk criterion exists. If either criterion exists, the condition is identified as high-risk; if neither exists, it is identified as low-risk. This process enables accurate identification and classification of potential thermal anomaly risks under full-envelope operating conditions.

[0043] Specifically, the mean temperature residuals for each operating condition are calculated, and each high-risk operating condition is screened. The specific screening method is as follows: It should be explained that this invention uses temperature as the core criterion because, in the field of automotive backlight thermal management, temperature is a direct, core, and quantifiable engineering indicator for measuring the thermal performance and heat dissipation design of devices. First, temperature distribution directly reflects the heat dissipation effect and thermal safety status of each module in the backlight. The junction temperature, hot spot distribution, and temperature rise rate of key components such as LED modules, FPC substrates, light guide plates, and metal heat sinks can all be clearly expressed through temperature data from measurement points. Whether in simulation modeling or real-vehicle testing, temperature is the only physical measure that can cross multiple factors including materials, structure, power, and environment, allowing for the comparison and evaluation of different design schemes and optimization strategies. Second, temperature is the only measurement parameter that is highly unified, traceable, and engineering-closed-loop in simulation, bench testing, and actual vehicle applications. In engineering practice, temperature sensors (including thermocouples, thermistors, etc.) are easy to install, have small measurement errors, and strong data standardization, making them suitable as engineering benchmarks for model calibration, anomaly detection, and performance evaluation. Conversely, if heat flux, thermal resistance, or theoretical parameters are used as optimization targets, they are difficult to observe directly and are easily affected by material, process, and assembly deviations, leading to reduced reliability in engineering applications. Third, temperature residuals are the best criterion for evaluating thermal coupling parameters, the accuracy of thermal models, and dynamic adaptability. This scheme uses indicators such as mean temperature residuals and maximum temperature residuals to comprehensively quantify the goodness of fit between the model output and the reference operating conditions. Using temperature as a criterion facilitates the definition of the objective function for parameter inversion and the convergence determination of parameter sensitivity analysis and adaptive optimization algorithms, greatly improving the convergence, robustness, and interpretability of the optimization process.

[0044] Temperature datasets for each operating condition segment are extracted from the original datasets for each operating condition segment.

[0045] The temperature dataset for each operating condition includes the temperature output values ​​of all sampling points for each operating condition.

[0046] For a given operating condition, calculate the arithmetic mean of the temperature output values ​​at all sampling points in that operating condition to obtain the average temperature of that operating condition.

[0047] Obtain the average reference temperature corresponding to the operating condition segment, and combine it with the average temperature of the operating condition segment to obtain the residual of the average temperature of the operating condition segment.

[0048] The reference temperature mean refers to the baseline temperature data used to compare and evaluate the accuracy of the simulated temperature. In this embodiment, the empirical average data is used as the reference temperature mean.

[0049] The residual temperature mean for a given operating condition refers to the numerical difference between the mean temperature and the reference mean temperature under that operating condition. Specifically, it is defined as follows: , where ΔT mean T represents the residual of the mean temperature for this operating condition. sim，mean The average temperature for this operating condition is T. ref，mean This is the average reference temperature for this operating condition range.

[0050] By iterating through each operating condition segment, the mean temperature residual for each operating condition segment is obtained.

[0051] Operating condition segments with a mean temperature residual greater than or equal to a preset mean temperature residual threshold are designated as risky operating condition segments, while operating condition segments with a mean temperature residual less than a preset mean temperature residual threshold are designated as normal operating condition segments. In this way, risky operating condition segments are selected from each operating condition segment.

[0052] It should be noted that if the residual mean temperature value for a certain operating condition is greater than or equal to the preset residual mean temperature value threshold, it indicates that the simulation prediction for that operating condition deviates from the baseline data, and there may be a large modeling error or thermal coupling parameter mismatch, which could easily lead to engineering risks such as underestimation of safety margin or local overheating. Therefore, the corresponding operating condition is marked as a risk operating condition.

[0053] Continuing with the aforementioned embodiments, taking the batch thermal simulation of a 10-inch automotive backlight panel as an example, the system pre-sets 10 typical operating conditions, and the system executes the batch simulation task for each condition. Operating condition 1: Ambient temperature 25℃, LED power 2W, wind speed 0m / s; Operating condition 2: Ambient temperature 25℃, LED power 4W, wind speed 1.0m / s; Operating condition 3: Ambient temperature 25℃, LED power 6W, wind speed 2.0m / s; Operating condition 4: Ambient temperature 40℃, LED power 2W, wind speed 0m / s; Operating condition 5: Ambient temperature 40℃, LED power 4W, wind speed 1.0m / s; Operating condition 6: Ambient temperature 40℃, LED power 4W, wind speed 1.0m / s; Operating condition 6: Ambient temperature 40℃, LED power 4W, wind speed 1.0m / s; Operating conditions were as follows: Operating condition 40℃, LED power 6W, wind speed 2.0m / s; Operating condition 7: Ambient temperature 60℃, LED power 2W, wind speed 0m / s; Operating condition 8: Ambient temperature 60℃, LED power 4W, wind speed 1.0m / s; Operating condition 9: Ambient temperature 60℃, LED power 6W, wind speed 2.0m / s; Operating condition 10: After 800 hours of high-temperature aging, ambient temperature 45℃, LED power 4W, wind speed 1.0m / s. In each operating condition, the system collected temperature data from 6 key sampling points, including: the point directly below the LED chip, the center of the FPC, the edge of the FPC, the light guide plate interface, the heat sink inlet, and the heat sink outlet. Taking operating condition segment 1 as an example, the simulated temperatures at the measurement points are as follows: 35.8℃ directly below the LED chip, 34.5℃ at the center of the FPC, 33.4℃ at the edge of the FPC, 32.0℃ at the light guide plate interface, 31.5℃ at the heat sink inlet, and 30.7℃ at the heat sink outlet. The temperature dataset for operating condition segment 1 is the set of temperatures at these six measurement points. The average simulated temperature for operating condition segment 1 is (35.8 + 34.5 + 33.4 + 32.0 + 31.5 + 30.7) / 6 = 32.98℃. The system synchronously calls the reference average temperature for operating condition segment 1, set to 32.0℃. Therefore, the residual of the average temperature for operating condition segment 1 is 32.98℃ - 32.0℃ = 0.98℃.Using the same method, the simulated mean temperature, reference mean temperature, and residual for each of the operating conditions 2 to 10 are as follows: Operating condition 2: simulated mean temperature 34.5℃, reference mean temperature 33.3℃, ​​residual 1.2℃; Operating condition 3: simulated mean temperature 36.7℃, reference mean temperature 34.8℃, residual 1.9℃; Operating condition 4: simulated mean temperature 43.2℃, reference mean temperature 41.5℃, residual 1.7℃; Operating condition 5: simulated mean temperature 45.7℃, reference mean temperature 43.1℃, residual 2.6℃. The simulation temperature for condition 6 is 47.8℃, the reference temperature is 44.8℃, and the residual is 3.0℃; the simulation temperature for condition 7 is 62.2℃, the reference temperature is 60.8℃, and the residual is 1.4℃; the simulation temperature for condition 8 is 64.5℃, the reference temperature is 62.0℃, and the residual is 2.5℃; the simulation temperature for condition 9 is 67.0℃, the reference temperature is 64.4℃, and the residual is 2.6℃; the simulation temperature for condition 10 is 48.2℃, the reference temperature is 46.0℃, and the residual is 2.2℃. The system sets the temperature mean residual threshold to 2.0℃. It iterates through the 10 operating conditions, filtering for those with residuals greater than or equal to 2.0℃, resulting in: Operating Condition 5 (residual 2.6℃), Operating Condition 6 (residual 3.0℃), Operating Condition 8 (residual 2.5℃), Operating Condition 9 (residual 2.6℃), and Operating Condition 10 (residual 2.2℃). All of these operating conditions are automatically marked as high-risk operating conditions, forming a list of high-risk operating conditions.

