Heat dissipation control method and device for high-level computing power autonomous driving FPC sensor module

Abstract

The present invention relates to a heat dissipation control method and device for a high-order computing power autonomous driving FPC sensor module, comprising the following steps: collecting temperature data of a target FPC sensor module through a preset distributed thermistor array to obtain three-dimensional temperature field data; performing heat flow path calculation on the target FPC sensor module based on the three-dimensional temperature field data to obtain a heat flux density distribution diagram; performing topology optimization calculation on the target FPC sensor module based on the heat flux density distribution diagram to obtain a heat dissipation structure layout scheme; dynamically optimizing and solving the heat dissipation structure layout scheme through a branch and bound algorithm to obtain a heat dissipation control parameter set; inputting the heat dissipation control parameter set into a preset execution unit, and performing real-time heat dissipation control on the target FPC sensor module to obtain an ideal heat dissipation state, thereby solving the problem that traditional methods are often unable to accurately and effectively cool local hot spots and lack the flexibility to adapt to dynamic thermal changes under different workloads.

Description

Technical Field

[0001] The present invention relates to the technical field of FPC sensor modules, and in particular to a heat dissipation control method and device for a high-order computing power autonomous driving FPC sensor module. Background Art

[0002] With the rapid development of autonomous driving technology, the complexity and computing power requirements of automotive electronic systems are also increasing. In particular, high-level computing power autonomous driving platforms rely on high-performance computing modules to process data from various sensors (such as cameras, radars, lidars, etc.) to achieve functions such as environmental perception, path planning, and decision control. The FPC (flexible printed circuit board) components in these sensor modules generate a lot of heat during operation, and excessive temperatures will directly affect the working efficiency and life of electronic components, and even cause system failures. Therefore, how to effectively manage the heat generated by these high-density integrated sensor modules has become an important challenge to ensure the reliability and stability of autonomous driving systems.

[0003] Most current heat dissipation solutions focus on traditional cooling methods, such as air cooling or liquid cooling systems, but these methods are not always applicable to FPC sensor modules with compact designs and strict space constraints. Traditional methods often cannot accurately target local hot spots for effective cooling, and lack the flexibility to adapt to dynamic thermal changes under different workloads. In addition, existing heat dissipation designs are usually considered in the late stages of product development, which makes the space for optimizing heat dissipation performance very limited and difficult to achieve ideal heat dissipation effects. Therefore, a more intelligent, targeted and real-time adjustable heat dissipation control strategy is needed to meet the growing demand.

[0004] In order to solve the above problems, researchers proposed a method for collecting and analyzing three-dimensional temperature field data based on a distributed thermistor array. This method can obtain detailed temperature distribution inside the FPC sensor module. Combined with advanced heat flow path calculation and topology optimization algorithms, it can not only determine the most effective heat dissipation path, but also dynamically adjust the heat dissipation structure according to actual operating conditions, thereby improving heat dissipation efficiency and reducing energy consumption. Furthermore, the branch and bound algorithm is used to optimize the heat dissipation structure layout solution, which can reduce unnecessary resource waste while ensuring system performance. Summary of the invention

[0005] The main purpose of the present invention is to provide a heat dissipation control method and device for a high-level computing power autonomous driving FPC sensor module, which solves the problem that traditional methods are often unable to accurately and effectively cool local hot spots and lack the flexibility to adapt to dynamic thermal changes under different workloads.

[0006] To achieve the above object, the present invention provides a heat dissipation control method for a high-order computing power autonomous driving FPC sensor module, comprising the following steps:

[0007] The temperature data of the target FPC sensor module is collected through a preset distributed thermal array to obtain three-dimensional temperature field data;

[0008] Calculating the heat flow path of the target FPC sensor module based on the three-dimensional temperature field data to obtain a heat flux density distribution diagram;

[0009] Based on the heat flux density distribution diagram, a topology optimization calculation is performed on the target FPC sensor module to obtain a heat dissipation structure layout solution;

[0010] The heat dissipation structure layout scheme is dynamically optimized and solved by the branch and bound algorithm to obtain the heat dissipation control parameter set;

[0011] The heat dissipation control parameter set is input into a preset execution unit, and the target FPC sensor module is subjected to real-time heat dissipation control through the execution unit to obtain an ideal heat dissipation state.

[0012] Furthermore, the temperature data of the target FPC sensor module is collected by a preset distributed thermal array to obtain three-dimensional temperature field data, including:

[0013] The target FPC sensor module is subjected to high-frequency scanning sampling by a preset distributed thermal array to obtain an original temperature signal sequence;

[0014] Filtering the original temperature signal sequence to obtain a denoised temperature data stream;

[0015] Performing Hilbert-Huang transform processing based on the de-noised temperature data stream to obtain instantaneous temperature characteristics;

[0016] The instantaneous temperature characteristics are spatially reconstructed by an adaptive grid subdivision algorithm to obtain a temperature field grid matrix; wherein the temperature field grid matrix includes grid node coordinates and temperature data corresponding to the grid node coordinates;

[0017] Performing Kriging interpolation operation on the temperature field grid matrix to obtain a continuous temperature field distribution;

[0018] The continuous temperature field distribution is gradient calculated to obtain a temperature field gradient vector, and three-dimensional temperature field data is obtained based on the temperature field gradient vector and the temperature field grid matrix; wherein the three-dimensional temperature field data includes the temperature value of the hot spot area, the temperature fluctuation frequency and the temperature gradient direction.

[0019] Furthermore, the heat flow path calculation of the target FPC sensor module based on the three-dimensional temperature field data to obtain a heat flux density distribution diagram includes:

[0020] Numerically discretizing the three-dimensional temperature field data by a finite difference method to obtain temperature gradient field data;

[0021] Performing thermal conductivity analysis on the target FPC sensor module based on the temperature gradient field data to obtain thermal conductivity spatial distribution characteristics;

[0022] By using the variational method, based on the thermal conductivity spatial distribution characteristics, the target FPC sensor module is subjected to heat flow path analysis to obtain a heat flow channel topology structure, and the heat flow channel topology structure is subjected to connectivity analysis to obtain a heat flow transmission network;

[0023] Performing spectral analysis on the thermal impedance in the target FPC sensor module based on the heat flow transmission network to obtain impedance characteristic distribution, and extracting key heat flow paths of the heat flow transmission network based on the impedance characteristic distribution;

[0024] The thermal field intensity of the key heat flow path is calculated by the fast multipole method to obtain the heat flow field intensity data, and the gradient of the heat flow field intensity data is tracked to obtain the heat flow convergence area;

[0025] Gaussian curvature analysis is performed on the heat flux density in the heat flux convergence area to obtain a density distribution characteristic map, and heat flux density calculation is performed on the density distribution characteristic map to obtain a heat flux density distribution map.

[0026] Furthermore, the target FPC sensor module is subjected to topological optimization calculation based on the heat flux density distribution diagram to obtain a heat dissipation structure layout scheme, including:

[0027] Performing multi-scale decomposition on the heat flux density distribution map to obtain a heat flux characteristic spectrum, and performing wavelet transformation on the heat flux characteristic spectrum to obtain heat flux spectrum characteristics; wherein the heat flux spectrum characteristics include heat flux frequency, heat flux amplitude and heat flux phase;

[0028] Performing a heat flow network analysis on the target FPC sensor module through the heat flow spectrum characteristics to obtain a heat flow network topology structure, and performing community discovery on the heat flow network topology structure to obtain a heat flow community distribution feature; wherein the heat flow community distribution feature includes the number of heat flow communities, the scale of heat flow communities, and the connectivity of heat flow communities;

[0029] Based on the heat flow community distribution characteristics, a heat flow optimization calculation is performed on the target FPC sensor module to obtain a heat flow optimization parameter;

[0030] Based on the heat flow optimization parameters, a topological structure of the target FPC sensor module is generated to obtain a heat dissipation structure topological map, and a finite element analysis is performed on the heat dissipation structure topological map to obtain heat dissipation structure performance parameters; wherein the heat dissipation structure performance parameters include heat dissipation structure temperature, heat dissipation structure stress data and heat dissipation structure deformation data;

[0031] Based on the heat dissipation structure performance parameters and the heat dissipation structure topology diagram, a heat dissipation structure layout calculation is performed on the target FPC sensor module to obtain a heat dissipation structure layout solution.

[0032] Furthermore, the heat dissipation structure layout scheme is dynamically optimized and solved by the branch and bound algorithm to obtain a heat dissipation control parameter set, including:

[0033] An initial state evaluation is performed on the heat dissipation structure layout scheme by a branch and bound algorithm to obtain an evaluation result, and fuzzy logic processing is performed on the evaluation result to obtain an optimization direction index; wherein the optimization direction index includes an optimization path, an optimization weight and an optimization target of the heat dissipation structure;

[0034] Based on the optimization direction index, a multi-objective optimization solution is performed on the heat dissipation structure layout scheme to obtain a preliminary optimization scheme, and a constraint condition analysis is performed on the preliminary optimization scheme to obtain an optimization constraint set; wherein the optimization constraint set includes constraints on structural strength, material heat resistance and heat dissipation efficiency;

[0035] Performing genetic recombination and mutation operations on the preliminary optimization scheme through a genetic algorithm to obtain a variant optimization scheme, and performing Bayesian optimization on the variant optimization scheme to obtain a Bayesian optimization scheme; wherein the Bayesian optimization scheme includes improvement suggestions and parameter adjustments for the heat dissipation structure in the target FPC sensor module;

[0036] Based on the Bayesian optimization scheme, topology optimization is performed on the heat dissipation structure layout scheme to obtain a topology optimization scheme, and structural modal analysis is performed on the topology optimization scheme to obtain a modal response curve; wherein the modal response curve includes the vibration response of the heat dissipation structure at different frequencies;

[0037] Performing random perturbation test on the modal response curve through Monte Carlo simulation to obtain the thermal stress distribution of the heat dissipation structure, and performing finite element analysis on the thermal stress distribution to obtain a thermal stress field intensity distribution diagram; wherein the thermal stress field intensity distribution diagram is used to reflect the thermal stress change of the heat dissipation structure under different loads;

[0038] Based on the thermal stress field intensity distribution diagram, a thermal-mechanical coupling analysis is performed on the heat dissipation structure to obtain a thermal-mechanical coupling parameter set of the heat dissipation structure, and a multi-physical field simulation is performed on the thermal-mechanical coupling parameter set to obtain a heat dissipation control parameter set; wherein the heat dissipation control parameter set includes temperature control parameters, fan speed control parameters and coolant flow control parameters of the heat dissipation structure.

