Component junction temperature resistance detection method and system based on natural convection heat dissipation

Abstract

This invention discloses a method and system for detecting the thermal resistance of component joints based on natural convection heat dissipation, belonging to the field of component joint thermal resistance detection technology. The method includes: collecting relevant parameters of the component joint area, mapping them using a dynamic evolution model to form a dataset, extracting features and establishing corresponding relationships using a gap thermal resistance identification algorithm, simulating the thermal resistance transfer process using an unsteady-state simulation model to generate a time-series change trajectory, constructing an association matrix by fusing features through a multi-dimensional data intelligent analysis gateway, and finally completing the quantitative characterization of thermal resistance and outputting the results through model interaction verification. This invention achieves precise advancement in parameter capture, feature extraction, trajectory optimization, and matrix construction. The various units of the system work collaboratively to form a complete technical link from parameter acquisition and data processing to result output, effectively adapting to the complex dynamic characteristics of heterogeneous interface thermal resistance in natural convection scenarios, and improving the accuracy and reliability of component joint thermal resistance detection.

Description

Technical Field

[0001] This invention relates to the field of component bonding thermal resistance detection technology, and in particular to a method and system for detecting the degree of component bonding thermal resistance based on natural convection heat dissipation. Background Technology

[0002] Natural convection, as a crucial method for heat dissipation, is widely used in various integrated component applications. The thermal resistance at component joints directly impacts heat dissipation efficiency and equipment reliability. With increasing component integration and the rise of heterogeneous material bonding scenarios, factors such as gap morphology and interface characteristics at joints lead to complex and dynamic changes in thermal resistance. Furthermore, the instability of the flow field in natural convection environments further complicates thermal resistance testing. Currently, the industry has an increasingly urgent need for correlation detection between component bonding levels and thermal resistance parameters. This necessitates the development of specialized testing solutions adapted to natural convection scenarios to accurately capture the patterns of thermal resistance changes and effectively characterize the bonding state, providing data support for equipment design optimization, fault warning, and lifespan assessment.

[0003] Existing technologies have significant shortcomings in detecting component bonding thermal resistance under natural convection scenarios: First, they fail to fully adapt to the dynamic characteristics of natural convection environments, neglecting the correlation between changes in the environmental flow field and the thermal resistance transfer at component bonding sites during the detection process, resulting in test results that fail to reflect the true thermal resistance state under actual operating conditions; Second, they lack comprehensive consideration of factors affecting thermal resistance and the ability to integrate multi-dimensional data, failing to form a complete technical link from parameter acquisition to result output, and are unable to effectively isolate the interference of complex factors such as interface characteristics and gap morphology, resulting in insufficient accuracy in characterizing the correlation between thermal resistance and component bonding degree, making it difficult to meet the requirements of high-requirement application scenarios for the reliability and accuracy of test results. Summary of the Invention

[0004] In order to overcome the shortcomings and deficiencies of the existing technology, the present invention provides a method and system for detecting the thermal resistance of component bonding based on natural convection heat dissipation.

[0005] The technical solution adopted in this invention is a method for detecting the thermal resistance of component joints based on natural convection heat dissipation, comprising the following steps: S1, collecting heat flux density, interface temperature gradient, and gap characteristic correlation parameters of the component joint under natural convection environment through a multi-dimensional thermal resistance data intelligent analysis gateway; S2, dynamically mapping the collected parameters through a heterogeneous interface thermal resistance dynamic evolution model to form a thermal resistance evolution trend correlation dataset; S3, extracting gap thermal resistance correlation features from the dataset using a natural convection joint gap thermal resistance identification algorithm to establish a correspondence between gap parameters and thermal resistance response; S4, simulating the thermal resistance transfer process of the extracted features using an unsteady-state joint thermal resistance simulation model to generate an unsteady-state thermal resistance time-series change trajectory; S5, feeding back the time-series change trajectory to the multi-dimensional thermal resistance data intelligent analysis gateway for multi-dimensional feature fusion to construct a thermal resistance-joint state correlation matrix; S6, completing the quantitative characterization of the component joint thermal resistance based on the correlation matrix, and outputting the detection result through interactive verification between the heterogeneous interface thermal resistance dynamic evolution model and the unsteady-state joint thermal resistance simulation model.

[0006] Furthermore, the expression for the dynamic evolution model of heterogeneous interface thermal resistance in S2 is: , in, For the dynamic thermal resistance of heterogeneous interfaces, For interface adaptation coefficients, This represents the percentage of the interface contact area. These are the thermal conductivity parameters of heterogeneous materials. For time variables, The evolution decay coefficient, For gap morphology parameters, This is a reference thermal resistance value.

[0007] Furthermore, the expression for the natural convection joint gap thermal resistance identification algorithm in S3 is as follows: , in, To identify the obtained gap thermal resistance, For natural convection temperature difference, These are parameters representing convective flow characteristics. These are the geometric characteristic parameters of the gap. To identify the weighting coefficients, This is a parameter for heat transfer efficiency. This is the initial value of the gap thermal resistance. It is a time variable.