[0054] See Figure 5 The figure shows the time-series simulation curves of the average temperature for each operating condition segment. It illustrates the dynamic process of the average temperature changing over time for operating conditions 1 through 5 during the simulation. Each solid line represents the upward trend of the average simulated temperature for that operating condition segment within its simulation timeframe, while the dashed lines correspond to the reference average temperature for each operating condition segment. As time progresses, the curves gradually rise and stabilize, indicating that the automotive backlight panel gradually heats up from the initial ambient temperature to a steady state within each operating condition segment. The stable value at the endpoint of each curve is the final output average simulated temperature for that operating condition segment.

[0055] Thermally coupled parameter disturbance groups are applied to each risky operating condition segment, and the degree of disturbance fluctuation in each risky operating condition segment is analyzed to determine each high-risk operating condition segment.

[0056] Specifically, thermally coupled parameter disturbance groups are applied to each risky operating condition segment, and the specific process is as follows: Determine the thermal coupling parameters of the disturbance applied to each risky operating condition segment, and denot them as the thermal coupling parameters of each disturbance applied to each risky operating condition segment. Based on the preset disturbance amplitude of the thermal coupling parameters of each disturbance applied to each disturbance, form a thermal coupling parameter disturbance group for each risky operating condition segment.

[0057] For each high-risk operating condition, the thermal coupling parameters for applying the disturbance are generally selected based on the sensitivity analysis of the mean temperature residual of that operating condition or the characteristics of the system structure. The selection principle of this invention is to select parameters that have the largest temperature rise error and the highest correlation with the main heat flow path of that operating condition, such as the interfacial thermal resistance between the LED module and the FPC, the thermal conductivity of the FPC substrate, and the convective heat transfer coefficient between the heat sink and the external environment.

[0058] The preset disturbance amplitude needs to be set in combination with material performance tolerances, process assembly fluctuations, and typical aging test data. For example, the interface thermal resistance can be set to ±10% or ±20%, the FPC thermal conductivity can be set to ±10%, and the heat transfer coefficient can be set to ±20%.

[0059] For each risky operating condition, a disturbance application step is performed, and the disturbance application fluctuation data for each risky operating condition are statistically obtained.

[0060] The fluctuation data for each disturbance include the rate of change sequence of the mean residual of the disturbance temperature and the absolute change sequence of the mean residual of the disturbance temperature.

[0061] Specifically, the disturbance application steps are as follows: A1. For a risky operating condition segment, according to the thermal coupling parameter disturbance group corresponding to the risky operating condition segment, change the thermal coupling parameter of one disturbance to the corresponding disturbance amplitude each time, while the thermal coupling parameters of the remaining disturbances remain unchanged.

[0062] A2, under the perturbation of the thermal coupling parameter, the simulation task of the risk condition segment is re-executed to obtain the original dataset of the risk condition segment, denoted as the perturbation dataset, and the mean temperature residual under the perturbation of the thermal coupling parameter is analyzed and denoted as the perturbation temperature mean residual.

[0063] A3. Extract the mean temperature residual for this risky operating condition segment, and calculate the rate of change and absolute change of the mean temperature residual in combination with the mean temperature residual of the disturbance.

[0064] A4. Repeat A2 and A3 until all thermal coupling parameters of the disturbance are simulated. From this, the rate of change sequence of the mean residual of the disturbance temperature and the absolute change sequence of the mean residual of the disturbance temperature are obtained. These are then combined and recorded as the disturbance fluctuation data of the risk condition.

[0065] Continuing with the above embodiments, taking the aforementioned risky operating conditions 5, 6, 8, 9, and 10 as examples: For each risky operating condition, assume the disturbance applied thermal coupling parameters are LED interface thermal resistance, FPC thermal conductivity, and heat sink convective heat transfer coefficient. The system sets the disturbance amplitude for each parameter as follows: LED interface thermal resistance disturbance amplitude is ±10% and ±20%; FPC thermal conductivity disturbance amplitude is ±10%; and heat sink convective heat transfer coefficient disturbance amplitude is ±20%. Taking operating condition segment 5 as an example, the specific disturbance implementation process is as follows: A1, Select the LED interface thermal resistance as the disturbance parameter. First, increase its initial value by 10% from 0.23 K·cm² / W to 0.253 K·cm² / W, keeping other parameters unchanged; A2, Rerun the simulation of operating condition segment 5 under this parameter to obtain the temperature of the new 6 sampling points. The arithmetic mean of the temperature after disturbance is 47.1℃; A3, Extract the average temperature of the simulation of operating condition segment 5 (45.7℃). The average temperature of the disturbance is 47.1℃, and the difference between the two is 1.4℃ (47.1℃-45.7℃), with a change rate of 1.4 / 45.7≈3.1%; A4, Change the LED interface thermal resistance to 0.27 K·cm² / W. 6 K·cm² / W (increased by 20%), simulation again, the average perturbation temperature is 48.6℃, the difference from the average simulation temperature (45.7℃) is 2.9℃ (48.6℃-45.7℃), the change rate is 6.3%; A5, restore the LED interface thermal resistance to the initial value, reduce the FPC thermal conductivity by 10% (from 180W / m·K to 162W / m·K), the average perturbation temperature is 46.6℃, the difference is 0.9℃ (46.6℃-45.7℃), the change rate is 2.0%; A6, finally increase the heat transfer coefficient of the heat sink by 20%, the average perturbation temperature is 44.0℃, the difference is -1.7℃ (44.0℃-45.7℃), the change rate is -3.7%. All the above disturbance data were recorded sequentially to form the mean residual change rate sequence (3.1%, 6.3%, 2.0%, -3.7%) and absolute change sequence (1.4℃, 2.9℃, 0.9℃, 1.7℃) of disturbance temperature in operating condition segment 5, which were then saved as disturbance application fluctuation data for operating condition segment 5. The system then performed the same disturbance procedure on the remaining risk operating conditions (operating conditions 6, 8, 9, and 10) to obtain the corresponding disturbance application fluctuation data.