[0039] Furthermore, the heat-mechanical coupling analysis of the heat dissipation structure is performed based on the thermal stress field intensity distribution diagram to obtain a heat-mechanical coupling parameter set of the heat dissipation structure, including:

[0040] Performing Helmholtz decomposition on the thermal stress field intensity distribution diagram to obtain a thermal-force field potential function, and performing variational principle derivation on the thermal-force field potential function to obtain a coupling control equation;

[0041] Discretizing the coupled control equations by a finite volume method to obtain thermal-mechanical grid units;

[0042] Solving the Navier-Stokes equations for the thermal-mechanical grid unit to obtain flow field distribution characteristics, and performing Rayleigh number analysis on the flow field distribution characteristics to obtain a convective heat transfer coefficient;

[0043] Performing a Boltzmann transport equation calculation based on the convective heat transfer coefficient to obtain a heat conduction flux, and performing a Maxwell stress tensor analysis based on the heat conduction flux to obtain a stress distribution field;

[0044] Performing thermal-mechanical coupling iterative calculation on the stress distribution field by nonlinear finite element method to obtain a coupled iterative sequence, and performing Richardson extrapolation processing on the coupled iterative sequence to obtain a converged solution set;

[0045] Based on the converged solution set, thermal-mechanical coupling parameters of the heat dissipation structure are extracted to obtain a thermal-mechanical coupling parameter set of the heat dissipation structure.

[0046] Furthermore, the heat dissipation control parameter set is input into a preset execution unit, and the target FPC sensor module is subjected to real-time heat dissipation control by the execution unit to obtain an ideal heat dissipation state, including:

[0047] Parsing the heat dissipation control parameter set to obtain a control parameter sequence;

[0048] Adaptively slice-code the control parameter sequence to obtain a parameter control instruction set;

[0049] The optimal trajectory planning of the parameter control instruction set is carried out through the preset Hamilton-Jacobi equation to obtain the control trajectory sequence, and the control trajectory sequence is subjected to Lyapunov stability analysis to obtain the control trajectory sequence in the steady-state control interval;

[0050] Inputting the control trajectory sequence of the steady-state control interval into a preset execution unit, and performing nonlinear robust control on the execution unit to obtain an execution response curve;

[0051] Performing wavelet packet decomposition processing on the execution response curve to obtain multi-level response characteristics; wherein the multi-level response characteristics include temperature dynamic characteristics, fan speed characteristics and coolant flow characteristics;

[0052] Based on the multi-level response characteristics, real-time heat dissipation control is performed on the target FPC sensor module to obtain an ideal heat dissipation state.

[0053] The present invention also provides a heat dissipation control device for a high-order computing power autonomous driving FPC sensor module, comprising:

[0054] The acquisition module is used to collect temperature data of the target FPC sensor module through a preset distributed thermal array to obtain three-dimensional temperature field data;

[0055] A calculation module, used to calculate the heat flow path of the target FPC sensor module based on the three-dimensional temperature field data to obtain a heat flux density distribution diagram;

[0056] An optimization module, used to perform topology optimization calculation on the target FPC sensor module based on the heat flux density distribution diagram to obtain a heat dissipation structure layout solution;

[0057] A solution module, used for dynamically optimizing and solving the heat dissipation structure layout scheme through a branch and bound algorithm to obtain a heat dissipation control parameter set;

[0058] The control module is used to input the heat dissipation control parameter set into a preset execution unit, and perform real-time heat dissipation control on the target FPC sensor module through the execution unit to obtain an ideal heat dissipation state.

[0059] The present invention also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the above methods when executing the computer program.

[0060] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of any of the above-mentioned methods are implemented.

[0061] The heat dissipation control method of the high-order computing power autonomous driving FPC sensor module provided by the present invention comprises the following steps: collecting temperature data of the target FPC sensor module through a preset distributed thermal array to obtain three-dimensional temperature field data; calculating the heat flow path of the target FPC sensor module based on the three-dimensional temperature field data to obtain a heat flux density distribution map; performing topological optimization calculation on the target FPC sensor module based on the heat flux density distribution map to obtain a heat dissipation structure layout scheme; dynamically optimizing and solving the heat dissipation structure layout scheme through a branch and bound algorithm to obtain a heat dissipation control parameter set; inputting the heat dissipation control parameter set into a preset execution unit, and performing real-time heat dissipation control on the target FPC sensor module to obtain an ideal heat dissipation state, solving the problem that traditional methods often cannot accurately and effectively cool down local hot spots, and lack flexibility to adapt to dynamic thermal changes under different workloads, and realizing the use of the heat flux density distribution map as input to perform topological optimization calculation on the FPC sensor module to determine the optimal heat dissipation structure layout scheme. This method not only takes into account space utilization, but also takes into account thermal performance, and can achieve the best heat dissipation effect in a limited space. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 It is a schematic diagram of the steps of a heat dissipation control method of a high-order computing power autonomous driving FPC sensor module in one embodiment of the present invention;

[0063] Figure 2 It is a structural block diagram of a heat dissipation control system of a high-order computing power autonomous driving FPC sensor module in one embodiment of the present invention;

[0064] Figure 3 It is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention.

[0065] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings in conjunction with the embodiments. DETAILED DESCRIPTION

[0066] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0067] like Figure 1 As shown, Figure 1 This is a schematic diagram of the steps of a heat dissipation control method for a high-order computing power autonomous driving FPC sensor module in one embodiment of the present invention;

[0068] In one embodiment of the present invention, a heat dissipation control method for a high-order computing power autonomous driving FPC sensor module is provided, comprising the following steps:

[0069] Step S1, collecting temperature data of a target FPC sensor module through a preset distributed thermal array to obtain three-dimensional temperature field data.

[0070] Specifically, in order to achieve the step of "collecting temperature data of the target FPC sensor module through a preset distributed thermal array to obtain three-dimensional temperature field data" described above, it is first necessary to deploy multiple thermal sensors at key positions of the FPC (flexible printed circuit) sensor module. These sensors constitute a preset distributed thermal array. The design and installation of the array are carefully planned based on the internal structure of the module and the expected heat distribution to ensure that all areas where high temperatures may be generated can be fully covered, thereby accurately reflecting the temperature conditions of the entire module. When the autonomous driving vehicle is running, the FPC sensor module will generate a lot of heat due to processing high-order computing tasks. At this time, the thermal sensors distributed in the module will start working. They can monitor the temperature changes at their respective positions in real time and convert this information into electrical signals and send them to the data acquisition system. As the vehicle continues to drive, each sensor point will continuously provide temperature readings. After these readings are sorted and analyzed by the data acquisition system, a set of temperature values ​​representing different time points and different spatial positions is formed, which is the so-called three-dimensional temperature field data. For example, in a car that is automatically navigating on a city road, its FPC sensor module is used to process data from sensors such as cameras and radars to support complex driving decisions. Due to heavy computing tasks, certain specific areas inside the module may experience local overheating. Through the pre-arranged distributed thermistor array, engineers can accurately capture the location and intensity of these hot spots and draw a detailed temperature distribution map. This picture not only shows the temperature difference on the surface of the module, but also deeply reveals the heat transfer between the internal layers, providing a solid foundation for subsequent heat flow path calculations. This approach allows designers to optimize the heat dissipation solution without affecting performance, ensuring the stability and safety of the autonomous driving system.

[0071] Step S2, performing heat flow path calculation on the target FPC sensor module based on the three-dimensional temperature field data to obtain a heat flux density distribution diagram.

[0072] Specifically, when implementing the process of "calculating the heat flow path of the target FPC sensor module based on the three-dimensional temperature field data to obtain a heat flux density distribution map" described above, the three-dimensional temperature field data obtained in the previous step must first be used as a basis. These data contain temperature information at different positions and time points inside the FPC sensor module, which is the key to understanding how heat is distributed within the module. Once accurate three-dimensional temperature field data is available, it is necessary to analyze these data through specially designed algorithms to determine the specific path for heat to be transferred from the heat source to the surrounding environment. In this process, engineers will use physical principles such as heat conduction, convection, and radiation to establish mathematical models, combined with actual material properties and geometric structures, to simulate the flow of heat within the module. Through such simulations, it is possible to predict which areas will become the main channels for heat and where thermal resistance may be formed, thereby affecting the heat dissipation efficiency. In order to make this simulation closer to the actual situation, the influence of external cooling conditions such as air cooling or liquid cooling, as well as the dynamic change factors generated under the working state of the module, will also be considered. For example, in the high-level computing power FPC sensor module used in autonomous vehicles, when it processes data from cameras, radars and other sensors, certain components such as processors or memory chips may generate a lot of heat. Through the pre-built three-dimensional temperature field data, we can know the location and intensity of these hot spots. Then, using the heat flow path calculation method, we can further analyze how the heat diffuses from these high-temperature areas to low-temperature areas. In the end, we can not only draw a detailed heat flux density distribution map, which shows the amount of heat passing through per unit area, but also identify potential heat dissipation bottlenecks, providing a basis for subsequent optimization of the heat dissipation structure. In this way, it is possible to ensure that the module can operate within an ideal temperature range under various working conditions without affecting its performance, thereby ensuring the stability and reliability of the autonomous driving system.

[0073] Step S3, performing topology optimization calculation on the target FPC sensor module based on the heat flux density distribution diagram to obtain a heat dissipation structure layout solution.