[0008] Furthermore, the expression for the unsteady-state bonding thermal resistance simulation model in S4 is as follows: , in, These are unsteady-state thermal resistance simulation values. To simulate thermal conductivity, The second derivative of the temperature gradient at the joint is given. For simulation correction coefficients, For the bonding tightness parameter, For the dynamic thermal resistance of heterogeneous interfaces, For time variables, For the simulated attenuation coefficient, The obtained gap thermal resistance is used for identification.

[0009] Furthermore, the multi-dimensional feature fusion expression used by the thermal resistance multi-dimensional data intelligent analysis gateway in S5 is as follows: , in, To fuse feature values, These are the fusion weights for the unsteady-state thermal resistance simulation value, the identified gap thermal resistance, and the dynamic thermal resistance, respectively. These are unsteady-state thermal resistance simulation values. To identify the obtained gap thermal resistance, For the dynamic thermal resistance of heterogeneous interfaces, These are the correlation coefficients for multi-dimensional parameters.

[0010] Furthermore, the thermal resistance quantification characterization expression in S6 is as follows: , in, The results are for the quantitative characterization of thermal resistance. For the dynamic thermal resistance of heterogeneous interfaces, To identify the obtained gap thermal resistance, These are unsteady-state thermal resistance simulation values. For time variables, For the bonding tightness parameter, These are the correlation coefficients for multi-dimensional parameters.

[0011] Further, step S3 includes the following steps: S31, synchronously capturing the heat flux density and interface temperature gradient at the joint using a natural convection joint gap thermal resistance identification algorithm to establish a heat flux-temperature correlation dataset; S32, extracting thermal resistance response features corresponding to gap width and contact flatness based on the dataset to form a feature vector sequence; S33, associating and matching the feature vector sequence with the evolution trend data output by the heterogeneous interface thermal resistance dynamic evolution model to filter gap thermal resistance sensitive features; S34, strengthening the filtered features through feature sensitivity weight allocation to construct a one-to-one correspondence between gap parameters and thermal resistance response.

[0012] Further, S4 includes the following steps: S41, the unsteady-state interface thermal resistance simulation model receives the gap thermal resistance sensitive characteristic data and sets the boundary conditions corresponding to the natural convection environment; S42, based on the boundary conditions, it simulates the thermal resistance transmission path and energy loss process between heterogeneous interfaces to generate instantaneous thermal resistance data; S43, it arranges the instantaneous thermal resistance data in a time series and corrects the thermal resistance transmission trajectory by combining the natural convection heat dissipation characteristics; S44, through interactive feedback with the heterogeneous interface thermal resistance dynamic evolution model, it optimizes the dynamic adaptability of the simulation trajectory and forms the final unsteady-state thermal resistance time series change trajectory.

[0013] Further, S5 includes the following steps: S51, the intelligent analysis gateway for multi-dimensional thermal resistance data receives the non-steady-state thermal resistance time-series change trajectory and the corresponding relationship data of gap thermal resistance; S52, the two types of data are decomposed into dimensions, and the correlation features of time dimension, spatial dimension and characteristic dimension are extracted; S53, the multi-dimensional features are integrated using feature fusion rules to eliminate data redundancy; S54, a thermal resistance-joint state correlation matrix is ​​constructed based on the integrated features to ensure that each element in the matrix corresponds to a unique combination relationship between thermal resistance and joint state.

[0014] A component joint thermal resistance detection system based on natural convection heat dissipation is applied to a component joint thermal resistance detection method based on natural convection heat dissipation. The system includes: a natural convection environment parameter acquisition unit, used to capture heat flux density, interface temperature gradient, and gap characteristic correlation parameters at the component joint, and transmit the acquired data to a heterogeneous interface thermal resistance dynamic evolution processing unit; a heterogeneous interface thermal resistance dynamic evolution processing unit, which receives the acquired data and performs dynamic correlation mapping through a heterogeneous interface thermal resistance dynamic evolution model, outputting a thermal resistance evolution trend correlation dataset to a natural convection gap thermal resistance identification unit; and a natural convection gap thermal resistance identification unit, which uses a natural convection joint gap thermal resistance identification algorithm. The thermal resistance correlation features of the gap are extracted, a corresponding relationship is established, and the data is transmitted to the unsteady-state thermal resistance simulation unit. The unsteady-state thermal resistance simulation unit simulates the thermal resistance transfer process through the unsteady-state bonding thermal resistance simulation model, generates a time-series change trajectory, and sends it to the multi-dimensional data intelligent analysis gateway unit. The multi-dimensional data intelligent analysis gateway unit performs multi-dimensional feature fusion on the received time-series change trajectory, constructs a thermal resistance-bonding state correlation matrix, and transmits the matrix to the thermal resistance quantification characterization output unit. The thermal resistance quantification characterization output unit completes the thermal resistance quantification of the component bonding degree based on the correlation matrix, and outputs the detection results through the interactive verification between the heterogeneous interface thermal resistance dynamic evolution model and the unsteady-state bonding thermal resistance simulation model.