[0066] Specifically, the degree of disturbance applied in each risky operating condition segment is analyzed to determine the high-risk operating condition segments. The specific process is as follows: The maximum values ​​of the rate of change sequence and the absolute change sequence of the mean residual of disturbance temperature are extracted for each risk condition segment to obtain the maximum residual rate of change and the maximum absolute change of residual for each risk condition segment.

[0067] The maximum residual change rate exceeding the preset residual change rate threshold is defined as the first risk criterion.

[0068] The maximum absolute change of residuals exceeding a preset absolute change threshold is defined as the second risk criterion.

[0069] It should be explained that the residual change rate threshold and the residual absolute change threshold are usually determined by product design standards, engineering experience, simulation and measurement comparisons, or industry reference values. In this embodiment, the residual change rate threshold is set to 3%, meaning that if the maximum change rate of the mean temperature residual after a disturbance is greater than 3%, it indicates that the model output is highly sensitive to disturbances of this parameter. The residual absolute change threshold is set to 1.5℃, meaning that if a disturbance causes the maximum absolute change of the mean temperature residual to exceed 1.5℃, it indicates that this parameter has a significant impact on the temperature rise deviation.

[0070] It should be noted that if the maximum residual change rate exceeds the preset residual change rate threshold, it indicates that the applied thermal coupling parameter perturbation will cause a significant change in the mean temperature residual. This suggests that the applied perturbation is highly sensitive to the model output or that the model fit is insufficient under the operating conditions, which is a risk warning and should be considered high-risk. Similarly, if the maximum absolute residual change exceeds the absolute residual change threshold, it also indicates that the applied perturbation has caused an unacceptable deviation between the model output and the baseline results, meaning that the applied thermal coupling parameters are not suitable for the current actual operating conditions, and this is also considered high-risk.

[0071] The degree of disturbance applied to risky operating conditions where either the first or second risk criterion exists is denoted as high disturbance applied fluctuation, and the degree of disturbance applied to risky operating conditions where neither the first nor the second risk criterion exists is denoted as low disturbance applied fluctuation.

[0072] In a specific embodiment, the presence of either a first or second risk criterion indicates that the thermally coupled parameters corresponding to the risky operating condition lack robustness to dynamic changes in the actual application environment, exhibiting high modeling uncertainty or thermal safety hazards, and is thus classified as high-perturbation fluctuation. The absence of either the first or second risk criterion indicates that the perturbation of the corresponding thermally coupled parameters will not significantly affect the model output; the current thermal model is stable and robust under this operating condition, and is classified as low-perturbation fluctuation, requiring no optimization.

[0073] By statistically analyzing the high-risk operating conditions caused by high disturbances, each high-risk operating condition is obtained.

[0074] Continuing with the above embodiments, disturbances were applied to the LED interface thermal resistance, FPC thermal conductivity, and heat sink convective heat transfer coefficient for risk conditions 5, 6, 8, 9, and 10, respectively. The mean residual change rate and absolute change rate sequences of the disturbance temperatures were recorded for each condition. The data are summarized as follows: For risk condition 5, the mean residual change rate sequence of the disturbance temperatures was 3.1%, 6.3%, 2.0%, and -3.7%, with absolute change rates of 1.4℃, 2.9℃, 0.9℃, and 1.7℃. For risk condition 6, the mean residual change rate sequence of the disturbance temperatures was 2.8%, 9.0%, 2.2%, and -4.0%, with absolute change rates of 1.1℃, 2.3℃, 0.8℃, and 1.9℃. The residual change rate sequence for the mean temperature disturbance in risk condition segment 8 is 1.4%, 2.5%, 1.2%, -1.7%, with absolute changes of 0.9℃, 1.1℃, 0.7℃, and 1.2℃. The residual change rate sequence for the mean temperature disturbance in risk condition segment 9 is 4.1%, 2.8%, 1.9%, -3.5%, with absolute changes of 1.8℃, 1.3℃, 0.9℃, and 1.6℃. The residual change rate sequence for the mean temperature disturbance in risk condition segment 10 is 1.0%, 2.2%, 1.1%, -1.5%, with absolute changes of 0.6℃, 1.0℃, 0.5℃, and 1.2℃. The system sets the residual change rate threshold to 3% and the absolute residual change threshold to 1.5℃. For each risk condition segment, the system extracts the maximum values ​​of the rate of change sequence and the absolute change sequence, as follows: Risk condition segment 5: maximum residual rate of change 6.3%, maximum absolute residual change 2.9℃; Risk condition segment 6: maximum residual rate of change 9.0%, maximum absolute residual change 2.3℃; Risk condition segment 8: maximum residual rate of change 2.5%, maximum absolute residual change 1.1℃; Risk condition segment 9: maximum residual rate of change 4.1%, maximum absolute residual change 1.8℃; Risk condition segment 10: maximum residual rate of change 2.2%, maximum absolute residual change 1.2℃; The system judges according to the following criteria: Risk condition segment 5: maximum residual rate of change (6.3%) > residual rate of change threshold (3%), maximum absolute residual change (2.9℃) > absolute residual change threshold (1.5℃), indicating the existence of the first risk criterion and the second risk criterion, which is recorded as high disturbance applied fluctuation. Risk Condition 6: Maximum residual change rate (9.0%) > 3%, maximum absolute change (2.3℃) > 1.5℃. Both first and second risk criteria are met, and this condition is classified as high-disturbance applied fluctuation. Risk Condition 8: Maximum residual change rate (2.5%) < 3%, maximum absolute change (1.1℃) < 1.5℃. Neither of these exceeds the threshold, and neither the first nor second risk criteria are met. This condition is classified as low-disturbance applied fluctuation. Risk Condition 9: Maximum residual change rate (4.1%) > 3%, maximum absolute change (1.8℃) > 1.5℃. Both first and second risk criteria are met, and this condition is classified as high-disturbance applied fluctuation.Risk condition segment 10: Maximum residual change rate (2.2%) < 3%, maximum absolute change (1.2℃) < 1.5℃. Neither of these exceeds the threshold, and there are no first or second risk criteria. This is recorded as a low-disturbance applied fluctuation. The system statistically analyzes the risk conditions with high-disturbance applied fluctuations and ultimately identifies risk conditions segment 5, 6, and 9 as high-risk segments. A list of high-risk segments is automatically generated, providing precise targets for subsequent parameter inversion and optimization.