[0074] Specifically, when implementing the key step of "performing topological optimization calculations on the target FPC sensor module based on the heat flux density distribution map to obtain a heat dissipation structure layout plan" mentioned above, the heat flux density distribution map obtained previously is first relied on. This figure intuitively shows the flow path and intensity of heat inside the FPC sensor module, providing a basis for the subsequent optimization work. By analyzing this figure, engineers can identify which areas are the main heat sources, which paths carry the most heat transfer, and which places may become heat dissipation bottlenecks. Once this information is clear, the next step is to input the data of the heat flux density distribution map into a specially designed topological optimization algorithm. This algorithm comprehensively considers factors such as material properties, geometry, space constraints, and the expected working environment to determine the most effective heat dissipation structure layout plan. In this process, the topological optimization calculation is not only to find one or several optimal heat dissipation paths, but also to comprehensively review the spatial layout of the entire FPC module to ensure that the position and shape of each component are carefully planned to maximize the heat dissipation efficiency while minimizing the impact on the original circuit function. For example, in the high-level computing power FPC sensor module used in autonomous vehicles, due to its compact design and high performance requirements, it is particularly important to arrange the heat dissipation structure reasonably in a limited space. For example, when processing massive data from sensors such as cameras and radars, the processor and other core components in the FPC module will generate a lot of heat. Through topology optimization calculations, we can adjust the position and size of the heat sink fins, increase the application of thermal pads, and even redesign the layout of some non-critical electronic components without affecting the overall performance of the module, thereby improving the speed and effect of heat conduction from high temperature areas to low temperature areas. Such optimization not only helps to maintain the system operating within an ideal temperature range, but also extends the service life of the equipment and improves the reliability and safety of the system. In addition, the optimized layout can also reduce unnecessary energy loss, reduce the burden on the cooling system, and thus improve the energy efficiency of the entire autonomous driving platform. In summary, through in-depth analysis of the heat flux density distribution map and implementation of topology optimization calculations, we can tailor an efficient heat dissipation solution for the FPC sensor module to ensure that it works stably and reliably in complex and changing practical application scenarios.

[0075] Step S4, dynamically optimizing and solving the heat dissipation structure layout solution by using a branch and bound algorithm to obtain a heat dissipation control parameter set.

[0076] Specifically, in order to achieve the step of "dynamically optimizing and solving the heat dissipation structure layout scheme through the branch and bound algorithm to obtain the heat dissipation control parameter set" described above, it is first necessary to calculate the heat dissipation structure layout scheme based on the previous topology optimization. At this point, we already have a set of theoretically optimal heat dissipation designs, but this is only an ideal model under static conditions. However, in practical applications, the working environment and load conditions of the FPC sensor module are constantly changing, so an optimization method that can adapt to these dynamic changes is needed to ensure that the heat dissipation performance is always kept in the best state. The branch and bound algorithm plays an important role in this case. It is an effective tool for solving complex optimization problems, which can efficiently search the possible solution space while ensuring the global optimal solution. In this process, the algorithm sets a series of constraints and objective functions, such as maximum allowable temperature, minimized energy consumption, etc., according to the current heat flux density distribution map and expected working conditions. Then, by evaluating different configurations of the heat dissipation structure layout scheme, the space of feasible solutions is gradually narrowed, and finally the optimal or near-optimal heat dissipation control parameter set is locked. For example, the workload of the high-order computing power FPC sensor module used in autonomous driving vehicles fluctuates with the changes in the vehicle driving environment. When the vehicle enters a busy urban road and the amount of sensor data processing surges, some areas inside the FPC module may heat up rapidly. At this time, the branch and bound algorithm can respond to this change in real time and quickly adjust the heat dissipation strategy. For example, it can reallocate cooling resources based on the latest three-dimensional temperature field data, dynamically adjust the fan speed or the flow of the liquid cooling system, and even change the angle and position of the heat sink fins to ensure that key components are always within a safe operating temperature range. At the same time, the algorithm will also consider the cost-effectiveness of long-term operation to avoid unnecessary energy consumption caused by excessive use of cooling equipment. In addition, the branch and bound algorithm also supports predictive optimization of future working conditions. Using machine learning technology, combined with historical data and real-time monitoring information, the heat dissipation path can be planned in advance to prevent potential overheating risks. For example, before the vehicle is expected to enter a high-temperature section, the system can increase the cooling intensity in advance to prepare for the upcoming high-intensity computing. In this way, even in the face of complex and changeable actual application scenarios, it can maintain an ideal heat dissipation state through continuous dynamic optimization and solution, and ensure the stability and reliability of the autonomous driving system. In summary, the dynamic optimization and solution of the heat dissipation structure layout scheme through the branch and bound algorithm not only improves the heat dissipation efficiency, but also enhances the system's adaptability, providing a solid thermal management foundation for high-level computing power autonomous driving platforms.

[0077] Step S5, inputting the heat dissipation control parameter set into a preset execution unit, and performing real-time heat dissipation control on the target FPC sensor module through the execution unit to obtain an ideal heat dissipation state.

[0078] Specifically, in order to achieve the above-described "inputting the heat dissipation control parameter set into the preset execution unit, and performing real-time heat dissipation control on the target FPC sensor module to obtain an ideal heat dissipation state", the entire process is closely centered on the heat dissipation control parameter set previously obtained through the branch and bound algorithm optimization solution. This parameter set contains all necessary instructions and set values, such as fan speed, flow regulation of the liquid cooling system, angle adjustment of the heat dissipation fins, etc., all of which are to ensure that the FPC sensor module can maintain an ideal temperature range under different working conditions. Once these optimized and calculated heat dissipation control parameters are obtained, the next step is to integrate them into the preset execution unit. The execution unit here can include various types of cooling equipment and their control systems, such as electric fans, pumps, valves and other hardware facilities, as well as corresponding software control logic. When these parameters are loaded into the execution unit, they become actual commands to guide the operation of the system, so that each execution element can respond accurately according to the current working requirements to achieve the best heat dissipation effect. For example, the high-order computing power FPC sensor module used in autonomous driving vehicles, its internal processor and other key components will generate a lot of heat when processing data from sensors such as cameras and radars. At this time, by dynamically adjusting the speed of the fan or the flow of the liquid cooling system, it is possible to respond quickly according to the temperature changes monitored in real time, thereby effectively managing heat accumulation and preventing overheating. Specifically, assuming that the vehicle is driving on a city road, as the traffic conditions change, the load of the FPC sensor module also fluctuates. When it is detected that the temperature rise in a certain area is close to the critical value, the system will immediately automatically increase the speed of the fan near the area or increase the fluid flow in the liquid cooling pipe according to the preset rules in the heat dissipation control parameter set to accelerate the dissipation of heat. At the same time, if the vehicle is expected to enter a long period of high-speed driving, the system can also take measures in advance, such as improving the cooling efficiency in advance, to prepare for the upcoming high-intensity computing. In addition, for some non-continuous high-temperature points, such as local hot spots, the system can improve the air flow path and enhance the heat dissipation effect by fine-tuning the angle of the heat sink fins. Such real-time regulation can not only quickly respond to sudden temperature rise problems, but also maintain a stable temperature environment during long-term operation, extend the service life of electronic components, and ensure the reliability and safety of the autonomous driving system. Ultimately, through this closed-loop feedback mechanism, the ideal heat dissipation state control of the FPC sensor module is achieved, ensuring that efficient and safe operating conditions can always be maintained even in complex and changing working environments.

[0079] In a specific embodiment, the temperature data of the target FPC sensor module is collected by a preset distributed thermal array to obtain three-dimensional temperature field data, including:

[0080] The target FPC sensor module is subjected to high-frequency scanning sampling by a preset distributed thermal array to obtain an original temperature signal sequence;

[0081] Filtering the original temperature signal sequence to obtain a denoised temperature data stream;

[0082] Performing Hilbert-Huang transform processing based on the de-noised temperature data stream to obtain instantaneous temperature characteristics;

[0083] The instantaneous temperature characteristics are spatially reconstructed by an adaptive grid subdivision algorithm to obtain a temperature field grid matrix; wherein the temperature field grid matrix includes grid node coordinates and temperature data corresponding to the grid node coordinates;

[0084] Performing Kriging interpolation operation on the temperature field grid matrix to obtain a continuous temperature field distribution;

[0085] The continuous temperature field distribution is gradient calculated to obtain a temperature field gradient vector, and three-dimensional temperature field data is obtained based on the temperature field gradient vector and the temperature field grid matrix; wherein the three-dimensional temperature field data includes the temperature value of the hot spot area, the temperature fluctuation frequency and the temperature gradient direction.

