Circuit board device thermal analysis system based on target detection

Abstract

The present invention relates to the technical field of thermal management of electronic devices, and in particular to a target detection-based thermal analysis system for circuit board devices. The system comprises a thermal processor, which is communicatively connected to a thermal map generation module, a thermal model construction and analysis module, an improvement and evaluation module, and a thermal response module. The thermal map generation module is provided with an image acquisition unit and an image processing unit. The present invention accurately identifies and locates devices on a PCB board through image processing and target detection technologies, and precisely associates thermal infrared images with the devices, so that thermal analysis more accurately reflects the actual heat generation of the PCB board. Through model construction and analysis, a more targeted basis is provided for thermal management. At the same time, a thermal analysis label is obtained based on information feedback, which helps to carry out targeted management of the PCB board based on the thermal analysis label. By performing simulation evaluation on the improvement of the improvement plan, it is judged whether the improvement plan of the PCB board meets the standards, and the improvement effect and feasibility of the improvement plan are intuitively understood.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic device thermal management, and in particular to a circuit board device thermal analysis system based on target detection. Background Art

[0002] As electronic devices develop towards miniaturization and high integration, the density of devices on circuit boards continues to increase, and thermal management issues are becoming increasingly prominent. If the heat generated by electronic devices during operation cannot be dissipated promptly and effectively, the device temperature will rise, affecting its performance and reliability, and even causing system failures. Therefore, accurate thermal analysis of circuit board devices and timely detection of hot spots and thermal problems are of great significance to ensuring the stable operation of electronic equipment.

[0003] The existing patent CN110942458A proposes a method and system for detecting and locating temperature anomaly defects. The system includes an infrared thermal imaging temperature measurement module, a visible light imaging module, a parameter setting module, a result display module, an alarm module, and an image analysis and positioning module. The temperature anomaly detection and positioning results are obtained through image analysis and processing. However, the existing technology still has the following problems in thermal analysis of circuit board components: it is difficult to accurately associate hot zones with specific electronic components and locate the specific source of thermal problems, resulting in reduced positioning accuracy of electronic components. It is also difficult to accurately identify overheating components and provide feedback on the cause type, which is not conducive to the management of overheating components. At the same time, the existing technology lacks a thermal management optimization suggestion mechanism and cannot provide targeted improvement solutions for circuit board design.

[0004] In view of the above technical defects, a solution is now proposed. Summary of the Invention

[0005] The object of the present invention is to provide a circuit board device thermal analysis system based on target detection to address the above-mentioned technical deficiencies. The present invention accurately identifies and locates devices on a PCB board through image processing and target detection technology, and precisely associates thermal infrared images with devices to obtain temperature data for each device, so that thermal analysis more accurately reflects the actual heat generation of the PCB board. During the association process, the image association matching accuracy is analyzed to improve the image registration accuracy and reliability. Through model construction and analysis, the temperature data of each device can be analyzed in detail, and local hotspots on the PCB board can be discovered in a timely manner, that is, the local hotspot detection capability is enhanced, providing a more targeted basis for thermal management. At the same time, a thermal analysis label is obtained based on information feedback, which helps to intuitively understand the thermal analysis feedback results of the PCB board on the one hand, and to perform targeted management of the PCB board based on the thermal analysis label on the other hand. By performing simulation evaluation on the improvement of the improvement plan, it is determined whether the improvement plan of the PCB board meets the standards. The information feedback helps to intuitively understand the improvement effect and feasibility of the improvement plan.