[0075] For each high-risk operating condition segment, parameter inversion is performed, and the optimal thermal coupling parameter solution, confidence level, and consistency score of each high-risk operating condition segment are output. This determines the thermal coupling parameter update strategy for each high-risk operating condition segment and the optimal thermal coupling parameter version.

[0076] like Figure 4 The flowchart shown illustrates the process for determining the optimal thermal coupling parameter version. First, parameter inversion is performed on the high-risk operating condition segment. An objective function and optimization constraints are set, and a constrained optimization algorithm is executed. The inverted input thermal coupling parameters are gradually adjusted until convergence, outputting the optimal thermal coupling parameter solution, confidence level, and consistency score. Next, a thermal coupling parameter update strategy is determined, and update criteria are evaluated. If neither the first nor the second update criterion is met, the original thermal coupling parameters are maintained, and a prompt pop-up is generated. If only one update criterion is met, the thermal coupling parameters of the main thermal path are updated, thus obtaining the published thermal coupling parameter set. If both the first and second update criteria are met, the optimal thermal coupling parameter solution is used as the published thermal coupling parameter set. Finally, simulation verification is performed on the published thermal coupling parameter set to determine the optimal thermal coupling parameter version.

[0077] Specifically, parameter inversion is performed for each high-risk operating condition segment, and the optimal thermally coupled parameter solution, confidence level, and consistency score for each high-risk operating condition segment are output. The specific process is as follows: By tracing the disturbance dataset that triggered the first or second risk criterion and the corresponding high-risk operating condition segment number from the thermally coupled parameter disturbance group, the inversion input dataset for each high-risk operating condition segment is obtained.

[0078] With minimizing the mean temperature residual as the objective function, adjustable boundary conditions and optimization constraints are set for the inversion input thermal coupling parameters.

[0079] In this embodiment, the purpose of parameter inversion is to minimize the deviation between the simulated temperature and the reference temperature; therefore, minimizing the mean temperature residual is taken as the optimization objective. Specifically, for a certain high-risk operating condition segment, the temperature output values ​​of all sampling points can form the mean temperature of that operating condition segment, denoted as T. sim,mean The average reference temperature for this operating condition is denoted as T. ref,meanThis invention defines the difference between the two as the temperature mean residual, thereby obtaining the residual function R, i.e.: The smaller the mean temperature residual, the more accurately the current thermal coupling parameters reflect the true thermal behavior of the operating condition. Therefore, this invention uses the mean temperature residual as the optimization objective function, requiring the optimization algorithm to minimize the residual function R within the allowable parameter adjustment range. This objective function definition ensures that the inversion process has a clear and quantifiable optimization direction, preventing the model from deviating from actual thermal behavior and ensuring the engineering applicability of the inversion results.

[0080] To ensure the feasibility, safety, and engineering rationality of parameter inversion, adjustable boundaries and optimization constraints need to be set in advance for the thermally coupled parameters input for inversion. The setting of adjustable boundaries is generally based on the following information: the physical upper and lower limits of material properties (e.g., the feasible range of thermal conductivity), structural assembly tolerances (e.g., the fluctuation range of interfacial thermal resistance with clamping force), aging test data (e.g., the drift range of equivalent thermal resistance caused by long-term compression decay of thermally conductive adhesive), and process deviations during manufacturing (e.g., the range of heat transfer coefficient variation). These boundaries ensure that parameter inversion does not exceed the actual physical capabilities of the material or structure.

[0081] The optimization constraints include, but are not limited to: (1) physical feasibility constraints, such as thermal conductivity, interfacial thermal resistance, and heat transfer coefficient must be positive and not lower than the theoretical lower limit of the material; (2) energy conservation constraints, requiring that the error between the input heat and the thermal conduction, convection, radiation, and energy storage terms be kept within the allowable proportion; (3) parameter change rate constraints, limiting the adjustment range of the inverted parameters compared with the original parameters to no more than the set change rate threshold, so as to prevent the algorithm from jumping out of the acceptable range of actual working conditions; (4) topological sparsity constraints, forcing the coupling parameters of non-physical thermal paths to remain at zero, avoiding the appearance of coupling terms that do not conform to the real structure. This ensures that the inverted parameters will not have physically unreasonable values; ensures that the inversion process is stable and convergent, and does not obtain results that are mathematically better but not usable in engineering; and ensures that the parameter changes between different working conditions are continuous, interpretable, and consistent with engineering.

[0082] A constrained optimization algorithm is used to perform parameter inversion, gradually adjusting the inversion input thermal coupling parameters and calculating the mean temperature residual until convergence.

[0083] Constrained optimization algorithms refer to optimization algorithms that simultaneously satisfy several physical boundary constraints and engineering constraints during the optimization of the objective function. Constrained optimization algorithms that can be used in this invention include, but are not limited to: Sequential Least Squares Programming (SLSQP, selected in this embodiment), penalty function method, trust region inversion algorithm, and L-BFGS-B constrained optimization algorithm. These algorithms can handle continuous variables, boundary conditions, and linear and nonlinear constraints, ensuring that the thermally coupled parameters remain within a set adjustable range during the optimization process. Thermally coupled parameters typically have strict physical boundaries; directly using unconstrained optimization may lead to parameter out-of-bounds errors, non-convergence, or inversion results that cannot be applied to practical engineering. The constrained optimization method used in this invention can obtain the optimal parameter solution while ensuring physical rationality, improving the model's credibility and robustness.