[0086] Specifically, in order to achieve the above-described "high-frequency scanning and sampling of the target FPC sensor module by a preset distributed thermal array to obtain the original temperature signal sequence", this process involves a series of closely connected technical steps. First, the method relies on a distributed thermal array pre-arranged at key positions inside the FPC (flexible printed circuit) sensor module. These sensors can not only accurately monitor the temperature changes at their respective positions, but also work in a high-frequency scanning sampling mode to ensure that every tiny temperature fluctuation is captured, thereby generating a raw temperature signal sequence. This high-frequency scanning sampling is essential for real-time tracking of the complex temperature dynamics inside the FPC module, especially in high-order computing power autonomous driving systems, which have heavy processing tasks and rapid and frequent temperature changes. Once the raw temperature signal sequence is obtained, it is then necessary to filter it to remove inaccurate information introduced by external interference or noise from the sensor itself, thereby obtaining a denoised temperature data stream. The selection and application of filtering technology needs to take into account various factors in the actual application scenario, such as the response characteristics of the sensor, environmental conditions, etc., to ensure that the final data is as close to the actual situation as possible. For example, during the driving of an autonomous vehicle, abnormal fluctuations in sensor readings may occur due to factors such as vehicle vibration and electromagnetic interference. Through a carefully designed filtering algorithm, these interferences can be effectively eliminated and the true temperature change trend can be retained. Based on the denoised temperature data stream, Hilbert-Huang transform (HHT) processing is further used to obtain instantaneous temperature characteristics. HHT is a powerful time-frequency analysis tool, especially suitable for the processing of non-stationary signals. It can decompose complex temperature changes into a series of intrinsic mode functions (IMFs) and extract instantaneous frequency and amplitude information from them. This is very useful for understanding the change law of temperature over time and space. In the scenario of autonomous driving, when the FPC sensor module processes large-scale data from sensors such as cameras and radars, some areas may heat up rapidly due to a sudden increase in computing load. Through HHT processing, these instantaneous temperature characteristics can be accurately captured, providing an important basis for subsequent spatial reconstruction. Subsequently, the instantaneous temperature characteristics are spatially reconstructed using an adaptive mesh subdivision algorithm to construct a temperature field grid matrix. This matrix contains the coordinates of each grid node and its corresponding temperature data, forming a discretized temperature distribution model. The adaptive mesh subdivision algorithm automatically adjusts the grid density according to the change of temperature gradient to ensure that there is enough resolution in the hot spot area to reflect subtle temperature differences. For example, near the core processing unit of the FPC module, the temperature change may be more drastic, so the algorithm will generate a finer grid structure here to improve the measurement accuracy. At the same time, in areas with more stable temperatures, a coarser grid can be used to save computing resources. After having the temperature field grid matrix, the next step is to perform Kriging interpolation on the matrix to obtain a continuous temperature field distribution.Kriging interpolation is a geostatistical method that predicts the value of unknown points based on the data of known points and takes into account spatial correlation. This method can not only fill the gaps between grids, but also smooth the temperature changes in the transition area, making the entire temperature field look more natural and coherent. This step is very important for a comprehensive understanding of the heat transfer path inside the FPC module because it provides a complete perspective from the local to the whole. In autonomous driving applications, the continuous temperature field distribution obtained by Kriging interpolation can help engineers identify those temperature difference areas that are not easy to detect but potentially affect performance, and then optimize the heat dissipation design. Next, the gradient of the continuous temperature field distribution is calculated to obtain the temperature field gradient vector. The temperature field gradient vector describes the rate and direction of temperature change in space, which is critical for understanding how heat flows and determining the main heat flow path. Based on the temperature field gradient vector and the temperature field grid matrix, we can construct three-dimensional temperature field data, which includes not only the temperature value of the hot spot area, but also covers information such as the frequency of temperature fluctuations and the direction of the temperature gradient. For example, during the operation of an autonomous vehicle, certain operating modes may cause temperature fluctuations of a specific frequency, and this information is instructive for formulating an effective heat dissipation strategy. In this way, we can not only see where the main heat sources are, but also understand how these hot spots evolve over time, thereby providing solid data support for dynamically adjusting the heat dissipation control parameter set. In summary, the above steps together constitute a complete temperature data acquisition and analysis process. From the initial high-frequency scanning sampling to the final three-dimensional temperature field data generation, each link is closely connected and linked together. Through such a set of refined methods, we can achieve accurate monitoring of temperature status in high-order computing power autonomous driving FPC sensor modules, ensuring that even in the face of complex and changing working environments, we can maintain ideal heat dissipation effects and ensure the stability and reliability of the system. In addition, this detailed and in-depth temperature analysis method not only helps prevent overheating problems, but also provides a valuable data basis for long-term maintenance and optimization, further improving the safety and efficiency of the autonomous driving platform.

[0087] In a specific embodiment, the heat flow path calculation of the target FPC sensor module based on the three-dimensional temperature field data to obtain a heat flux density distribution diagram includes:

[0088] Numerically discretizing the three-dimensional temperature field data by a finite difference method to obtain temperature gradient field data;

[0089] Performing thermal conductivity analysis on the target FPC sensor module based on the temperature gradient field data to obtain thermal conductivity spatial distribution characteristics;

[0090] By using the variational method, based on the thermal conductivity spatial distribution characteristics, the target FPC sensor module is subjected to heat flow path analysis to obtain a heat flow channel topology structure, and the heat flow channel topology structure is subjected to connectivity analysis to obtain a heat flow transmission network;

[0091] Performing spectral analysis on the thermal impedance in the target FPC sensor module based on the heat flow transmission network to obtain impedance characteristic distribution, and extracting key heat flow paths of the heat flow transmission network based on the impedance characteristic distribution;

[0092] The thermal field intensity of the key heat flow path is calculated by the fast multipole method to obtain the heat flow field intensity data, and the gradient of the heat flow field intensity data is tracked to obtain the heat flow convergence area;

[0093] Gaussian curvature analysis is performed on the heat flux density in the heat flux convergence area to obtain a density distribution characteristic map, and heat flux density calculation is performed on the density distribution characteristic map to obtain a heat flux density distribution map.

[0094] Specifically, in order to achieve the above-described "calculating the heat flow path of the target FPC sensor module based on the three-dimensional temperature field data to obtain a heat flux density distribution map", this process involves a series of complex and interrelated technical steps. First, the method starts from the three-dimensional temperature field data and numerically discretizes it by the finite difference method to obtain temperature gradient field data. The finite difference method is a commonly used numerical solution method that discretizes the continuous temperature field into a series of discrete points. The temperature value at each point can be approximated by the difference between adjacent points to represent the temperature gradient. This method can effectively capture the trend of temperature variation with space, especially in areas where the temperature changes drastically, such as near the core processing unit of the high-order computing power FPC sensor module in the autonomous driving vehicle, which is prone to local high temperature due to the large amount of calculation. Once the temperature gradient field data is obtained, the next step is to perform thermal conductivity analysis on the target FPC sensor module based on these data to determine the spatial distribution characteristics of thermal conductivity. Thermal conductivity is an indicator of the ability of a material to conduct heat, and materials in different locations may have different thermal conductivity values. Through in-depth analysis of the temperature gradient field data, it is possible to identify which materials or areas have higher thermal conductivity and which are relatively low. For example, inside the FPC module, some metal layers may exhibit high thermal conductivity, while the thermal conductivity of the insulating material is much lower. Through precise measurement and simulation, we are able to construct a detailed spatial distribution feature map, which provides the basis for subsequent heat flow path calculations. With the thermal conductivity spatial distribution characteristics, the next step is to use variational techniques to perform heat flow path analysis on the target FPC sensor module based on these characteristics to determine the heat flow channel topology, and perform connectivity analysis on this structure to finally obtain the heat flow transmission network. Variational method is a mathematical optimization method that finds the optimal state of the system under given conditions by minimizing or maximizing a specific energy function (i.e., energy functional). In this case, the energy functional is used to describe the heat transfer process within the module, including parameters such as thermal resistance and heat capacity of each path. By solving this functional, we can identify the most effective heat flow channel to ensure that heat can flow along the path with the least resistance. At the same time, connectivity analysis helps us understand how these channels are interconnected to form a complete heat flow transmission network, which is crucial for predicting how heat spreads throughout the module. Next, based on the obtained heat transfer network, the thermal impedance in the target FPC sensor module is spectrally analyzed to obtain the impedance characteristic distribution, and the key heat flow paths in the heat transfer network are further extracted based on these characteristics. Thermal impedance reflects the degree of resistance encountered during heat transfer, which is usually related to factors such as the thermal conductivity and thickness of the material. Through spectral analysis, we can decompose complex thermal impedance into different frequency components, thereby revealing which parts have the greatest impact on heat transfer.For example, in an autonomous driving scenario, when the processor in the FPC module runs at high load for a long time, certain specific paths may become particularly important due to the accumulated heat. By identifying these critical paths, we can focus resources on optimizing their heat dissipation performance and improve the reliability of the entire system. Then, for the extracted critical heat flow paths, the fast multipole method is used to calculate the thermal field strength to obtain the thermal flow field strength data, and the gradient of these data is tracked to locate the heat flow convergence area. The fast multipole method is an efficient numerical calculation technique that can significantly reduce the calculation time while maintaining high accuracy. By carefully calculating the thermal flow field strength on the critical path, we can determine where the most heat is gathered. Gradient tracking helps us find the location of these hot spots along the direction of the temperature gradient, which is very important for formulating targeted heat dissipation strategies. For example, during the driving of an autonomous vehicle, if it is found that the heat flow in a certain area is abnormally concentrated, additional measures can be taken, such as adding cooling fins or adjusting the coolant flow, to alleviate the overheating problem. Finally, in order to generate the final heat flux density distribution map, it is necessary to perform Gaussian curvature analysis on the heat flux density in the heat flux convergence area to obtain the density distribution feature map, and calculate the heat flux density for this feature map. Gaussian curvature analysis can help us quantify the changing trend of heat flux density, especially those nonlinear change patterns. In this way, we can more accurately depict the distribution of heat in the module, including those subtle differences that are not easy to detect but potentially affect performance. For example, when an autonomous driving vehicle enters a busy urban road section and the sensor data processing volume surges, some areas may experience regular temperature rise. Through the above method, we can not only intuitively see the location of these hot spots, but also understand their frequency characteristics over time, providing a scientific basis for dynamically adjusting heat dissipation control. In summary, the above steps together constitute a complete process from three-dimensional temperature field data analysis to heat flow path calculation, and then to the generation of heat flux density distribution map. Each step is closely linked and interlocking, ensuring a comprehensive understanding and precise control of the complex thermal behavior inside the high-order computing power FPC sensor module. Through such a systematic analysis method, even in the face of complex and changing working environments, it can provide reliable thermal management support for the autonomous driving platform to ensure its stability and safety. In addition, this method can also provide a valuable data foundation for long-term maintenance and optimization, further improving the safety and efficiency of the autonomous driving platform.

[0095] In a specific embodiment, the topology optimization calculation of the target FPC sensor module is performed based on the heat flux density distribution diagram to obtain a heat dissipation structure layout solution, including:

[0096] Performing multi-scale decomposition on the heat flux density distribution map to obtain a heat flux characteristic spectrum, and performing wavelet transformation on the heat flux characteristic spectrum to obtain heat flux spectrum characteristics; wherein the heat flux spectrum characteristics include heat flux frequency, heat flux amplitude and heat flux phase;

[0097] Performing a heat flow network analysis on the target FPC sensor module through the heat flow spectrum characteristics to obtain a heat flow network topology structure, and performing community discovery on the heat flow network topology structure to obtain a heat flow community distribution feature; wherein the heat flow community distribution feature includes the number of heat flow communities, the scale of heat flow communities, and the connectivity of heat flow communities;

[0098] Based on the heat flow community distribution characteristics, a heat flow optimization calculation is performed on the target FPC sensor module to obtain a heat flow optimization parameter;

[0099] Based on the heat flow optimization parameters, a topological structure of the target FPC sensor module is generated to obtain a heat dissipation structure topological map, and a finite element analysis is performed on the heat dissipation structure topological map to obtain heat dissipation structure performance parameters; wherein the heat dissipation structure performance parameters include heat dissipation structure temperature, heat dissipation structure stress data and heat dissipation structure deformation data;

[0100] Based on the heat dissipation structure performance parameters and the heat dissipation structure topology diagram, a heat dissipation structure layout calculation is performed on the target FPC sensor module to obtain a heat dissipation structure layout solution.