[0015] Beneficial Effects: This invention proposes a method and system for detecting the thermal resistance of component joints based on natural convection heat dissipation. Through multi-dimensional parameter acquisition and dynamic correlation mapping, it accurately captures the correlation between changes in the environmental flow field and the thermal resistance of component joints, achieving a comprehensive presentation of the thermal resistance change trajectory under unsteady-state conditions, making the detection results more consistent with actual application scenarios. Addressing the shortcomings of existing technologies in lacking multi-factor integration and precise correlation capabilities, this invention establishes a precise correspondence between thermal resistance and component joint state through gap thermal resistance feature extraction, thermal resistance transfer process simulation, and multi-dimensional feature fusion. This effectively eliminates the interference of complex factors such as interface characteristics and gap morphology, significantly improving the accuracy and reliability of thermal resistance quantification. Simultaneously, through an interactive verification mechanism of different technical modules, the stability of the detection results is further enhanced, forming a complete technical architecture from parameter acquisition and feature processing to result output. This not only meets the stringent requirements of high-integration components for thermal resistance detection accuracy but also provides more reliable data support for equipment heat dissipation performance optimization, fault warning, and life assessment, adapting to the detection needs of complex application scenarios such as heterogeneous material joints. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the overall process of the method of the present invention. Figure 2 This is a flowchart of method step S3 of the present invention; Figure 3 This is a flowchart of method step S4 of the present invention; Figure 4 This is a flowchart of step S5 of the method of the present invention; Figure 5 This is a diagram showing the system unit composition of the present invention. Detailed Implementation

[0017] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0018] like Figure 1As shown, a method for detecting the thermal resistance of component joints based on natural convection heat dissipation is characterized by the following steps: S1, collecting heat flux density, interface temperature gradient, and gap characteristic correlation parameters of the component joint under natural convection environment through a multi-dimensional thermal resistance data intelligent analysis gateway; S2, dynamically mapping the collected parameters through a heterogeneous interface thermal resistance dynamic evolution model to form a thermal resistance evolution trend correlation dataset; S3, extracting gap thermal resistance correlation features from the dataset using a natural convection joint gap thermal resistance identification algorithm to establish a correspondence between gap parameters and thermal resistance response; S4, simulating the thermal resistance transfer process of the extracted features using an unsteady-state joint thermal resistance simulation model to generate an unsteady-state thermal resistance time-series change trajectory; S5, feeding back the time-series change trajectory to the multi-dimensional thermal resistance data intelligent analysis gateway for multi-dimensional feature fusion to construct a thermal resistance-joint state correlation matrix; S6, completing the quantitative characterization of the component joint thermal resistance based on the correlation matrix, and outputting the detection result through interactive verification between the heterogeneous interface thermal resistance dynamic evolution model and the unsteady-state joint thermal resistance simulation model.

[0019] Step S1 involves collecting relevant parameters of the component joint under natural convection conditions through a multi-dimensional thermal resistance data intelligent analysis gateway. This gateway integrates a multi-channel sensor data receiving module and a real-time data transmission module, and has the ability to simultaneously capture multiple thermal and structural characteristic parameters. During the data acquisition process, a high-precision heat flux sensor array was deployed in a distributed manner for the heat flux density parameters at the component joints. The sensor spacing was set to 2 mm, covering the entire joint interface area, and the sampling frequency was set to 100 Hz to ensure the capture of instantaneous changes in heat flux density. For the interface temperature gradient parameters, temperature sensing nodes were deployed on both sides of the heterogeneous material joint surface and inside the gap. The number of nodes was determined according to the component joint area, with one sensing node set per square centimeter. Temperature data at different locations were collected synchronously, and the temperature gradient was calculated by the temperature difference between adjacent nodes. The acquisition of gap feature correlation parameters was carried out using micro-displacement sensing technology and interface morphology scanning technology. The measurement accuracy of the micro-displacement sensor was controlled within 0.1 micrometers, and the scanning frequency was 50 Hz. Parameters such as gap width, gap uniformity, and interface contact point distribution density were acquired in real time. At the same time, auxiliary parameters such as airflow velocity and ambient temperature in the natural convection environment were recorded. All acquired parameters were aggregated through the high-speed data transmission interface of the gateway, with a transmission rate of 1 Gbps, ensuring that there was no data loss or delay during the acquisition process, providing comprehensive and accurate raw data support for subsequent thermal resistance analysis.