[0084] The thermal coupling parameters at convergence are the optimal thermal coupling parameter solution, and the confidence and consistency scores are calculated.

[0085] The parameter inversion process of this invention includes the following strict time-series steps: First, the original dataset corresponding to the high-risk operating condition is input into the optimization module, and the aforementioned objective function, adjustable boundary, and constraint conditions are invoked. Second, the optimization algorithm performs a small perturbation on each thermal coupling parameter to be adjusted based on the initial thermal coupling parameter values, and calculates the mean temperature residual after the perturbation in real time. Third, the algorithm adjusts the parameter search direction and step size according to the residual changes, using gradient, quasi-Newton, or trust region methods. Fourth, in each iteration, all constraint conditions are checked for satisfaction; if violated, the search interval is shrunk or replanned. Fifth, the iteration is repeated until the mean temperature residual converges to the minimum value. Finally, the thermal coupling parameter value output at the convergence moment is the optimal thermal coupling parameter solution for the high-risk operating condition. Simultaneously, the confidence level and consistency score are calculated based on the residual convergence speed, constraint satisfaction degree, and consistency of the multi-round inversion results.

[0086] In this embodiment, after completing the parameter inversion for each high-risk operating condition segment, the reliability and stability of the inversion results need to be quantitatively evaluated to output confidence and consistency scores. This process includes the following specific steps: During the constrained optimization algorithm inversion process, the system records the numerical sequence of the mean temperature residual after each parameter adjustment, calculates the number of iteration steps required for the residual to converge from the initial value to the minimum value, and statistically analyzes the magnitude of the residual decrease at each step. The residual iteration sequence {R0, R1, R2, ..., R...} is used as the basis for the calculation. n} Calculate the amount of descent ΔR for each time. i =R i-1-Ri, where i is the residual number of the residual iteration sequence, i=1,2,...,n, and n is the number of residuals in the residual iteration sequence. The average convergence step size and maximum single-step descent are further calculated. If the residual convergence speed is fast and the final value is significantly lower than the initial value (e.g., the residual descent rate exceeds 90%), the parameter inversion is judged to have high convergence confidence. Throughout the parameter inversion process, the system continuously monitors whether each thermally coupled parameter remains within the preset physical boundaries and process allowable range. All parameters must meet the adjustable boundaries (e.g., parameter values ​​do not exceed the upper and lower physical limits, and the energy closure error is less than 3%); otherwise, the inversion result is considered invalid. The proportion of the final convergence result that satisfies all constraints is statistically analyzed and used as the basis for constraint satisfaction scoring. If all constraints are satisfied and the parameter change rate is within a reasonable range, it is considered a high degree of constraint satisfaction. The system repeats the inversion process multiple times with different initial parameters, recording the final converged thermally coupled parameters and the corresponding minimum residual value in each round. Statistical analysis of the results from multiple rounds reveals that if the optimal parameter solutions obtained in each round are highly consistent within the allowable fluctuation range (e.g., the maximum parameter deviation does not exceed 5%), and the final residual values ​​are similar, it indicates that the inversion process has good stability and convergence consistency. The system calculates a parameter consistency score based on this. Finally, the system combines the residual convergence speed score, constraint satisfaction score, and multi-round inversion consistency score in a weighted manner to form a total confidence score and a consistency score. Specific scoring methods can employ linear weighting or interval criterion methods to ensure that the scoring criteria are clear, quantifiable, and verifiable.

[0087] Continuing with the above embodiments, for the high-risk operating condition segment 5 that has been screened out, the system first traces the original dataset and thermal coupling parameter perturbation group that triggered the risk judgment for this segment, and determines the inversion input data, including temperature time-series data, power data, environmental variable data, and existing initial values ​​of thermal coupling parameters from 6 sampling points. To optimize the solution, the system sets the adjustable range of LED interface thermal resistance to 0.20~0.30 K·cm. 2 / W, FPC thermal conductivity is 160~200W / m·K, and heat sink convective heat transfer coefficient is 20~60W / m 2 The parameter is set to K, and constraints are set: the parameter must be positive, must not exceed the above range, and the energy closure error must not exceed 3%. With the goal of minimizing the mean temperature residual, the sequential least squares programming (SLSQP) algorithm is used to perform constrained parameter inversion. After the initial parameters are input, the algorithm gradually adjusts the LED interface thermal resistance (from 0.23 K·cm⁻¹). 2 / W to 0.24K·cm 2 / W to 0.25K·cm 2 / W to 0.26K·cm 2 / W), FPC thermal conductivity (from 180W / m·K to 175W / m·K to 170W / m·K), heat transfer coefficient (from 45W / m²·K to 50W / m²·K). 2 ·K to 55W / m2 For each adjustment, the algorithm automatically simulates and recalculates the average temperature of operating condition segment 5 (45.7℃ to 44.8℃ to 43.9℃), compares it with the reference average temperature (43.1℃), and updates the residual sequence {2.6, 1.7, 0.8, 0.5, 0.2, 0.1}℃. At each step, the algorithm checks whether the parameters exceed limits and whether the energy closure constraints are met. If any violation occurs, it immediately backtracks to ensure that all constraints are strictly satisfied. Ultimately, the residual converges to 0.1℃, and the parameters stabilize at the LED interface thermal resistance of 0.247 K·cm. 2 / W, FPC thermal conductivity 170W / m·K, heat transfer coefficient 53W / m 2 K. Subsequently, based on the residual convergence speed (8 iterations in this round, residual temperature decreased from 2.6℃ to 0.1℃, a decrease of over 96%), the absence of all parameters exceeding limits, and the energy closure error consistently <2%, the system was scored as follows: convergence confidence score 95 points, constraint satisfaction score 100 points. Then, the inversion was repeated 3 times with different initial parameters perturbed by ±5%. All optimal parameter deviations were less than 3%, and the final residual values ​​were all within 0.15℃, resulting in a consistency score of 95 points. Finally, the system linearly weighted the calculation to obtain a total confidence score (confidence level) of 95 points and a consistency score of 95 points for this round of parameter inversion.

[0088] Specifically, the thermal coupling parameter update strategy for each high-risk operating condition section is determined, and the optimal thermal coupling parameter version is determined. The specific determination process is as follows: A confidence level greater than a preset confidence threshold is used as the first update criterion.

[0089] The consistency score being greater than the preset consistency score is used as the second update criterion.