[0101] Specifically, in order to achieve the above-described "topology optimization calculation of the target FPC sensor module based on the heat flux density distribution map to obtain the heat dissipation structure layout plan", this process requires a series of complex and delicate steps. First, starting from the heat flux density distribution map, multi-scale decomposition is performed to obtain the heat flux characteristic spectrum, and wavelet transform is further applied to extract the heat flux spectrum characteristics. These features include information such as heat flux frequency, heat flux amplitude and heat flux phase, which together constitute a comprehensive description of heat flow behavior. Multi-scale decomposition can reveal heat transfer patterns at different scales, while wavelet transform helps to identify specific heat flow dynamics in the time-frequency domain. For example, in the high-order computing power FPC sensor module used in autonomous driving vehicles, some key components such as processors generate a lot of heat when processing data from sensors such as cameras and radars. This heat is not only unevenly distributed in space, but also presents complex dynamic characteristics over time. Through multi-scale decomposition and wavelet transform, we can accurately capture these dynamic changes and provide a solid foundation for subsequent analysis. Next, the heat flow network analysis of the target FPC sensor module is performed using the obtained heat flow spectrum characteristics to determine the heat flow network topology, and community discovery is performed on this basis to obtain the heat flow community distribution characteristics. Heat flow network analysis is a method that regards heat flow paths as network nodes and edges, which can help us understand how heat propagates in the network. Community discovery refers to identifying those areas that are closely connected internally but relatively independent externally, namely the so-called heat flow communities. Each community may represent a local hotspot or an important heat dissipation path. Through in-depth analysis of the heat flow network, we can know the number, size and connectivity of heat flow communities. For example, in an FPC module, several heat flow communities may be formed around the core processing unit, each of which carries different heat loads and is connected to each other through specific paths. Understanding the distribution of these communities is crucial to optimizing heat dissipation design because it directly affects whether heat can be effectively transferred from high temperature areas to low temperature areas. With the heat flow community distribution characteristics, the next step is to perform heat flow optimization calculations on the target FPC sensor module based on these characteristics to obtain heat flow optimization parameters. This step aims to find the most effective heat dissipation strategy to ensure that heat can be quickly discharged along the path with the least resistance. Thermal flow optimization parameters may involve material selection, geometry adjustment, cooling system configuration and other aspects. For example, if a certain thermal flow community is large in size but has low connectivity, you can consider increasing the thermal conduction path in that area or improving the layout of the cooling equipment to improve heat dissipation efficiency. Through such optimization calculations, we can find a set of optimal parameter combinations so that the entire FPC module always remains within an ideal temperature range during operation. Then, based on the obtained thermal flow optimization parameters, the topology of the target FPC sensor module is generated to obtain a heat dissipation structure topology map, and the map is subjected to finite element analysis to evaluate the heat dissipation structure performance parameters.The heat dissipation structure topology diagram shows the optimized heat dissipation path and component arrangement, while finite element analysis is a numerical simulation method used to predict the response of the structure under various load conditions. In this way, we can understand the temperature distribution, stress data and deformation of the heat dissipation structure in detail. For example, in the application scenario of autonomous driving, when the FPC module is in a high-intensity computing state, finite element analysis can help us identify where there may be overheating risks and which areas are subjected to greater mechanical stress, so as to take measures in advance to prevent them. In addition, the air flow path can be improved and the heat dissipation effect can be enhanced by adjusting the angle, thickness and other parameters of the heat dissipation fins. Finally, the heat dissipation structure layout of the target FPC sensor module is calculated based on the heat dissipation structure performance parameters and the heat dissipation structure topology diagram, and the heat dissipation structure layout plan is finally obtained. The goal of this stage is to comprehensively consider all factors, including but not limited to heat flow optimization parameters, heat dissipation structure performance indicators, etc., to develop a complete set of heat dissipation design plans. For example, in actual operation, engineers can re-plan the position and number of heat dissipation fins, adjust the application of thermal pads, and even change the layout of some non-critical electronic components based on the results of finite element analysis to achieve the best heat dissipation effect. At the same time, it is also necessary to consider the cost-effectiveness of long-term operation to avoid unnecessary energy consumption caused by excessive use of cooling equipment. Through such a set of refined design processes, even in the face of complex and changing working environments, reliable thermal management support can be provided for high-order computing power FPC sensor modules to ensure their stability and reliability. In summary, the above steps together constitute a complete process from heat flux density distribution map analysis to heat dissipation structure layout solution generation. Each step is closely linked and interlocking, ensuring a comprehensive understanding and precise control of the complex thermal behavior inside the FPC sensor module. Through this systematic analysis and optimization method, we can significantly improve the heat dissipation efficiency and extend the service life of the equipment without affecting performance, providing strong technical guarantees for the safe and reliable operation of the autonomous driving platform.

[0102] In a specific embodiment, the heat dissipation structure layout scheme is dynamically optimized and solved by a branch and bound algorithm to obtain a heat dissipation control parameter set, including:

[0103] An initial state evaluation is performed on the heat dissipation structure layout scheme by a branch and bound algorithm to obtain an evaluation result, and fuzzy logic processing is performed on the evaluation result to obtain an optimization direction index; wherein the optimization direction index includes an optimization path, an optimization weight and an optimization target of the heat dissipation structure;

[0104] Based on the optimization direction index, a multi-objective optimization solution is performed on the heat dissipation structure layout scheme to obtain a preliminary optimization scheme, and a constraint condition analysis is performed on the preliminary optimization scheme to obtain an optimization constraint set; wherein the optimization constraint set includes constraints on structural strength, material heat resistance and heat dissipation efficiency;

[0105] Performing genetic recombination and mutation operations on the preliminary optimization scheme through a genetic algorithm to obtain a variant optimization scheme, and performing Bayesian optimization on the variant optimization scheme to obtain a Bayesian optimization scheme; wherein the Bayesian optimization scheme includes improvement suggestions and parameter adjustments for the heat dissipation structure in the target FPC sensor module;

[0106] Based on the Bayesian optimization scheme, topology optimization is performed on the heat dissipation structure layout scheme to obtain a topology optimization scheme, and structural modal analysis is performed on the topology optimization scheme to obtain a modal response curve; wherein the modal response curve includes the vibration response of the heat dissipation structure at different frequencies;

[0107] Performing random perturbation test on the modal response curve through Monte Carlo simulation to obtain the thermal stress distribution of the heat dissipation structure, and performing finite element analysis on the thermal stress distribution to obtain a thermal stress field intensity distribution diagram; wherein the thermal stress field intensity distribution diagram is used to reflect the thermal stress change of the heat dissipation structure under different loads;

[0108] Based on the thermal stress field intensity distribution diagram, a thermal-mechanical coupling analysis is performed on the heat dissipation structure to obtain a thermal-mechanical coupling parameter set of the heat dissipation structure, and a multi-physical field simulation is performed on the thermal-mechanical coupling parameter set to obtain a heat dissipation control parameter set; wherein the heat dissipation control parameter set includes temperature control parameters, fan speed control parameters and coolant flow control parameters of the heat dissipation structure.

[0109] Specifically, in order to achieve the above-described "dynamic optimization and solution of the heat dissipation structure layout scheme by the branch and bound algorithm to obtain the heat dissipation control parameter set", this process involves a series of complex and interrelated technical steps. First, the method starts with the initial state evaluation of the heat dissipation structure layout scheme, and uses the branch and bound algorithm to comprehensively analyze these schemes to obtain the evaluation results. The branch and bound algorithm is an effective tool for solving combinatorial optimization problems. It can systematically search the possible solution space and ensure that the global optimal solution is found. In the high-order computing power FPC sensor module used in autonomous driving vehicles, due to its compact design and high performance requirements, the evaluation of the heat dissipation structure layout scheme must consider multiple factors, such as heat distribution, material selection, and cooling system configuration. By comprehensively considering these factors, it can be determined which areas are the main heat source points, which paths carry the most heat transfer, and which places may become heat dissipation bottlenecks. Once the evaluation results are obtained, the next step is to perform fuzzy logic processing on them to obtain optimization direction indicators. Fuzzy logic processing allows us to make decisions under conditions of uncertainty and ambiguity, which is very useful for dealing with complex thermal management problems. The optimization direction indicators include the optimization path, optimization weight and optimization target of the heat dissipation structure, which provide guidance for subsequent optimization work. For example, inside the FPC module, some key components such as the processor may generate a lot of heat, while other areas are relatively stable. Through fuzzy logic processing, we can identify those hot spots that need special attention and determine the optimization path with the highest priority to ensure that limited resources can be used on the cutting edge. At the same time, different optimization weights can be set according to the needs of actual application scenarios. For example, when driving on urban roads, more attention is paid to the immediate cooling effect, while when driving on highways, more emphasis may be placed on long-term stability. Based on the above optimization direction indicators, the next step is to perform multi-objective optimization on the heat dissipation structure layout scheme to obtain a preliminary optimization scheme. This step aims to balance multiple conflicting objectives, such as structural strength, material heat resistance and heat dissipation efficiency. The multi-objective optimization solution not only takes into account the current working conditions, but also predicts possible changes in the future to ensure that the design scheme is flexible and adaptable enough. For example, in the scenario of autonomous driving, when the vehicle enters a busy urban road section, the sensor data processing volume surges and the temperature inside the FPC module rises rapidly. At this point, the preliminary optimization solution needs to be able to quickly adjust the heat dissipation strategy to cope with sudden high temperature conditions without affecting the overall performance. In addition, the preliminary optimization solution needs to be subjected to constraint analysis to obtain the optimization constraint set to ensure that the final design meets all necessary physical and technical requirements. Then, a genetic algorithm is used to perform genetic recombination and mutation operations on the preliminary optimization solution to generate a variant optimization solution. Genetic algorithms imitate the evolutionary process in nature and explore new solutions through operations such as selection, crossover, and mutation.In this process, each solution is regarded as an "individual", and its attributes (such as the position of the heat sink fins, fan speed, etc.) are used as "genes". Through continuous iteration, the proportion of excellent individuals in the group can be gradually increased, and finally converge to the optimal or near-optimal solution. Next, the variation optimization solution is Bayesian optimized to obtain the Bayesian optimization solution. Bayesian optimization is an optimization method based on a probabilistic model. It can efficiently find the global optimal solution, especially for non-convex optimization problems in high-dimensional space. The Bayesian optimization solution not only includes suggestions for improving the heat dissipation structure, but also involves specific parameter adjustments, providing direct operational guidance for practical applications. For example, in an autonomous driving vehicle, if it is found that the temperature fluctuation at a certain frequency is more obvious, this situation can be improved by adjusting the fan speed or changing the coolant flow. With the Bayesian optimization solution, the next step is to perform topological optimization on the heat dissipation structure layout solution based on it to obtain the topological optimization solution. Topological optimization refers to the redistribution of materials within a given volume to achieve the best mechanical and thermal properties. In this way, the heat dissipation effect can be significantly improved without adding extra weight. After completing the topology optimization, the scheme needs to be subjected to structural modal analysis to obtain the modal response curve. The modal response curve shows the vibration characteristics of the heat dissipation structure at different frequencies, which is very important for understanding the dynamic behavior of the structure. For example, during the operation of an autonomous vehicle, the FPC module may experience periodic mechanical stress changes due to factors such as road bumps. Through the modal response curve, it is possible to identify in advance which frequencies may cause resonance, so that measures can be taken to avoid it. Finally, in order to further verify and optimize the reliability of the heat dissipation structure, it is necessary to perform random perturbation tests on the modal response curve through Monte Carlo simulation to obtain the thermal stress distribution of the heat dissipation structure. Monte Carlo simulation is a statistical method that estimates the performance of a complex system through a large number of random sampling. On this basis, a finite element analysis is performed on the thermal stress distribution to generate a thermal stress field intensity distribution map. This distribution map intuitively reflects the thermal stress changes of the heat dissipation structure under different load conditions, helping us identify potential risk points. Based on this information, it is also necessary to perform a thermal-mechanical coupling analysis on the heat dissipation structure to obtain the thermal-mechanical coupling parameter set of the heat dissipation structure, and finally obtain the heat dissipation control parameter set through multi-physics field simulation. The heat dissipation control parameter set includes the temperature control parameters of the heat dissipation structure, the fan speed control parameters, and the coolant flow control parameters. These are key instructions to ensure that the FPC sensor module can maintain an ideal heat dissipation state under various working conditions. In summary, the above steps together constitute a complete process from the evaluation of the heat dissipation structure layout plan to the generation of the heat dissipation control parameter set. Each step is closely linked and interlocking, ensuring a comprehensive understanding and precise control of the complex thermal behavior inside the high-order computing power FPC sensor module.This systematic analysis and optimization method can provide reliable thermal management support for the autonomous driving platform, ensuring its stability and safety even in complex and changing working environments. In addition, this method can also provide valuable data foundation for long-term maintenance and optimization, further improving the safety and efficiency of the autonomous driving platform.