[0006] The object of the present invention can be achieved by the following technical solution: a circuit board device thermal analysis system based on target detection includes a thermal processor, the thermal processor is communicatively connected to a thermal map generation module, a thermal model construction and analysis module, an improvement and evaluation module, and a thermal response module, and the thermal map generation module is provided with an image acquisition unit and an image processing unit;

[0007] The image acquisition unit is used to acquire the optical characteristic image and thermal infrared characteristic image of the PCB board, and the image processing unit is used to perform image matching and fusion analysis on the optical characteristic image and the thermal infrared characteristic image to obtain a device-temperature map;

[0008] The thermal processor is used to store the response device-temperature map and display the device-temperature map on the display panel;

[0009] The thermal model construction and analysis module is used to construct a three-dimensional heat conduction model of the PCB board and perform thermal management analysis and calibration operations to obtain the overheating device + defect type + defect hazard score, and set the overheating device + defect type + defect hazard score as the thermal analysis label;

[0010] The improvement evaluation module is used to evaluate and analyze the improvement effect of the improvement parameters in the improvement plan pre-generated for the PCB board, and to perform judgment processing on the obtained improvement effect scores to determine whether the improvement plan of the PCB board meets the standards.

[0011] Preferably, the image matching fusion analysis process is as follows:

[0012] S1: Place the PCB to be analyzed in a fixed position, and use an optical camera and a thermal infrared camera to obtain the optical characteristic image and thermal infrared characteristic image of the PCB;

[0013] S2: Input the optical feature image into a pre-set target detection model, which processes the optical feature image and identifies the category information and location coordinates of each device;

[0014] S3: Preprocessing the optical feature image and the thermal infrared feature image, and registering the preprocessed optical feature image and the thermal infrared feature image through feature point matching and transformation matrix calculation;

[0015] S4: Extracting temperature data of each device in the PCB board, including the maximum temperature, minimum temperature, and average temperature. Obtaining a device-temperature map of the PCB board based on the temperature data and position coordinates of each device.

[0016] Preferably, S3: the process of achieving registration is as follows:

[0017] S31: Set the pads, vias, and device corners in the PCB board as feature points;

[0018] S32: Obtain visual description features of each feature point, where the visual description features include temperature gradient direction and texture pattern;

[0019] S33: Based on the visual description features of each feature point and a preset nearest neighbor algorithm, the Hamming distance of the visual description features of the optical feature image and the thermal infrared feature image is calculated, and feature point matching pairs whose Hamming distance is less than a preset Hamming distance threshold are screened.

[0020] Preferably, it also includes S34: randomly extracting n groups of feature point matching pairs, where n is a natural number greater than 3, and constructing the spatial coordinates of the thermal infrared feature image and the spatial coordinates of the optical feature image respectively, and selecting a point to be mapped in the thermal infrared feature image, and mapping the spatial coordinates of the thermal infrared feature image to the optical feature image coordinate system through the existing affine transformation technology to obtain a mapping point, and obtaining the ratio of the number of feature point matching pairs corresponding to the preset error threshold and n when the error value between the mapping point and the corresponding actual point is less than the preset error threshold, and setting the ratio of the number of feature point matching pairs corresponding to the preset error threshold and n as the matching accuracy, and performing discrimination processing on the matching accuracy to obtain an adjustment signal or a matching signal.

[0021] Preferably, the thermal management analysis calibration operation process is as follows:

[0022] A three-dimensional heat conduction model of the PCB is constructed based on the PCB design drawings and device-temperature diagrams. The PCB is divided into multiple grid cells in the three-dimensional heat conduction model. The thermal conductivity and specific heat capacity of each grid cell are set. Environmental conditions and constraints are also set. Environmental conditions include ambient temperature and air flow rate, while constraints include device power consumption and interface contact thermal resistance.

[0023] Set the simulation duration, and use the 3D heat conduction model to obtain the power consumption of each device during the simulation duration. Set the device power consumption as the internal heat source, and then obtain the heat conduction process within the PCB during the simulation of each internal heat source.

[0024] Based on the three-dimensional heat conduction model, the temperature distribution characteristic images and heat flux density distribution characteristic images of the PCB board at different time points are obtained through simulation.