[0090] The confidence threshold and consistency score threshold are typically determined based on practical engineering experience, historical project data, industry standards, and statistical analysis of experimental inversion. A range of 80 to 90 points is generally recommended. Specifically, the confidence threshold can be set based on the convergence speed of parameter inversion, the magnitude of residual reduction, and the low number of false alarms and missed alarms observed in multiple batches of actual verification, such as 85 points. The consistency score threshold is set based on historical statistical scores showing small deviations in parameter solutions, stable results, and reproducibility across multiple rounds of inversion, such as 90 points. The main purpose of setting these thresholds is to balance the reliability and flexibility of parameter updates, ensuring engineering safety while avoiding excessive conservatism that could prevent the model from adaptively optimizing.

[0091] A confidence score greater than the confidence threshold indicates that the parameter inversion converges quickly, the residuals decrease significantly, and all constraints are met, demonstrating a stable and reliable inversion process and highly reliable optimal thermally coupled parameters. A consistency score greater than the consistency score threshold indicates that the inversion results with multiple sets of different initial parameters are highly consistent and converge to similar optimal solutions, suggesting that the algorithm is not sensitive to initial values ​​and the inversion solution has high stability and repeatability. Parameter solutions that meet the above conditions have both physical meaning and can be reproduced under different starting points, thus providing a basis for automatic publishing and engineering solidification.

[0092] For high-risk operating conditions where neither the first nor the second update criterion is met, the thermal coupling parameter update strategy is recorded as maintaining the original thermal coupling parameters, and a prompt pop-up window is generated.

[0093] Failure to meet both the first and second update criteria indicates that the inversion results of the corresponding thermal coupling parameters fluctuate greatly, have poor reliability, and the model's convergence or reproducibility is insufficient. In this case, continuing to use the original thermal coupling parameters is the safest approach, and manual intervention should be suggested to prevent the simulation model from becoming distorted due to low-quality parameters.

[0094] The thermal coupling parameter update strategy for high-risk operating conditions that only satisfy any one update criterion is denoted as updating the thermal coupling parameters of the main thermal path. This means replacing the original thermal coupling parameters of the main thermal path with the corresponding thermal coupling parameters in the optimal thermal coupling parameter solution, while keeping the other thermal coupling parameters unchanged. This results in the release of the thermal coupling parameter set.

[0095] Satisfying any one of the update criteria indicates that the model has a certain convergence advantage or good stability, but some uncertainties still exist. In this case, it is recommended to update only the main thermal path parameters, that is, only replace the most critical thermal coupling parameters that have a large impact on the residual (such as LED interface thermal resistance), while leaving the other thermal coupling parameters unchanged, thereby balancing model optimization and engineering safety.

[0096] The thermal coupling parameter update strategy for high-risk operating conditions that satisfy both the first and second update criteria is denoted as a full update, which means that the optimal thermal coupling parameter solution is used as the published thermal coupling parameter set.

[0097] Satisfying both the first and second update criteria indicates that the parameter inversion results meet the standards in terms of reliability and consistency, the model converges stably, and the inversion solution is highly reliable. At this point, the entire set of optimal thermally coupled parameter solutions can be solidified into a published thermally coupled parameter set to achieve global adaptive optimal correction.

[0098] The published thermal coupling parameter set was simulated and verified to determine the optimal thermal coupling parameter version.

[0099] In a specific embodiment, the published thermal coupling parameter set is simulated and verified to determine the optimal thermal coupling parameter version. The specific process is as follows: Based on the published thermal coupling parameter set, thermal simulation is re-executed for all operating condition segments in sequence according to the full envelope operating condition script.

[0100] Output the original verification dataset for each operating condition segment, and calculate the mean residual of the verification temperature for each operating condition segment.

[0101] If the residual of the mean temperature in all operating conditions is less than the threshold of the residual of the mean temperature, then the thermal coupling parameter set is solidified and released as the optimal parameter version.

[0102] If any operating condition is not met, the process will backtrack to the parameter inversion stage.

[0103] If the residual values ​​of the mean temperature are less than the threshold value for all operating conditions, it indicates that the current thermal coupling parameter set can cover all actual operating conditions, achieving accurate global modeling and meeting the requirements of subsequent design, control, and safety assessment. This set can be solidified as the optimal thermal coupling parameter version and used as the basis for model finalization and archiving. If any operating condition fails to meet the requirements, it indicates that the current parameter set cannot accurately fit the actual thermal behavior under certain operating conditions, and residuals remain, posing potential error risks. In this case, parameter inversion or adjustment of the optimization strategy is necessary to ensure that the final parameter solution has full operating condition adaptability.

[0104] To prevent falling into an infinite loop, this invention sets a maximum number of backtracking attempts (e.g., 3 to 5). If, after multiple backtracking attempts to the parameter inversion stage, the operating condition still fails to meet the standards, the system should automatically pause, output all historical optimization records and non-converged operating condition reports, prompting the user to perform manual analysis or adjust the model structure. This ensures process automation and avoids the model falling into meaningless, blind loops, balancing efficiency and engineering safety.

[0105] After completing the parameter inversion for high-risk operating conditions 5, 6, and 9, the system statistically obtains the optimal thermal coupling parameter solutions, confidence scores, and consistency scores for each condition. Assumptions: The confidence score and consistency score for the parameter inversion of high-risk operating condition 5 are 95; the confidence score and consistency score for the parameter inversion of high-risk operating condition 6 are 83; and the confidence score and consistency score for the parameter inversion of high-risk operating condition 9 are 78; the consistency score is 77. Based on the set thresholds (confidence threshold 85, consistency score threshold 90), the system makes the following automatic decision: High-risk operating condition 5 simultaneously satisfies both the first and second update criteria (95 > 85, 94 > 90), and adopts a full update strategy, that is, all the optimal thermal coupling parameters of high-risk operating condition 5 (such as LED interface thermal resistance 0.247 K·cm² / W, FPC thermal conductivity 170 W / m·K, and heat transfer coefficient 53 W / m²·K) are replaced into the published thermal coupling parameter set. High-risk operating condition segment 6 only meets the consistency score (91 points > 90 points), but the confidence level is slightly lower (83 points < 85 points). The main thermal path update strategy is adopted, replacing only the main thermal path (such as the LED interface thermal resistance) with the inversion optimal value, while keeping other parameters unchanged. High-risk operating condition segment 9 does not meet any update criteria, so the system determines to maintain the original parameters and displays a pop-up message indicating that the parameter inversion for high-risk operating condition segment 9 has not passed the automatic release threshold and requires manual review. After summarizing all released thermal coupling parameters, the system automatically calls the full-envelope operating condition script based on this parameter set to re-execute thermal simulation for all 10 operating condition segments. The new simulation results are as follows (excerpt of core data): The residual temperature mean for operating condition 1 is 0.7℃, for operating condition 2 it is 0.9℃, for operating condition 3 it is 1.1℃, for operating condition 4 it is 0.8℃, for operating condition 5 it is 0.2℃, for operating condition 6 it is 0.7℃, for operating condition 7 it is 1.0℃, for operating condition 8 it is 1.3℃, for operating condition 9 it is 1.9℃, and for operating condition 10 it is 1.2℃. The residual temperature mean threshold remains at 2.0℃. This round of verification results shows that the residual temperature mean for all operating conditions is less than the residual temperature mean threshold. The system automatically determines that the released thermal coupling parameter set has passed the full-condition simulation verification, solidifies this parameter set as the optimal thermal coupling parameter version, and archives all simulation and comparison data, as well as version change logs.