[0110] In a specific embodiment, the heat-mechanical coupling analysis of the heat dissipation structure is performed based on the thermal stress field intensity distribution diagram to obtain a heat-mechanical coupling parameter set of the heat dissipation structure, including:

[0111] Performing Helmholtz decomposition on the thermal stress field intensity distribution diagram to obtain a thermal-force field potential function, and performing variational principle derivation on the thermal-force field potential function to obtain a coupling control equation;

[0112] Discretizing the coupled control equations by a finite volume method to obtain thermal-mechanical grid units;

[0113] Solving the Navier-Stokes equations for the thermal-mechanical grid unit to obtain flow field distribution characteristics, and performing Rayleigh number analysis on the flow field distribution characteristics to obtain a convective heat transfer coefficient;

[0114] Performing a Boltzmann transport equation calculation based on the convective heat transfer coefficient to obtain a heat conduction flux, and performing a Maxwell stress tensor analysis based on the heat conduction flux to obtain a stress distribution field;

[0115] Performing thermal-mechanical coupling iterative calculation on the stress distribution field by nonlinear finite element method to obtain a coupled iterative sequence, and performing Richardson extrapolation processing on the coupled iterative sequence to obtain a converged solution set;

[0116] Based on the converged solution set, thermal-mechanical coupling parameters of the heat dissipation structure are extracted to obtain a thermal-mechanical coupling parameter set of the heat dissipation structure.

[0117] Specifically, in order to achieve the above-described "thermal-mechanical coupling analysis of the heat dissipation structure based on the thermal stress field intensity distribution map to obtain the thermal-mechanical coupling parameter set of the heat dissipation structure", this process involves a series of complex and interrelated technical steps. First, starting from the thermal stress field intensity distribution map, the Helmholtz decomposition is performed on it to extract the thermal-mechanical field potential function. Helmholtz decomposition is a method of decomposing a vector field into two parts, irrotational and irradiated, and it is used here to analyze the relationship between heat transfer and mechanical stress. By performing this mathematical processing on the data in the thermal stress field intensity distribution map, we can more clearly understand how heat affects the mechanical behavior of the structure, and vice versa. For example, in the high-order computing power FPC sensor module used in autonomous driving vehicles, some key components such as processors generate a lot of heat when processing data from sensors such as cameras and radars. This heat will not only cause the temperature to rise, but may also cause the material to expand or deform, thereby changing its mechanical properties. Once the thermal-mechanical field potential function is obtained, the next step is to derive the function by variational principle to obtain the coupling control equation. The variational principle provides a method to find the optimal solution of the system. By minimizing or maximizing a specific energy function (i.e., the control equation), the optimal state that satisfies all constraints can be found. In this case, the variational principle helps us establish a mathematical model that can simultaneously consider heat transfer and mechanical stress changes, ensuring a close coupling between the two. For example, inside the FPC module, when heat is transferred from the high temperature area to the low temperature area, it is accompanied by the thermal expansion and contraction of the material. The coupled control equation accurately describes this dynamic change and provides a theoretical basis for subsequent numerical simulation. With the coupled control equation, it is necessary to discretize it through the finite volume method to obtain the thermal-mechanical grid unit. The finite volume method is a numerical method widely used in the fields of fluid mechanics and heat transfer. It divides the continuous space into multiple small units and solves the conservation law in each unit. This method can not only accurately capture local details, but also effectively handle complex boundary conditions. For the FPC module, the application of the finite volume method allows us to carefully analyze the thermal-mechanical response of each component, especially in those places with complex geometries or interfaces of multiple materials. The thermal-mechanical grid units generated in this way provide a reliable discretization framework for subsequent calculations. Next, the Navier-Stokes equations are solved for these thermal-mechanical grid cells to obtain the flow field distribution characteristics, and the Rayleigh number analysis of the flow field distribution characteristics is performed to obtain the convective heat transfer coefficient. The Navier-Stokes equations describe the basic laws of fluid motion, including changes in velocity, pressure, and density. In this case, it is used to simulate the flow of coolant or other cooling media in the FPC module, revealing how heat is transferred through the fluid.Rayleigh number analysis is an important indicator for evaluating the intensity of natural convection. It reflects the movement trend of fluid due to temperature difference under the action of gravity. By calculating the convective heat transfer coefficient, we can understand the efficiency of heat exchange between different areas, which is crucial for optimizing heat dissipation design. For example, in an autonomous driving scenario, if the convective heat transfer coefficient of a specific area is found to be low, the situation can be improved by adjusting the fan speed or changing the coolant flow rate. Then, the Boltzmann transport equation is calculated based on the obtained convective heat transfer coefficient to obtain the heat conduction flux, and the heat conduction flux is further analyzed by Maxwell stress tensor to obtain the stress distribution field. The Boltzmann transport equation describes the movement law of particles at the microscopic level, and it is used here to simulate the heat propagation process in solid materials. Maxwell stress tensor analysis is used to study the mechanical stress distribution caused by temperature gradient, which helps us identify where large mechanical loads may be borne. For example, near the core processing unit of the FPC module, due to heavy computing tasks, there may be a significant temperature gradient in this area, resulting in additional stress inside the material. Through the above analysis, we can take measures to mitigate it in advance and avoid potential structural damage risks. Finally, the stress distribution field is iteratively calculated by thermal-mechanical coupling through nonlinear finite element method to obtain coupled iterative sequences, and these sequences are processed by Richardson extrapolation method to finally obtain converged solution sets. Nonlinear finite element method is a powerful numerical tool, especially suitable for solving complex problems involving geometric nonlinearity and material nonlinearity. Here, it is used to simulate the structural response under thermal-mechanical coupling to ensure that all physical phenomena are fully considered. Richardson extrapolation rule is a technique to accelerate convergence, which improves the calculation accuracy by combining multiple iteration results. In this way, we can ensure that the final solution is stable and reliable. Based on the converged solution set, the thermal-mechanical coupling parameters of the heat dissipation structure are extracted to obtain the thermal-mechanical coupling parameter set. This parameter set includes a series of key indicators that reflect the performance of the heat dissipation structure, such as temperature control parameters, fan speed control parameters, and coolant flow control parameters, which together constitute the operating instructions to ensure that the FPC sensor module can maintain an ideal heat dissipation state under various working conditions. In summary, the above steps together constitute a complete process from thermal stress field intensity distribution map analysis to thermal-mechanical coupling parameter set generation. Each step is closely linked to each other, ensuring a comprehensive understanding and precise control of the complex thermal-mechanical behavior inside the high-level computing power FPC sensor module. Through this systematic analysis and optimization method, even in the face of complex and changing working environments, it can provide reliable thermal management support for the autonomous driving platform and ensure its stability and safety.

[0118] In a specific embodiment, the step of inputting the heat dissipation control parameter set into a preset execution unit, and performing real-time heat dissipation control on the target FPC sensor module through the execution unit to obtain an ideal heat dissipation state includes:

[0119] Parsing the heat dissipation control parameter set to obtain a control parameter sequence;

[0120] Adaptively slice-code the control parameter sequence to obtain a parameter control instruction set;

[0121] The optimal trajectory planning of the parameter control instruction set is carried out through the preset Hamilton-Jacobi equation to obtain the control trajectory sequence, and the control trajectory sequence is subjected to Lyapunov stability analysis to obtain the control trajectory sequence in the steady-state control interval;

[0122] Inputting the control trajectory sequence of the steady-state control interval into a preset execution unit, and performing nonlinear robust control on the execution unit to obtain an execution response curve;

[0123] Performing wavelet packet decomposition processing on the execution response curve to obtain multi-level response characteristics; wherein the multi-level response characteristics include temperature dynamic characteristics, fan speed characteristics and coolant flow characteristics;

[0124] Based on the multi-level response characteristics, real-time heat dissipation control is performed on the target FPC sensor module to obtain an ideal heat dissipation state.