[0025] Preferably, the temperature distribution characteristic image and the heat flux density distribution characteristic image are input as input layers to a pre-set thermal defect recognition model to obtain an output result of the thermal defect recognition model, the output result including the defect type and the defect hazard score;

[0026] At the same time, the temperature value of each component in the three-dimensional heat conduction model during the simulation time is obtained, the component whose temperature value exceeds the preset temperature value threshold is obtained, and the component whose temperature value exceeds the preset temperature value threshold is set as a risk component. At the same time, the duration corresponding to the temperature value of the risk component exceeds the preset temperature value threshold is obtained, and the risk component whose duration exceeds the preset duration threshold is set as an overheating component;

[0027] Set the overheating device + defect type + defect severity score as the thermal analysis label.

[0028] Preferably, the improvement effect evaluation feedback analysis process is as follows: based on the thermal analysis tag, improvement parameters in the improvement scheme pre-generated for the PCB board are obtained, the improvement parameters in the improvement scheme pre-generated for the PCB board are reset in the three-dimensional heat conduction model, and improvement effect parameters of the overheating device before and after optimization are obtained, and the improvement effect parameters include temperature drop value and heat flux density peak drop.

[0029] Preferably, a pre-set weight factor coefficient is assigned to each parameter in the improvement effect parameters, each parameter in the simulation effect parameters is multiplied by the corresponding weight factor coefficient, the sum of the multiplications of each parameter in the simulation effect parameters and the corresponding weight factor coefficient is set as the improvement effect score, and the improvement effect score is judged and processed to obtain an unqualified signal or a qualified signal.

[0030] The beneficial effects of the present invention are as follows:

[0031] (1) The present invention accurately identifies and locates devices on a PCB board through image processing and target detection technology, and precisely associates thermal infrared images with the devices to obtain temperature data for each device, so that thermal analysis can more accurately reflect the actual heat generation of the PCB board. In the association process, the image association matching accuracy is analyzed to improve the image registration accuracy and reliability.

[0032] (2) Through model construction and analysis, the present invention can analyze the temperature data of each device in detail and timely discover local hot spots on the PCB board, that is, enhance the local hot spot detection capability, and provide a more targeted basis for thermal management. At the same time, based on information feedback, a thermal analysis label is obtained, which helps to intuitively understand the thermal analysis feedback results of the PCB board. On the other hand, the PCB board can be targetedly managed based on the thermal analysis label. By performing simulation evaluation on the improvement of the improvement plan, it is judged whether the improvement plan of the PCB board meets the standards. Information feedback helps to intuitively understand the improvement effect and feasibility of the improvement plan. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The present invention will be further described below with reference to the accompanying drawings;

[0034] Figure 1It is a flow chart of the system of the present invention;

[0035] Figure 2 It is a local analysis reference diagram of the present invention;

[0036] Figure 3 2 is a diagram of analysis steps in Example 1 of the present invention. DETAILED DESCRIPTION

[0037] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0038] References to "embodiments" herein mean that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute a separate or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments;

[0039] Example 1:

[0040] See also Figures 1 to 3 As shown, the present invention is a circuit board device thermal analysis system based on target detection, including a thermal processor, the thermal processor is communicatively connected to a thermal map generation module, a thermal model construction and analysis module, an improvement and evaluation module, and a thermal response module, and the thermal map generation module is provided with an image acquisition unit and an image processing unit;

[0041] The thermal processor is connected to the thermal map generation module in a two-way communication manner, the thermal processor is connected to the thermal model construction and analysis module in a one-way communication manner, the thermal model construction and analysis module is connected to the improvement evaluation module and the thermal response module in a one-way communication manner, and the improvement evaluation module is connected to the thermal response module in a one-way communication manner;

[0042] The image acquisition unit is used to acquire the optical characteristic image and thermal infrared characteristic image of the PCB board;

[0043] The image processing unit is used to perform image matching and fusion analysis on the optical feature image and the thermal infrared feature image to obtain the device-temperature map. The specific image matching and fusion analysis process is as follows:

[0044] S1: Place the PCB to be analyzed in a fixed position and use an optical camera and a thermal infrared camera to acquire optical and thermal infrared characteristic images of the PCB. It should be noted that the image resolution should meet the requirements for device identification and temperature analysis. In actual applications, first, when the PCB is in normal working condition, infrared thermal imaging images and visible light images are simultaneously acquired. During acquisition, ensure that the PCB has been operating stably for at least 30 minutes to obtain the steady-state thermal distribution.