[0020] Step S2 uses a dynamic evolution model of heterogeneous interface thermal resistance to dynamically correlate and map various collected parameters. The model incorporates a database of heterogeneous material thermal properties and an interface evolution rule base, enabling it to adaptively adjust the mapping logic based on different material combinations and bonding conditions. During implementation, the collected heat flux density, interface temperature gradient, gap characteristic parameters, and environmental auxiliary parameters are first imported into the model according to a preset data format. The model performs time-series alignment processing on various parameters, with alignment accuracy controlled within 1 millisecond to ensure consistency in the time dimension between parameters. Then, based on the fundamental principles of heat conduction and interface evolution theory, nonlinear correlation equations are established between the parameters. These equations incorporate inherent property parameters such as material thermal conductivity, interface contact stiffness, and gap medium thermal conductivity, which are automatically matched and called from the model database. Finally, the model uses a dynamic iterative calculation method to refine the correlation equations. The iteration step size is adaptively adjusted according to the parameter change rate, ranging from 0.01 to 0.1 seconds. The thermal resistance evolution coefficient is updated after each iteration, and the iteration continues until the preset convergence condition is met. The convergence threshold is set to 0.001. Finally, a thermal resistance evolution trend association dataset is formed through correlation mapping, which includes multi-dimensional information such as time series, parameter values, and thermal resistance evolution coefficient. The dataset is stored in the order of timestamps, and each data record includes more than 20 feature indicators, which fully presents the dynamic law of the thermal resistance of heterogeneous interfaces changing with various parameters, providing a structured data foundation for subsequent gap thermal resistance feature extraction.

[0021] Step S3 utilizes a natural convection junction gap thermal resistance identification algorithm to extract gap thermal resistance correlation features from the thermal resistance evolution trend-related dataset. The algorithm integrates core functions such as feature selection, pattern recognition, and parameter matching, enabling precise location of features strongly correlated with gap thermal resistance. During implementation, the associated dataset is first cleaned to remove outliers and noise. Outlier identification uses the 3σ criterion, where values ​​exceeding the mean ± 3 standard deviations are considered outliers. Missing data is filled using linear interpolation with an accuracy of 0.001. Subsequently, a combination of principal component analysis and mutual information entropy calculation is used for initial feature selection, identifying candidate features with a correlation coefficient greater than 0.8. These candidate features include 15 indicators such as gap width change rate, peak interface temperature gradient, and heat flux density fluctuation amplitude. Finally, a support vector machine algorithm is used to construct... A feature classification model was constructed, dividing candidate features into highly sensitive, moderately sensitive, and weakly sensitive features. The classification weight of highly sensitive features was set to 0.7, moderately sensitive features to 0.2, and weakly sensitive features to 0.1. The influence of key features was strengthened through weight allocation. Finally, based on the classified features, a correspondence between gap parameters and thermal resistance response was established. The correspondence data was stored in matrix form, with the row dimension representing the gap parameter type and the column dimension representing the thermal resistance response index. Each matrix element represents the correlation strength between the corresponding parameter and the index, achieving a precise mapping between gap parameters and thermal resistance response.

[0022] Step S4 employs an unsteady-state joint thermal resistance simulation model to simulate the thermal resistance transfer process of the extracted gap thermal resistance correlation features. The model, built upon the finite element analysis principle, can recreate the dynamic transfer behavior of thermal resistance in heterogeneous interfaces and gaps under natural convection conditions. During implementation, firstly, a three-dimensional simulation model is established based on the actual dimensions and geometry of the component joint structure. Tetrahedral elements are used for mesh generation, with mesh density adjusted according to structural complexity. The mesh size ranges from 0.1 to 1 mm to ensure a balance between simulation accuracy and computational efficiency. Subsequently, the extracted highly sensitive gap thermal resistance features are used as model input parameters, and boundary conditions corresponding to the natural convection environment are set, including ambient temperature, airflow velocity, and thermal radiation coefficient. The boundary conditions are dynamically updated according to the collected actual environmental parameters at a frequency of 10 Hz. Finally, the model uses an explicit time integration method to calculate the thermal resistance transfer process, with a time step set to 0.00. For 5 seconds, the thermal resistance values ​​of each node at the junction are calculated by solving the heat conduction equation and the convection heat transfer equation at each time step, while the thermal resistance transfer path and energy loss distribution are recorded. During the simulation, the thermal resistance change trend is monitored in real time. When the thermal resistance change rate is less than 0.01%, it is determined that the steady state stage has been reached and the simulation calculation is stopped. Finally, an unsteady thermal resistance time series change trajectory is generated, which includes thermal resistance values ​​at each time step, transfer path coordinates, energy loss ratio and other information. The trajectory data is arranged in chronological order, and a key data point is extracted every 100 time steps to form a simplified but accurate time series trajectory curve, which provides dynamic thermal resistance data for subsequent feature fusion.