[0125] Specifically, in order to achieve the above-described "inputting the heat dissipation control parameter set into the preset execution unit, and performing real-time heat dissipation control on the target FPC sensor module to obtain an ideal heat dissipation state", this process involves a series of complex and interrelated technical steps. First, starting from the heat dissipation control parameter set, it is parsed to obtain a control parameter sequence. These parameters may include temperature control parameters, fan speed control parameters, and coolant flow control parameters, which are key instructions to ensure that the FPC sensor module can maintain an ideal heat dissipation state under various working conditions. For example, in the high-order computing power FPC sensor module used in autonomous driving vehicles, when processing data from sensors such as cameras and radars, different components will generate different heat, so each parameter needs to be accurately adjusted to optimize the heat dissipation effect. Once the control parameter sequence is obtained, the next step is to adaptively slice the sequence to generate a parameter control instruction set. Adaptive slice coding is a data compression technology that can dynamically adjust the encoding strategy according to actual needs, thereby improving transmission efficiency and reducing redundant information. In this way, a complex control parameter sequence can be converted into a set of concise and efficient instructions, which is convenient for the understanding and operation of subsequent execution units. For example, in an autonomous driving scenario, if it is found that the temperature fluctuations in a certain area are more frequent, these key parameters can be prioritized through adaptive shard coding to ensure the rapid response of the cooling system. With the parameter control instruction set, the next step is to plan the optimal trajectory through the preset Hamilton-Jacobi equation to obtain the control trajectory sequence. The Hamilton-Jacobi equation is a mathematical tool for solving optimal control problems. It can find the path that makes the system reach the optimal state under given initial conditions and objective functions. In this process, not only the current working conditions are considered, but also possible changes in the future are predicted to ensure that the design scheme has sufficient flexibility and adaptability. For example, during the driving of an autonomous vehicle, before it is expected to enter a long period of high-speed driving, the system can plan the optimal cooling path in advance to prepare for the upcoming high-intensity operations. Then, Lyapunov stability analysis is performed on the control trajectory sequence to determine the control trajectory sequence of the steady-state control interval. Lyapunov stability analysis is a method for evaluating the stability of a system. It can help us identify which control trajectories can keep the system within a stable range. The steady-state control interval refers to those areas where the system can still maintain its performance even under the influence of external disturbances or internal changes. By performing this analysis on the control trajectory sequence, we can screen out the most suitable control scheme to ensure that the FPC sensor module is always in an ideal heat dissipation state. For example, when a vehicle is driving on a city road, changes in traffic conditions may cause load fluctuations in the FPC module. Through Lyapunov stability analysis, we can select those control trajectories that can maintain stability in complex environments.Next, the control trajectory sequence of the steady-state control interval is input into the preset execution unit, and the execution unit is subjected to nonlinear robust control to obtain the execution response curve. Nonlinear robust control is a control method that can cope with uncertainty and interference. It ensures that the system can accurately perform the predetermined control task even in a complex dynamic environment. The execution response curve shows the actual performance of the execution unit (such as fans, pumps, valves, etc.) under different control instructions, which is very important for verifying the control effect. For example, in the scenario of autonomous driving, if it is found that the response time of a fan is long, its performance can be improved by adjusting the control algorithm to ensure the efficient operation of the cooling system. Finally, the execution response curve is processed by wavelet packet decomposition to obtain multi-level response characteristics, including temperature dynamic characteristics, fan speed characteristics, and coolant flow characteristics. Wavelet packet decomposition is a signal processing technology that can decompose complex response curves into multiple levels of characteristics, helping us to understand the behavior of each component more deeply. For example, by analyzing the temperature dynamic characteristics, we can understand the temperature change trend of the FPC module in different time periods; the fan speed characteristics reflect the working status of the cooling device, and the coolant flow characteristics reveal the transmission of the cooling medium. Based on these multi-level response characteristics, the target FPC sensor module can be subjected to real-time heat dissipation control, and finally an ideal heat dissipation state can be obtained. In summary, the above steps together constitute a complete process from the analysis of the heat dissipation control parameter set to the implementation of real-time heat dissipation control. Each step is closely linked and interlocked, ensuring a comprehensive understanding and precise control of the complex thermal behavior inside the high-order computing power FPC sensor module. Through this systematic analysis and optimization method, even in the face of complex and changing working environments, reliable thermal management support can be provided for the autonomous driving platform to ensure its stability and safety.

[0126] The above describes the heat dissipation control method of the high-order computing power autonomous driving FPC sensor module in the embodiment of the present invention. The following describes the heat dissipation control system of the high-order computing power autonomous driving FPC sensor module in the embodiment of the present invention. Please refer to Figure 2 , an embodiment of the heat dissipation control system of the high-order computing power autonomous driving FPC sensor module in the embodiment of the present invention includes:

[0127] The acquisition module 21 is used to collect temperature data of the target FPC sensor module through a preset distributed thermal array to obtain three-dimensional temperature field data;

[0128] A calculation module 22, used to calculate the heat flow path of the target FPC sensor module based on the three-dimensional temperature field data to obtain a heat flux density distribution diagram;

[0129] An optimization module 23 is used to perform topology optimization calculation on the target FPC sensor module based on the heat flux density distribution diagram to obtain a heat dissipation structure layout solution;

[0130] A solution module 24 is used to dynamically optimize and solve the heat dissipation structure layout scheme through a branch and bound algorithm to obtain a heat dissipation control parameter set;

[0131] The control module 25 is used to input the heat dissipation control parameter set into a preset execution unit, and perform real-time heat dissipation control on the target FPC sensor module through the execution unit to obtain an ideal heat dissipation state.

[0132] In this embodiment, for the specific implementation of each unit in the above system embodiment, please refer to the above method embodiment, which will not be repeated here.

[0133] Reference Figure 3 The present invention also provides a computer device in an embodiment, wherein the internal structure of the computer device can be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a display screen, an input device, a network interface and a database connected through a system bus. Among them, the processor designed by the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0134] Those skilled in the art will understand that Figure 3 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0135] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the above method is implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0136] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided by the present invention and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-speed data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM.

[0137] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0138] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A heat dissipation control method for a high-order computing power autonomous driving FPC sensor module, characterized in that: The following steps are involved: The temperature data of the target FPC sensor module is collected through a preset distributed thermal array to obtain three-dimensional temperature field data; Calculating the heat flow path of the target FPC sensor module based on the three-dimensional temperature field data to obtain a heat flux density distribution diagram; Based on the heat flux density distribution diagram, a topology optimization calculation is performed on the target FPC sensor module to obtain a heat dissipation structure layout solution; The heat dissipation structure layout scheme is dynamically optimized and solved by the branch and bound algorithm to obtain the heat dissipation control parameter set; The heat dissipation control parameter set is input into a preset execution unit, and the target FPC sensor module is subjected to real-time heat dissipation control by the execution unit to obtain an ideal heat dissipation state; The topology optimization calculation of the target FPC sensor module is performed based on the heat flux density distribution diagram to obtain a heat dissipation structure layout solution, including: Performing multi-scale decomposition on the heat flux density distribution map to obtain a heat flux characteristic spectrum, and performing wavelet transformation on the heat flux characteristic spectrum to obtain heat flux spectrum characteristics; wherein the heat flux spectrum characteristics include heat flux frequency, heat flux amplitude and heat flux phase; Performing a heat flow network analysis on the target FPC sensor module through the heat flow spectrum characteristics to obtain a heat flow network topology structure, and performing community discovery on the heat flow network topology structure to obtain a heat flow community distribution feature; wherein the heat flow community distribution feature includes the number of heat flow communities, the scale of heat flow communities, and the connectivity of heat flow communities; Based on the heat flow community distribution characteristics, a heat flow optimization calculation is performed on the target FPC sensor module to obtain a heat flow optimization parameter; Based on the heat flow optimization parameters, a topological structure of the target FPC sensor module is generated to obtain a heat dissipation structure topological map, and a finite element analysis is performed on the heat dissipation structure topological map to obtain heat dissipation structure performance parameters; wherein the heat dissipation structure performance parameters include heat dissipation structure temperature, heat dissipation structure stress data and heat dissipation structure deformation data; Based on the heat dissipation structure performance parameters and the heat dissipation structure topology diagram, a heat dissipation structure layout calculation is performed on the target FPC sensor module to obtain a heat dissipation structure layout solution; The heat dissipation structure layout scheme is dynamically optimized and solved by the branch and bound algorithm to obtain a heat dissipation control parameter set, including: An initial state evaluation is performed on the heat dissipation structure layout scheme by a branch and bound algorithm to obtain an evaluation result, and fuzzy logic processing is performed on the evaluation result to obtain an optimization direction index; wherein the optimization direction index includes an optimization path, an optimization weight and an optimization target of the heat dissipation structure; Based on the optimization direction index, a multi-objective optimization solution is performed on the heat dissipation structure layout scheme to obtain a preliminary optimization scheme, and a constraint condition analysis is performed on the preliminary optimization scheme to obtain an optimization constraint set; wherein the optimization constraint set includes constraints on structural strength, material heat resistance and heat dissipation efficiency; Performing genetic recombination and mutation operations on the preliminary optimization scheme through a genetic algorithm to obtain a variant optimization scheme, and performing Bayesian optimization on the variant optimization scheme to obtain a Bayesian optimization scheme; wherein the Bayesian optimization scheme includes improvement suggestions and parameter adjustments for the heat dissipation structure in the target FPC sensor module; Based on the Bayesian optimization scheme, topology optimization is performed on the heat dissipation structure layout scheme to obtain a topology optimization scheme, and structural modal analysis is performed on the topology optimization scheme to obtain a modal response curve; wherein the modal response curve includes the vibration response of the heat dissipation structure at different frequencies; Performing random perturbation test on the modal response curve through Monte Carlo simulation to obtain the thermal stress distribution of the heat dissipation structure, and performing finite element analysis on the thermal stress distribution to obtain a thermal stress field intensity distribution diagram; wherein the thermal stress field intensity distribution diagram is used to reflect the thermal stress change of the heat dissipation structure under different loads; Based on the thermal stress field intensity distribution diagram, a thermal-mechanical coupling analysis is performed on the heat dissipation structure to obtain a thermal-mechanical coupling parameter set of the heat dissipation structure, and a multi-physical field simulation is performed on the thermal-mechanical coupling parameter set to obtain a heat dissipation control parameter set; wherein the heat dissipation control parameter set includes temperature control parameters, fan speed control parameters and coolant flow control parameters of the heat dissipation structure.