[0045] S2: Input the optical feature image into a pre-set target detection model. The target detection model processes the optical feature image and identifies the category information and location coordinates of each component. The category information includes various components such as resistors, capacitors, inductors, and chips. That is, the visible light image is processed and the location and type of each component on the PCB are identified through a deep learning network. The identification results include the component type and location coordinates.

[0046] For example, the target detection model breaks down the optical feature image into pixels and uses a convolutional neural network (CNN) to extract features layer by layer. These features range from device edges and corners to the overall shape and texture of the device (e.g., the circular shape of a capacitor or the rectangular outline of a chip).

[0047] Use a large amount of labeled data (circuit board images with component locations and categories) to train the model so that it can remember "what features correspond to which components", such as "rectangle + pin features = chip" and "circle + color ring = resistor";

[0048] S3: Preprocessing the optical feature image and the thermal infrared feature image. The preprocessing includes noise reduction, enhancement, and size unification. The preprocessed optical feature image and the thermal infrared feature image are registered through feature point matching and transformation matrix calculation.

[0049] S4: Extracting temperature data of each device on the PCB board, including the maximum temperature, minimum temperature, average temperature, etc., and obtaining a device-temperature map of the PCB board based on the temperature data and position coordinates of each device;

[0050] The thermal processor is used to store the response device-temperature map and display the device-temperature map on the display panel so that the temperature of each device in the PCB board and the temperature distribution of the PCB board can be intuitively understood;

[0051] In the embodiments of the present invention, target detection technology is used to accurately associate hot spots with specific electronic components, accurately locating the source of thermal problems; device-level thermal characteristic analysis is provided to evaluate the contribution of each component to the overall thermal distribution; thermal analysis results are combined with PCB board design data to provide precise guidance for thermal design optimization; the high degree of automation can significantly improve the efficiency and accuracy of PCB board thermal analysis;

[0052] S3: The process of registering the pre-processed optical feature image and thermal infrared feature image through feature point matching and transformation matrix calculation is as follows:

[0053] S31: Set the pads, vias, device corners, etc. on the PCB board as feature points;

[0054] S32: Obtain visual description features of each feature point, where the visual description features include temperature gradient direction, texture pattern, etc.

[0055] S33: Based on the visual description features of each feature point and a preset nearest neighbor algorithm, the Hamming distance of the feature points in the optical feature image and the thermal infrared feature image is calculated, and matching pairs of feature points whose Hamming distance is less than a preset Hamming distance threshold are selected;

[0056] That is, the similarity between visual description features is measured to match the feature points in the optical feature image and the thermal infrared feature image, and the Hamming distance is calculated using a preset nearest neighbor algorithm. The smaller the Hamming distance, the more similar the two feature points are.

[0057] The nearest neighbor algorithm is an instance-based machine learning method, which achieves classification or regression prediction by calculating the distance between samples.

[0058] S34: randomly extracting n groups of feature point matching pairs, where n is a natural number greater than 3, and constructing the spatial coordinates of the thermal infrared feature image and the spatial coordinates of the optical feature image respectively, and selecting a point to be mapped in the thermal infrared feature image, and mapping the spatial coordinates of the thermal infrared feature image to the optical feature image coordinate system through the existing affine transformation technology to obtain a mapping point, obtaining the ratio of the number of feature point matching pairs corresponding to the error value between the mapping point and the corresponding actual point being less than a preset error threshold to n, and setting the ratio of the number of feature point matching pairs corresponding to the error value between the mapping point and the actual point being less than the preset error threshold to n as the matching accuracy, and performing a discrimination process on the matching accuracy, and generating an adjustment signal if the matching accuracy is less than the preset matching accuracy threshold, and generating a matching signal if the matching accuracy is greater than or equal to the preset matching accuracy threshold;