[0023] Step S5 feeds back the unsteady thermal resistance time-series change trajectory to the thermal resistance multi-dimensional data intelligent analysis gateway for multi-dimensional feature fusion. The gateway has a built-in data fusion engine and feature processing module, supporting efficient integration and in-depth analysis of multi-source data. During implementation, the gateway first receives the unsteady thermal resistance time-series change trajectory data and the previously established gap parameter-thermal resistance response correspondence data. The data format conversion module unifies the two types of data into a standardized format, retaining data precision to four decimal places. Then, the multi-dimensional feature extraction process is initiated, extracting features such as the thermal resistance peak occurrence time, thermal resistance stabilization time, and thermal resistance change cycle from the time dimension; features such as the maximum, minimum, average, and standard deviation of thermal resistance from the numerical dimension; and features such as the rate of increase, rate of decrease, and fluctuation frequency of thermal resistance from the trend dimension, extracting a total of 25 multi-dimensional feature indicators. Next, a weighted fusion algorithm is used to integrate the extracted features, and the thermal resistance detection results are analyzed based on these features. The influence of each feature is assigned a fusion weight, which is determined by the analytic hierarchy process and ranges from 0.02 to 0.15. Key features such as peak thermal resistance, settling time, and rise rate have weights higher than 0.1. During the fusion process, a consistency verification mechanism is used to ensure the reliability of the fusion results. The verification threshold is set at 0.005. When the consistency deviation of the fusion results exceeds the threshold, the weights are readjusted and the fusion calculation is performed again. Finally, a thermal resistance-bonding state correlation matrix is ​​constructed based on the integrated features. The matrix has a dimension of 100×100, and the matrix elements represent the matching degree between a specific combination of thermal resistance features and the corresponding bonding state. The matching degree ranges from 0 to 1. This matrix enables a precise correlation between thermal resistance features and bonding states.

[0024] Step S6 completes the quantitative characterization of component bonding thermal resistance based on the thermal resistance-bonding state correlation matrix. The detection results are output through interactive verification between the heterogeneous interface thermal resistance dynamic evolution model and the unsteady-state bonding thermal resistance simulation model, forming a closed-loop link for data processing and result verification. During implementation, firstly, based on the matching degree values ​​in the correlation matrix, the fuzzy comprehensive evaluation method is used to quantitatively calculate the component bonding thermal resistance. Five quantization levels are set, each corresponding to a specific matching degree interval and thermal resistance range. The quantization level division adopts an equidistant division principle, and the thermal resistance difference between adjacent levels remains consistent. Then, the quantization calculation results are transmitted to the heterogeneous interface thermal resistance dynamic evolution model. The model derives the corresponding parameter combinations based on the quantization results, compares them with the original collected parameters, and calculates the parameter deviation rate. The deviation rate calculation formula is the difference between the derived parameters and the original parameters divided by the original parameters. Simultaneously, the quantization results are transmitted to the unsteady-state bonding thermal resistance simulation model. The model performs reverse simulation based on the quantization results, generating a reverse-simulated thermal resistance trajectory. Similarity is calculated between this trajectory and the original simulation trajectory using cosine similarity, with values ​​ranging from 0 to 1. The quantization result is considered valid if the parameter deviation rate is less than 5% and the trajectory similarity is greater than 0.95; otherwise, the correlation matrix weights are readjusted for further quantization. The final output includes detection results such as thermal resistance quantization values, bonding level, verification deviation rate, and similarity values. These results are presented in a standardized report format, including complete information such as data source, processing flow, quantization basis, and verification results, ensuring the reliability and traceability of the detection results.

[0025] Preferably, the expression for the dynamic evolution model of heterogeneous interface thermal resistance in S2 is: , in, For the dynamic thermal resistance of heterogeneous interfaces, For interface adaptation coefficients, This represents the percentage of the interface contact area. These are the thermal conductivity parameters of heterogeneous materials. For time variables, The evolution decay coefficient, For gap morphology parameters, This is a reference thermal resistance value.