[0106] like Figure 2 The diagram shown is a framework diagram of an adaptive optimization system for thermal coupling parameters of automotive backlight panels. It includes a full-envelope operating condition script compilation module, a risk operating condition segment screening module, a high-risk operating condition segment determination module, and an optimal thermal coupling parameter version determination module.

[0107] The full-envelope working condition script compilation module is used to obtain the target automotive backlight panel information, compile the full-envelope working condition script, divide the time axis into continuous working condition segments, and import the thermal model and thermal coupling parameter set.

[0108] The risk operating condition segment screening module is used to automatically run thermal simulations in batches sequentially using full-envelope operating condition scripts. During the simulation of each operating condition segment, it collects the original dataset of each operating condition segment, extracts the temperature dataset of each operating condition segment based on the original dataset of each operating condition segment, calculates the mean residual temperature of each operating condition segment, and screens each risk operating condition segment.

[0109] The high-risk operating condition segment determination module is used to apply thermally coupled parameter perturbation groups to each high-risk operating condition segment, analyze the degree of fluctuation of the perturbation application in each high-risk operating condition segment, and then determine each high-risk operating condition segment.

[0110] The optimal thermal coupling parameter version determination module is used to perform parameter inversion for each high-risk operating condition segment, output the optimal thermal coupling parameter solution, confidence level and consistency score for each high-risk operating condition segment, thereby determining the thermal coupling parameter update strategy for each high-risk operating condition segment and determining the optimal thermal coupling parameter version.

[0111] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims can be performed in a different order than that shown in the embodiments and still achieve the desired results. Additionally, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are also possible or may be advantageous.

[0112] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular...

[0113] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. An adaptive optimization method for thermal coupling parameters of automotive backlight panels, characterized in that, The method includes: Obtain the target automotive backlight panel information, compile the full-envelope working condition script, divide the time axis into continuous working condition segments, and import the thermal model and thermal coupling parameter set; The thermal simulation is automatically run in batches sequentially using the full-envelope working condition script. During the simulation of each working condition segment, the original dataset of each working condition segment is collected. Based on the original dataset of each working condition segment, the temperature dataset of each working condition segment is extracted, the mean temperature residual of each working condition segment is calculated, and each risky working condition segment is screened. Thermally coupled parameter disturbance groups are applied to each risky operating condition segment, and the degree of disturbance application fluctuation in each risky operating condition segment is analyzed to determine each high-risk operating condition segment. For each high-risk operating condition segment, parameter inversion is performed, and the optimal thermal coupling parameter solution, confidence level, and consistency score of each high-risk operating condition segment are output. This determines the thermal coupling parameter update strategy for each high-risk operating condition segment and the optimal thermal coupling parameter version.

2. The adaptive optimization method for thermal coupling parameters of automotive backlight panels as described in claim 1, characterized in that, The specific process of obtaining the target automotive backlight panel information and compiling the full-envelope circuit operating script is as follows: Obtain information about the target automotive backlight panel, including its structural dimensions, partition layout, material type, physical performance parameters, typical operating conditions, and application environment parameters. Set the initial values ​​of the thermal coupling parameters and the allowable disturbance range; Design a full-envelope line operating condition parameter script covering the entire life cycle and extreme conditions, divide the simulation time axis into multiple continuous operating condition segments, and output the full-envelope line operating condition script and operating condition segment division table.

3. The adaptive optimization method for thermal coupling parameters of automotive backlight panels as described in claim 1, characterized in that, The process of automatically running thermal simulations in batches using a full-envelope operating condition script and collecting raw datasets for each operating condition segment during the simulation is as follows: Call the full-envelope working condition script to load the thermal model and thermal coupling parameter set; The simulation tasks for each operating condition segment are executed sequentially according to the time sequence based on the full-coverage line operating condition script; During the simulation run of each working condition segment, the collected simulation output dataset is associated with and recorded as the working condition segment number to which the simulation output dataset belongs. The collected simulation output data is tagged and managed according to the working condition section number to form the original dataset of each working condition section. The original datasets for each operating condition segment include temperature datasets, power datasets, and environmental variable datasets for each operating condition segment.

4. The adaptive optimization method for thermal coupling parameters of automotive backlight panels as described in claim 1, characterized in that, The method for calculating the mean temperature residual for each operating condition segment and screening for risky operating conditions is as follows: Temperature datasets for each operating condition segment are extracted from the original datasets for each operating condition segment. The temperature dataset for each operating condition includes the temperature output values ​​of all sampling points for each operating condition. For a given operating condition, calculate the arithmetic mean of the temperature output values ​​of all sampling points in that operating condition to obtain the average temperature of that operating condition. Obtain the average reference temperature corresponding to the operating condition segment, and combine it with the average temperature of the operating condition segment to obtain the residual of the average temperature of the operating condition segment; By iterating through each working condition segment, the mean temperature residual for each working condition segment is obtained; The operating condition segment where the mean temperature residual is greater than or equal to the preset mean temperature residual threshold is recorded as the risk operating condition segment, and thus each risk operating condition segment is selected from each operating condition segment.