[0059] The thermal response module is used to respond to the adjustment signal or the matching signal, and immediately display the preset warning text corresponding to the adjustment signal or the matching signal, so as to adjust and optimize the registration process of the optical feature image and the thermal infrared feature image according to the text feedback to improve the registration accuracy and reliability.

[0060] Example 2:

[0061] The thermal model construction and analysis module is used to construct a three-dimensional heat conduction model of the PCB board and perform thermal management analysis and calibration operations to obtain thermal analysis labels. The specific thermal management analysis and calibration operation process is as follows:

[0062] A 3D thermal conduction model of the PCB is constructed based on the PCB design drawings and device-temperature diagrams. The PCB is divided into multiple grid cells in the 3D thermal conduction model. The thermal properties, such as thermal conductivity and specific heat capacity, are set for each grid cell. Environmental conditions and constraints are also set. Environmental conditions include ambient temperature and air velocity, while constraints include device power consumption and interface contact thermal resistance.

[0063] Set the simulation duration, and use the 3D heat conduction model to obtain the power consumption of each device during the simulation duration. Set the device power consumption as the internal heat source, and then obtain the heat conduction process within the PCB during the simulation of each internal heat source.

[0064] Based on the three-dimensional heat conduction model, the temperature distribution characteristic images and heat flux density distribution characteristic images of the PCB board at different time points are obtained through simulation;

[0065] The temperature distribution characteristic image and the heat flux density distribution characteristic image are input as input layers to a pre-set thermal defect recognition model to obtain the output results of the thermal defect recognition model. The output results include defect type and defect hazard score (0-100). The defect type includes layout defect, material defect, etc.

[0066] At the same time, the temperature value of each component in the three-dimensional heat conduction model during the simulation time is obtained, the component whose temperature value exceeds the preset temperature value threshold is obtained, and the component whose temperature value exceeds the preset temperature value threshold is set as a risk component. At the same time, the duration corresponding to the temperature value of the risk component exceeds the preset temperature value threshold is obtained, and the risk component whose duration exceeds the preset duration threshold is set as an overheating component;

[0067] The overheating device, defect type, and defect hazard score are set as thermal analysis tags. The thermal response module is used to respond to the thermal analysis tag and immediately display the preset warning text corresponding to the thermal analysis tag. This not only helps to intuitively understand the PCB board thermal analysis feedback results, but also allows for targeted management of PCB boards based on the thermal analysis tags.

[0068] The improvement evaluation module is used to evaluate and analyze the improvement effect of the improvement parameters in the pre-generated improvement plan of the PCB board, and to judge whether the improvement plan of the PCB board meets the standards. The specific improvement effect evaluation feedback analysis process is as follows:

[0069] Based on the thermal analysis tag, the improvement parameters of the pre-generated improvement plan for the PCB board are obtained. The improvement parameters of the pre-generated improvement plan for the PCB board are reset in the 3D heat conduction model to obtain the improvement effect parameters of the overheating components before and after optimization. The improvement effect parameters include the temperature reduction value and the heat flux density peak reduction.

[0070] For example: the improved parameter is to increase the size of the heat sink (such as heat sink material = copper alloy, size = 30mm×30mm×5mm);

[0071] Assign a preset weight factor coefficient to each parameter in the improvement effect parameter, multiply each parameter in the simulation effect parameter by the corresponding weight factor coefficient, set the sum of the multiplications of each parameter in the simulation effect parameter by the corresponding weight factor coefficient as the improvement effect score, and perform judgment processing on the improvement effect score. If the improvement effect score is less than a preset improvement effect score threshold, a failure signal is generated. If the improvement effect score is greater than or equal to the preset improvement effect score threshold, a qualified signal is generated. The thermal response module is used to respond to the failure signal or the qualified signal, and immediately display the preset warning text corresponding to the failure signal or the qualified signal. On the one hand, it helps to intuitively understand the improvement effect and feasibility of the improvement plan, and on the other hand, it helps to timely adjust the improvement parameters, such as replacing higher thermal conductivity materials, thereby improving the thermal management effect of the PCB board.