[0026] Specifically, the dynamic evolution model of heterogeneous interface thermal resistance is constructed based on heat conduction theory and the evolution law of heterogeneous interfaces. It is clear that the dynamic change of heterogeneous interface thermal resistance is jointly dominated by the evolution of interface contact state and the change of gap characteristics. Therefore, the interface fit coefficient, the proportion of interface contact area, and the thermal conductivity parameters of heterogeneous materials are selected as core influencing factors. The dynamic evolution rate of thermal resistance over time is characterized by the partial derivative with respect to the time variable. At the same time, the evolution decay coefficient, gap morphology parameters, and reference thermal resistance benchmark value are introduced to reflect the inhibitory effect of gap change on thermal resistance evolution. The square root form is used to integrate this part of the influence to reflect the nonlinear relationship between gap parameters and reference thermal resistance, and finally a complete model expression is formed. The interface adaptation coefficient is determined based on the type of heterogeneous material combination, ranging from 0.1 to 0.9. The interface contact area ratio is obtained through interface morphology scanning, ranging from 0.3 to 1.0. The thermal conductivity parameters of the heterogeneous materials are determined based on measured thermophysical properties. The evolution attenuation coefficient is adjusted based on the stability of the natural convection environment, ranging from 0.05 to 0.3. The gap morphology parameters are collected through micro-displacement sensing technology. The reference thermal resistance benchmark value is set based on the measured statistical results of the thermal resistance of similar component joints. This model accurately captures the dynamic change law of the thermal resistance of heterogeneous interfaces under natural convection environment. During implementation, real-time parameters are obtained through a multi-dimensional thermal resistance data intelligent analysis gateway and substituted into the model to calculate the dynamic thermal resistance value, providing basic data for subsequent gap thermal resistance feature extraction. This solves the defect that traditional static models cannot reflect the time-series evolution characteristics of thermal resistance and improves the accuracy and dynamic adaptability of thermal resistance correlation datasets.

[0027] Preferably, the expression for the natural convection joint gap thermal resistance identification algorithm in S3 is: , in, To identify the obtained gap thermal resistance, For natural convection temperature difference, These are parameters representing convective flow characteristics. These are the geometric characteristic parameters of the gap. To identify the weighting coefficients, This is a parameter for heat transfer efficiency. This is the initial value of the gap thermal resistance. It is a time variable.

[0028] Specifically, the natural convection junction gap thermal resistance identification algorithm model is derived based on the principle of natural convection heat transfer and the characteristics of gap thermal resistance transfer. It combines the fact that gap thermal resistance is directly related to the convection temperature difference, flow characteristics, and gap geometry under natural convection conditions, and is also affected by heat transfer efficiency and initial thermal resistance state. Therefore, the product of natural convection temperature difference, convection flow characteristic parameters, and gap geometric characteristic parameters is used as the numerator to reflect the positive factors that promote the formation of gap thermal resistance. The integral of the product of identification weight coefficient, heat transfer efficiency parameter, and initial value of gap thermal resistance over time is used as the denominator to reflect the weakening effect of heat transfer process on gap thermal resistance. The gap thermal resistance identification expression is constructed by the ratio of the numerator and denominator. The natural convection temperature difference is calculated using the interface temperature gradient. Convection flow characteristic parameters are determined based on measured airflow velocities in the natural convection environment, ranging from 0.2 to 0.8. Gap geometric characteristic parameters are acquired using interface scanning technology, ranging from 0.1 to 1.0. Identification weight coefficients are determined based on feature sensitivity analysis, ranging from 0.4 to 0.9. Heat transfer efficiency parameters are calculated using measured heat flux density results, ranging from 0.3 to 0.9. The initial value of gap thermal resistance is determined through pre-detection. This model achieves a precise correlation between gap parameters and thermal resistance response. During implementation, a dataset of thermal resistance evolution trends is received and substituted into the model to extract gap thermal resistance correlation features. This solves the problems of traditional algorithms struggling to isolate interference from the natural convection environment and accurately identify gap thermal resistance, providing reliable algorithmic support for establishing the correspondence between gap parameters and thermal resistance response, and ensuring the relevance and accuracy of feature extraction results.

[0029] Preferably, the expression for the unsteady-state bonding thermal resistance simulation model in S4 is: , in, These are unsteady-state thermal resistance simulation values. To simulate thermal conductivity, The second derivative of the temperature gradient at the joint is given. For simulation correction coefficients, For the bonding tightness parameter, For the dynamic thermal resistance of heterogeneous interfaces, For time variables, For the simulated attenuation coefficient, The obtained gap thermal resistance is used for identification.

[0030] Specifically, the unsteady-state joint thermal resistance simulation model is based on finite element thermal analysis theory and unsteady-state heat transfer laws. It fully considers the multi-factor coupling characteristics of thermal resistance transfer under unsteady conditions. The product of the second derivative of the temperature gradient at the joint and the simulated thermal conductivity is used as a basic term to reflect the contribution of the heat conduction process to the thermal resistance. The partial derivative of the product of the simulation correction coefficient, the joint tightness parameter and the dynamic thermal resistance of the heterogeneous interface with respect to time is introduced to characterize the influence of the dynamic changes in the joint state on the thermal resistance transfer. The product of the simulation attenuation coefficient and the identified gap thermal resistance is subtracted to reflect the inhibitory effect of the gap thermal resistance on the overall thermal resistance transfer. The unsteady-state thermal resistance simulation expression is constructed by superimposing and subtracting the three terms. The simulated thermal conductivity coefficient is set based on the thermal conductivity characteristics of the component materials, ranging from 0.3 to 1.0. The simulation correction coefficient is determined based on the deviation analysis between the simulation results and measured data, ranging from 0.2 to 0.7. The bonding tightness parameter is obtained through interface contact state detection, ranging from 0.3 to 0.9. The simulation attenuation coefficient is adjusted according to the stability of the unsteady process, ranging from 0.1 to 0.5. This model recreates the real process of thermal resistance transfer under unsteady conditions. During implementation, it receives the extracted gap thermal resistance correlation features, sets natural convection boundary conditions, and inputs them into the model for simulation calculation, generating the unsteady thermal resistance time-series change trajectory. This overcomes the limitation of steady-state simulation in reflecting the dynamic changes of thermal resistance, accurately simulating the evolution trajectory of thermal resistance in the time dimension, and providing high-quality dynamic thermal resistance data for subsequent multi-dimensional feature fusion.