2. The heat dissipation control method of the high-order computing power autonomous driving FPC sensor module according to claim 1 is characterized in that: The temperature data of the target FPC sensor module is collected by a preset distributed thermal array to obtain three-dimensional temperature field data, including: The target FPC sensor module is subjected to high-frequency scanning sampling by a preset distributed thermal array to obtain an original temperature signal sequence; Filtering the original temperature signal sequence to obtain a denoised temperature data stream; Performing Hilbert-Huang transform processing based on the de-noised temperature data stream to obtain instantaneous temperature characteristics; The instantaneous temperature characteristics are spatially reconstructed by an adaptive grid subdivision algorithm to obtain a temperature field grid matrix; wherein the temperature field grid matrix includes grid node coordinates and temperature data corresponding to the grid node coordinates; Performing Kriging interpolation operation on the temperature field grid matrix to obtain a continuous temperature field distribution; The continuous temperature field distribution is gradient calculated to obtain a temperature field gradient vector, and three-dimensional temperature field data is obtained based on the temperature field gradient vector and the temperature field grid matrix; wherein the three-dimensional temperature field data includes the temperature value of the hot spot area, the temperature fluctuation frequency and the temperature gradient direction.

3. The heat dissipation control method of the high-order computing power autonomous driving FPC sensor module according to claim 1 is characterized in that: The heat flow path calculation of the target FPC sensor module based on the three-dimensional temperature field data to obtain a heat flux density distribution diagram includes: Numerically discretizing the three-dimensional temperature field data by a finite difference method to obtain temperature gradient field data; Performing thermal conductivity analysis on the target FPC sensor module based on the temperature gradient field data to obtain thermal conductivity spatial distribution characteristics; By using the variational method, based on the thermal conductivity spatial distribution characteristics, the target FPC sensor module is subjected to heat flow path analysis to obtain a heat flow channel topology structure, and the heat flow channel topology structure is subjected to connectivity analysis to obtain a heat flow transmission network; Performing spectral analysis on the thermal impedance in the target FPC sensor module based on the heat flow transmission network to obtain impedance characteristic distribution, and extracting key heat flow paths of the heat flow transmission network based on the impedance characteristic distribution; The thermal field intensity of the key heat flow path is calculated by the fast multipole method to obtain the heat flow field intensity data, and the gradient of the heat flow field intensity data is tracked to obtain the heat flow convergence area; Gaussian curvature analysis is performed on the heat flux density in the heat flux convergence area to obtain a density distribution characteristic map, and heat flux density calculation is performed on the density distribution characteristic map to obtain a heat flux density distribution map.

4. The heat dissipation control method of the high-order computing power autonomous driving FPC sensor module according to claim 1 is characterized in that: The heat-mechanical coupling analysis of the heat dissipation structure is performed based on the thermal stress field intensity distribution diagram to obtain a heat-mechanical coupling parameter set of the heat dissipation structure, including: Performing Helmholtz decomposition on the thermal stress field intensity distribution diagram to obtain a thermal-force field potential function, and performing variational principle derivation on the thermal-force field potential function to obtain a coupling control equation; Discretizing the coupled control equations by a finite volume method to obtain thermal-mechanical grid units; Solving the Navier-Stokes equations for the thermal-mechanical grid unit to obtain flow field distribution characteristics, and performing Rayleigh number analysis on the flow field distribution characteristics to obtain a convective heat transfer coefficient; Performing a Boltzmann transport equation calculation based on the convective heat transfer coefficient to obtain a heat conduction flux, and performing a Maxwell stress tensor analysis based on the heat conduction flux to obtain a stress distribution field; Performing thermal-mechanical coupling iterative calculation on the stress distribution field by nonlinear finite element method to obtain a coupled iterative sequence, and performing Richardson extrapolation processing on the coupled iterative sequence to obtain a converged solution set; Based on the converged solution set, thermal-mechanical coupling parameters of the heat dissipation structure are extracted to obtain a thermal-mechanical coupling parameter set of the heat dissipation structure.

5. The heat dissipation control method of the high-order computing power autonomous driving FPC sensor module according to claim 1 is characterized in that: The step of inputting the heat dissipation control parameter set into a preset execution unit, and performing real-time heat dissipation control on the target FPC sensor module through the execution unit to obtain an ideal heat dissipation state includes: Parsing the heat dissipation control parameter set to obtain a control parameter sequence; Adaptively slice-code the control parameter sequence to obtain a parameter control instruction set; The optimal trajectory planning of the parameter control instruction set is carried out through the preset Hamilton-Jacobi equation to obtain the control trajectory sequence, and the control trajectory sequence is subjected to Lyapunov stability analysis to obtain the control trajectory sequence in the steady-state control interval; Inputting the control trajectory sequence of the steady-state control interval into a preset execution unit, and performing nonlinear robust control on the execution unit to obtain an execution response curve; Performing wavelet packet decomposition processing on the execution response curve to obtain multi-level response characteristics; wherein the multi-level response characteristics include temperature dynamic characteristics, fan speed characteristics and coolant flow characteristics; Based on the multi-level response characteristics, real-time heat dissipation control is performed on the target FPC sensor module to obtain an ideal heat dissipation state.

6. A heat dissipation control device for a high-level computing power autonomous driving FPC sensor module, characterized in that: include: The acquisition module is used to collect temperature data of the target FPC sensor module through a preset distributed thermal array to obtain three-dimensional temperature field data; A calculation module, used to calculate the heat flow path of the target FPC sensor module based on the three-dimensional temperature field data to obtain a heat flux density distribution diagram; An optimization module, used to perform topology optimization calculation on the target FPC sensor module based on the heat flux density distribution diagram to obtain a heat dissipation structure layout solution; A solution module, used for dynamically optimizing and solving the heat dissipation structure layout scheme through a branch and bound algorithm to obtain a heat dissipation control parameter set; A control module, used for inputting the heat dissipation control parameter set into a preset execution unit, and performing real-time heat dissipation control on the target FPC sensor module through the execution unit to obtain an ideal heat dissipation state; The topology optimization calculation of the target FPC sensor module is performed based on the heat flux density distribution diagram to obtain a heat dissipation structure layout solution, including: Performing multi-scale decomposition on the heat flux density distribution map to obtain a heat flux characteristic spectrum, and performing wavelet transformation on the heat flux characteristic spectrum to obtain heat flux spectrum characteristics; wherein the heat flux spectrum characteristics include heat flux frequency, heat flux amplitude and heat flux phase; Performing a heat flow network analysis on the target FPC sensor module through the heat flow spectrum characteristics to obtain a heat flow network topology structure, and performing community discovery on the heat flow network topology structure to obtain a heat flow community distribution feature; wherein the heat flow community distribution feature includes the number of heat flow communities, the scale of heat flow communities, and the connectivity of heat flow communities; Based on the heat flow community distribution characteristics, a heat flow optimization calculation is performed on the target FPC sensor module to obtain a heat flow optimization parameter; Based on the heat flow optimization parameters, a topological structure of the target FPC sensor module is generated to obtain a heat dissipation structure topological map, and a finite element analysis is performed on the heat dissipation structure topological map to obtain heat dissipation structure performance parameters; wherein the heat dissipation structure performance parameters include heat dissipation structure temperature, heat dissipation structure stress data and heat dissipation structure deformation data; Based on the heat dissipation structure performance parameters and the heat dissipation structure topology diagram, a heat dissipation structure layout calculation is performed on the target FPC sensor module to obtain a heat dissipation structure layout solution; The heat dissipation structure layout scheme is dynamically optimized and solved by the branch and bound algorithm to obtain a heat dissipation control parameter set, including: An initial state evaluation is performed on the heat dissipation structure layout scheme by a branch and bound algorithm to obtain an evaluation result, and fuzzy logic processing is performed on the evaluation result to obtain an optimization direction index; wherein the optimization direction index includes an optimization path, an optimization weight and an optimization target of the heat dissipation structure; Based on the optimization direction index, a multi-objective optimization solution is performed on the heat dissipation structure layout scheme to obtain a preliminary optimization scheme, and a constraint condition analysis is performed on the preliminary optimization scheme to obtain an optimization constraint set; wherein the optimization constraint set includes constraints on structural strength, material heat resistance and heat dissipation efficiency; Performing genetic recombination and mutation operations on the preliminary optimization scheme through a genetic algorithm to obtain a variant optimization scheme, and performing Bayesian optimization on the variant optimization scheme to obtain a Bayesian optimization scheme; wherein the Bayesian optimization scheme includes improvement suggestions and parameter adjustments for the heat dissipation structure in the target FPC sensor module; Based on the Bayesian optimization scheme, topology optimization is performed on the heat dissipation structure layout scheme to obtain a topology optimization scheme, and structural modal analysis is performed on the topology optimization scheme to obtain a modal response curve; wherein the modal response curve includes the vibration response of the heat dissipation structure at different frequencies; Performing random perturbation test on the modal response curve through Monte Carlo simulation to obtain the thermal stress distribution of the heat dissipation structure, and performing finite element analysis on the thermal stress distribution to obtain a thermal stress field intensity distribution diagram; wherein the thermal stress field intensity distribution diagram is used to reflect the thermal stress change of the heat dissipation structure under different loads; Based on the thermal stress field intensity distribution diagram, a thermal-mechanical coupling analysis is performed on the heat dissipation structure to obtain a thermal-mechanical coupling parameter set of the heat dissipation structure, and a multi-physical field simulation is performed on the thermal-mechanical coupling parameter set to obtain a heat dissipation control parameter set; wherein the heat dissipation control parameter set includes temperature control parameters, fan speed control parameters and coolant flow control parameters of the heat dissipation structure.

7. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images