[0072] In summary, the present invention accurately identifies and locates devices on a PCB board through image processing and target detection technology, and precisely associates thermal infrared images with devices to obtain temperature data for each device, so that thermal analysis more accurately reflects the actual heat generation of the PCB board. During the association process, the image association matching accuracy is analyzed to improve the image registration accuracy and reliability. Through model construction and analysis, the temperature data of each device can be analyzed in detail, and local hotspots on the PCB board can be discovered in a timely manner, that is, the local hotspot detection capability is enhanced, providing a more targeted basis for thermal management. At the same time, a thermal analysis label is obtained based on information feedback, which helps to intuitively understand the thermal analysis feedback results of the PCB board on the one hand, and to perform targeted management of the PCB board based on the thermal analysis label on the other hand. By performing simulation evaluation on the improvement of the improvement plan, it is determined whether the improvement plan of the PCB board meets the standards. The information feedback helps to intuitively understand the improvement effect and feasibility of the improvement plan.

[0073] The threshold is set for result comparison and analysis in order to determine whether it is good or bad. The value of the threshold is set based on a combination of large-scale model analysis of sample data and manual experience to enter and store it. It can also be appropriately adjusted based on seasonal or common sense influencing conditions.

[0074] The size of the coefficient is to quantify each parameter to obtain a specific numerical value, which is convenient for subsequent comparison. The size of the coefficient depends on the amount of sample data and the preliminary setting of the corresponding operating coefficient for each set of sample data by technical personnel in this field; as long as it does not affect the proportional relationship between the parameter and the quantized value.

[0075] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. A circuit board device thermal analysis system based on target detection, characterized in that: The thermal processor is connected to a thermal map generation module, a thermal model construction and analysis module, an improvement and evaluation module, and a thermal response module. The thermal map generation module is provided with an image acquisition unit and an image processing unit. The image acquisition unit is used to acquire the optical characteristic image and thermal infrared characteristic image of the PCB board, and the image processing unit is used to perform image matching and fusion analysis on the optical characteristic image and the thermal infrared characteristic image to obtain a device-temperature map; The thermal processor is used to store a device-temperature map and display the device-temperature map on a display panel; The thermal model construction and analysis module is used to construct a three-dimensional heat conduction model of the PCB board and perform thermal management analysis and calibration operations to obtain the overheating device + defect type + defect hazard score, and set the overheating device + defect type + defect hazard score as the thermal analysis label; The improvement evaluation module is used to evaluate and analyze the improvement effect of the improvement parameters in the pre-generated improvement plan of the PCB board, and to judge the improvement effect score obtained to determine whether the improvement plan of the PCB board meets the standards; The thermal management analysis calibration operation process is as follows: A three-dimensional heat conduction model of the PCB is constructed based on the PCB design drawings and device-temperature diagrams. The PCB is divided into multiple grid cells in the three-dimensional heat conduction model. The thermal conductivity and specific heat capacity of each grid cell are set. Environmental conditions and constraints are also set. Environmental conditions include ambient temperature and air flow rate, while constraints include device power consumption and interface contact thermal resistance. Set the simulation duration, and use the 3D heat conduction model to obtain the power consumption of each device during the simulation duration. Set the device power consumption as the internal heat source, and then obtain the heat conduction process within the PCB during the simulation of each internal heat source. Based on the three-dimensional heat conduction model, the temperature distribution characteristic images and heat flux density distribution characteristic images of the PCB board at different time points are obtained through simulation; The temperature distribution characteristic image and the heat flux density distribution characteristic image are input as input layers to a pre-set thermal defect recognition model to obtain the output results of the thermal defect recognition model, which include defect type and defect hazard score; At the same time, the temperature value of each component in the three-dimensional heat conduction model during the simulation time is obtained, the component whose temperature value exceeds the preset temperature value threshold is obtained, and the component whose temperature value exceeds the preset temperature value threshold is set as a risk component. At the same time, the duration corresponding to the temperature value of the risk component exceeds the preset temperature value threshold is obtained, and the risk component whose duration exceeds the preset duration threshold is set as an overheating component; Set the overheating device + defect type + defect severity score as the thermal analysis label.