[0031] Preferably, the multi-dimensional feature fusion expression used by the thermal resistance multi-dimensional data intelligent analysis gateway in S5 is: , in, To fuse feature values, These are the fusion weights for the unsteady-state thermal resistance simulation value, the identified gap thermal resistance, and the dynamic thermal resistance, respectively. These are unsteady-state thermal resistance simulation values. To identify the obtained gap thermal resistance, For the dynamic thermal resistance of heterogeneous interfaces, These are the correlation coefficients for multi-dimensional parameters.

[0032] Specifically, the multi-dimensional feature fusion model of the thermal resistance multi-dimensional data intelligent analysis gateway is derived based on data fusion theory and weight allocation principles. Considering that unsteady-state thermal resistance simulation values, identified gap thermal resistance, and heterogeneous interface dynamic thermal resistance are the core dimensions affecting thermal resistance detection results, a weighted summation method is used to integrate these three core thermal resistance parameters. To eliminate the influence of differences in the dimensions of different parameters on the fusion results, the square root of the sum of the squares of each weight is introduced as a normalization factor. Simultaneously, a multi-dimensional parameter correlation coefficient is multiplied to strengthen the intrinsic correlation between the parameters, ultimately forming the feature fusion expression. The fusion weights of the unsteady-state thermal resistance simulation values, identified gap thermal resistance, and heterogeneous interface dynamic thermal resistance are determined using the analytic hierarchy process (AHP), with each weight ranging from 0.2 to 0.5, and the sum of the weights being 1.0. The multi-dimensional parameter correlation coefficient is calculated based on the correlation analysis of each dimension parameter, with a value range of 0.6 to 0.95. This model achieves efficient integration and redundancy elimination of multi-source thermal resistance data. During implementation, the intelligent analysis gateway for multi-dimensional thermal resistance data receives the unsteady thermal resistance time-series change trajectory and substitutes it into the model to complete multi-dimensional feature fusion. Its core significance lies in solving the problems of isolated parameters and low information utilization in traditional data processing. By fusion, it strengthens key features and weakens interference factors, providing comprehensive and accurate fused feature data for constructing the thermal resistance-junction state correlation matrix.

[0033] Preferably, the thermal resistance quantification expression in S6 is: , in, The results are for the quantitative characterization of thermal resistance. For the dynamic thermal resistance of heterogeneous interfaces, To identify the obtained gap thermal resistance, These are unsteady-state thermal resistance simulation values. For time variables, For the bonding tightness parameter, These are the correlation coefficients for multi-dimensional parameters.

[0034] Specifically, the thermal resistance quantification model is based on quantitative analysis theory and the correlation law between thermal resistance and bonding state. The quantification results need to comprehensively reflect the synergistic effect of the dynamic thermal resistance of the heterogeneous interface, the identified gap thermal resistance, and the simulated values ​​of unsteady-state thermal resistance. This is directly related to bonding tightness and is also affected by the correlation of multiple dimensions of parameters. Therefore, the integral of the product of the dynamic thermal resistance of the heterogeneous interface, the identified gap thermal resistance, and the simulated values ​​of unsteady-state thermal resistance over time is used as the basic term to reflect the cumulative effect of the three types of thermal resistance parameters over time. Multiplying this by the ratio of the bonding tightness parameter to the correlation coefficient of the multiple dimensions of parameters highlights the core influence of bonding tightness and corrects for the interference of parameter correlation, ultimately forming the thermal resistance quantification expression. The bonding tightness parameter is obtained by comprehensively calculating the interface contact area ratio and the gap width, with a value range of 0.2 to 0.9. The correlation coefficient of the multiple dimensions of parameters is consistent with weight 5, with a value range of 0.6 to 0.95. This model enables precise quantification of the thermal resistance of component bonding. During implementation, it is based on the thermal resistance-bonding state correlation matrix, which is substituted into the model to complete the thermal resistance quantification calculation. It solves the defects of the quantification method that lacks comprehensive consideration of multiple factors and insufficient quantification accuracy. By integrating key thermal resistance parameters and bonding characteristic parameters, it achieves a precise correspondence between thermal resistance and component bonding degree, providing a reliable quantitative basis for the output of test results.