5. The adaptive optimization method for thermal coupling parameters of automotive backlight panels as described in claim 1, characterized in that, The specific process of applying thermally coupled parameter disturbance groups to each risky operating condition segment is as follows: Determine the thermal coupling parameters of the disturbance applied to each risk condition segment, and denot them as the thermal coupling parameters of each disturbance applied to each risk condition segment. Based on the preset disturbance amplitude of each disturbance applied to each thermal coupling parameter, form a thermal coupling parameter disturbance group for each risk condition segment. For each risky operating condition segment, a disturbance application step is performed, and the disturbance application fluctuation data for each risky operating condition segment are statistically obtained. The fluctuation data applied for each disturbance include a sequence of the rate of change of the mean residual of the disturbance temperature and a sequence of the absolute change of the mean residual of the disturbance temperature.

6. The adaptive optimization method for thermal coupling parameters of automotive backlight panels as described in claim 5, characterized in that, The disturbance application step is specifically as follows: A1. For a risky operating condition segment, according to the thermal coupling parameter disturbance group corresponding to the risky operating condition segment, change the thermal coupling parameter of one disturbance to the corresponding disturbance amplitude each time, while the thermal coupling parameters of the other disturbances remain unchanged. A2, under the perturbation of the thermal coupling parameter, the simulation task of the risk condition segment is re-executed to obtain the original dataset of the risk condition segment, denoted as the perturbation dataset, and the mean temperature residual under the perturbation of the thermal coupling parameter is analyzed and denoted as the perturbation temperature mean residual. A3, extract the mean temperature residual for this risky operating condition segment, and calculate the rate of change and absolute change of the mean temperature residual in combination with the mean temperature residual of the disturbance. A4. Repeat A2 and A3 until all thermal coupling parameters of the disturbance are simulated. From this, the rate of change sequence of the mean residual of the disturbance temperature and the absolute change sequence of the mean residual of the disturbance temperature are obtained. These are then combined and recorded as the disturbance fluctuation data of the risk condition.

7. The adaptive optimization method for thermal coupling parameters of automotive backlight panels as described in claim 1, characterized in that, The analysis of the disturbance application fluctuation degree in each risky operating condition segment, and the determination of each high-risk operating condition segment, is carried out in the following specific process: Extract the maximum values ​​of the rate of change sequence and the absolute change sequence of the mean residual of disturbance temperature for each risk condition segment to obtain the maximum residual rate of change and the maximum absolute change of residual for each risk condition segment. The maximum residual change rate exceeding the preset residual change rate threshold is defined as the first risk criterion. The maximum absolute change in residuals exceeding a preset threshold for absolute change in residuals is defined as the second risk criterion. The degree of disturbance applied to the risky operating condition segment with the first risk criterion or the second risk criterion is defined as high disturbance applied fluctuation; The disturbance applied fluctuation level in the risk condition segment where neither the first risk criterion nor the second risk criterion exists is recorded as low disturbance applied fluctuation; By statistically analyzing the high-risk operating conditions caused by high disturbances, each high-risk operating condition is obtained.

8. The adaptive optimization method for thermal coupling parameters of automotive backlight panels as described in claim 1, characterized in that, The process involves performing parameter inversion for each high-risk operating condition segment, outputting the optimal thermally coupled parameter solution, confidence level, and consistency score for each high-risk operating condition segment. The specific process is as follows: From the thermally coupled parameter perturbation group, trace back the perturbation dataset that triggered the first risk criterion or the second risk criterion and the corresponding high-risk operating condition segment number to obtain the inversion input dataset for each high-risk operating condition segment; With minimizing the mean temperature residual as the objective function, adjustable boundary conditions and optimization constraints are set for the inversion input thermal coupling parameters. A constrained optimization algorithm is used to perform parameter inversion, gradually adjusting the inversion input thermal coupling parameters and calculating the mean temperature residual until convergence. The thermal coupling parameters at convergence are the optimal thermal coupling parameter solution, and the confidence and consistency scores are calculated.

9. The adaptive optimization method for thermal coupling parameters of automotive backlight panels as described in claim 1, characterized in that, The specific process for determining the thermal coupling parameter update strategy for each high-risk operating condition segment and determining the optimal thermal coupling parameter version is as follows: The confidence level being greater than a preset confidence threshold is used as the first update criterion; A consistency score greater than the preset consistency score will be used as the second update criterion. The thermal coupling parameter update strategy for high-risk operating conditions that do not meet the first update criterion and the second update criterion is recorded as maintaining the original thermal coupling parameters, and a prompt pop-up window is generated. The thermal coupling parameter update strategy for high-risk operating conditions that only satisfy any one update criterion is denoted as updating the thermal coupling parameters of the main thermal path. That is, the original thermal coupling parameters of the main thermal path are replaced with the corresponding thermal coupling parameters in the optimal thermal coupling parameter solution, while the other thermal coupling parameters remain unchanged. This results in the release of the thermal coupling parameter set. The thermal coupling parameter update strategy for high-risk operating conditions that satisfy both the first and second update criteria is denoted as a complete update, i.e., the optimal thermal coupling parameter solution is used as the published thermal coupling parameter set. The published thermal coupling parameter set was simulated and verified to determine the optimal thermal coupling parameter version.

10. An adaptive optimization system for thermal coupling parameters of an automotive backlight panel, used to implement the adaptive optimization method for thermal coupling parameters of an automotive backlight panel as described in any one of claims 1-9, characterized in that, The system includes: a full-envelope working condition script compilation module, a risk working condition section screening module, a high-risk working condition section determination module, and an optimal thermal coupling parameter version determination module; Among them, the full-envelope working condition script compilation module is used to obtain the target automotive backlight panel information, compile the full-envelope working condition script, divide the time axis into continuous working condition segments, and import the thermal model and thermal coupling parameter set. The risk operating condition segment screening module is used to automatically run thermal simulations in batches sequentially using full-envelope operating condition scripts, and to collect the original datasets of each operating condition segment during the simulation process. Based on the original datasets of each operating condition segment, the temperature datasets of each operating condition segment are extracted, the mean temperature residuals of each operating condition segment are calculated, and each risk operating condition segment is screened. The high-risk operating condition segment determination module is used to apply thermally coupled parameter perturbation groups to each high-risk operating condition segment, analyze the degree of fluctuation of perturbation application in each high-risk operating condition segment, and then determine each high-risk operating condition segment. The optimal thermal coupling parameter version determination module is used to perform parameter inversion for each high-risk operating condition segment, output the optimal thermal coupling parameter solution, confidence level and consistency score for each high-risk operating condition segment, thereby determining the thermal coupling parameter update strategy for each high-risk operating condition segment and determining the optimal thermal coupling parameter version.

Images