2. The circuit board device thermal analysis system based on target detection according to claim 1, characterized in that: The image matching fusion analysis process is as follows: S1: Place the PCB to be analyzed in a fixed position, and use an optical camera and a thermal infrared camera to obtain the optical characteristic image and thermal infrared characteristic image of the PCB; S2: Input the optical feature image into a pre-set target detection model, which processes the optical feature image and identifies the category information and location coordinates of each device; S3: Preprocessing the optical feature image and the thermal infrared feature image, and registering the preprocessed optical feature image and the thermal infrared feature image through feature point matching and transformation matrix calculation; S4: Extracting temperature data of each device in the PCB board, including the maximum temperature, minimum temperature, and average temperature. Obtaining a device-temperature map of the PCB board based on the temperature data and position coordinates of each device.

3. The circuit board device thermal analysis system based on target detection according to claim 2, characterized in that: S3: The process of achieving registration is as follows: S31: Set the pads, vias, and device corners in the PCB board as feature points; S32: Obtain visual description features of each feature point, where the visual description features include temperature gradient direction and texture pattern; S33: Based on the visual description features of each feature point and a preset nearest neighbor algorithm, the Hamming distance of the visual description features of the optical feature image and the thermal infrared feature image is calculated, and feature point matching pairs whose Hamming distance is less than a preset Hamming distance threshold are screened.

4. The circuit board device thermal analysis system based on target detection according to claim 3, characterized in that: It also includes S34: randomly extracting n groups of feature point matching pairs, where n is a natural number greater than 3, and constructing the thermal infrared feature image spatial coordinates and the optical feature image spatial coordinates respectively, and selecting a point to be mapped in the thermal infrared feature image, and mapping the spatial coordinates of the thermal infrared feature image to the optical feature image coordinate system through the existing affine transformation technology to obtain a mapping point, and obtaining the ratio of the number of feature point matching pairs corresponding to the error value between the mapping point and the corresponding actual point being less than a preset error threshold to n, and setting the ratio of the number of feature point matching pairs corresponding to the error value between the mapping point and the actual point being less than the preset error threshold to n as the matching accuracy, and performing discrimination processing on the matching accuracy to obtain an adjustment signal or a matching signal.

5. The circuit board device thermal analysis system based on target detection according to claim 1, characterized in that: The improvement effect evaluation feedback analysis process is as follows: based on the thermal analysis tag, improvement parameters in the pre-generated improvement plan for the PCB board are obtained, and the improvement parameters in the pre-generated improvement plan for the PCB board are reset in the three-dimensional heat conduction model to obtain improvement effect parameters of the overheating components before and after optimization. The improvement effect parameters include temperature reduction value and heat flux density peak reduction.

6. The circuit board device thermal analysis system based on target detection according to claim 5, characterized in that: A preset weight factor coefficient is assigned to each parameter in the improvement effect parameters, each parameter in the simulation effect parameters is multiplied by the corresponding weight factor coefficient, and the sum of the multiplications of each parameter in the simulation effect parameters and the corresponding weight factor coefficient is set as the improvement effect score, and the improvement effect score is judged and processed to obtain an unqualified signal or a qualified signal.

Images