[0035] Preferred, such as Figure 2 As shown, step S3 includes the following steps: S31, synchronously capturing the heat flux density and interface temperature gradient at the joint using a natural convection joint gap thermal resistance identification algorithm to establish a heat flux-temperature correlation dataset; S32, extracting thermal resistance response features corresponding to gap width and contact flatness based on the dataset to form a feature vector sequence; S33, associating and matching the feature vector sequence with the evolution trend data output by the heterogeneous interface thermal resistance dynamic evolution model to filter gap thermal resistance sensitive features; S34, strengthening the filtered features through feature sensitivity weight allocation to construct a one-to-one correspondence between gap parameters and thermal resistance response.

[0036] Specifically, step S3 achieves accurate extraction and establishment of the correlation between thermal resistance features of natural convection joint gaps through four sub-steps. In sub-step S31, the parameter capture module of the general natural convection joint gap thermal resistance identification algorithm synchronously collects and records the heat flux density and interface temperature gradient at the joint. The collection frequency is set to 100 Hz to ensure strict alignment of the two parameters in the time dimension. A heat flux-temperature correlation dataset, including timestamps, heat flux density values, and interface temperature gradient values, is established using data synchronization storage technology. The dataset is arranged in chronological order, with each data record retaining four decimal places of precision. In sub-step S32, based on this dataset, a feature extraction algorithm is used to explore the response relationship between the two core gap parameters—gap width and contact flatness—and changes in thermal resistance. Key response features such as peak thermal resistance, rate of change, and fluctuation amplitude are extracted. These features are arranged according to preset dimensions to form a feature vector sequence. Each feature vector includes 15 feature dimensions, with vector elements ranging from 0 to 1. In step S33, the thermal resistance evolution trend data output by the heterogeneous interface thermal resistance dynamic evolution model is called. A data association algorithm is used to match the feature vector sequence with the evolution trend data point by point, calculate the correlation coefficient between the two, and set the correlation coefficient threshold to 0.8. Sensitive features of gap thermal resistance with correlation coefficients higher than the threshold are selected. In step S34, based on the influence of the features on the thermal resistance detection results, the sensitivity weight of each sensitive feature is determined by the analytic hierarchy process (AHP). The weight values ​​range from 0.05 to 0.25. The selected sensitive features are then weighted and strengthened. Based on the strengthened feature data, a one-to-one correspondence table between gap parameters and thermal resistance response is constructed to achieve accurate mapping between gap parameters and thermal resistance response. This provides high-quality feature data for subsequent unsteady thermal resistance simulation. The significance of this weighting step is to improve the targeting and accuracy of feature extraction by refining the process, and to ensure the reliability of the correlation between gap parameters and thermal resistance response.

[0037] Preferred, such as Figure 3 As shown, S4 includes the following steps: S41, the unsteady-state interface thermal resistance simulation model receives the gap thermal resistance sensitive characteristic data and sets the boundary conditions corresponding to the natural convection environment; S42, based on the boundary conditions, it simulates the thermal resistance transmission path and energy loss process between heterogeneous interfaces to generate instantaneous thermal resistance data; S43, it arranges the instantaneous thermal resistance data in a time series and corrects the thermal resistance transmission trajectory by combining the natural convection heat dissipation characteristics; S44, through interactive feedback with the heterogeneous interface thermal resistance dynamic evolution model, it optimizes the dynamic adaptability of the simulation trajectory and forms the final unsteady-state thermal resistance time series change trajectory.

[0038] Specifically, step S4 achieves accurate simulation and temporal trajectory generation of the unsteady-state junction thermal resistance transfer process through four consecutive sub-steps, ensuring a high degree of consistency between the simulation results and actual operating conditions. In sub-step S41, the unsteady-state junction thermal resistance simulation model receives filtered gap thermal resistance sensitive characteristic data through a data interface. The data transmission rate is controlled at 1Gbps to ensure real-time performance. Simultaneously, based on actual monitoring data of the natural convection environment, boundary conditions such as ambient temperature, airflow velocity, and thermal radiation coefficient are set. The boundary condition parameters are dynamically updated every 10 milliseconds to ensure synchronization with actual environmental changes. In sub-step S42, based on the set boundary conditions, a finite element simulation algorithm is used to simulate the thermal resistance transfer path between heterogeneous interfaces, calculating the energy loss ratio during the transfer process. The energy loss calculation accuracy is retained to three decimal places. Instantaneous thermal resistance data is generated at each time step, with the time step set to 0.005 seconds. The instantaneous thermal resistance data includes key information such as thermal resistance value, transfer path coordinates, and energy loss ratio. In step S43, all instantaneous thermal resistance data are arranged chronologically to form the original thermal resistance transfer trajectory. Combining the flow field characteristics and heat exchange laws of natural convection heat dissipation, a trajectory correction algorithm is used to optimize the original trajectory. During the correction process, the focus is on adjusting the thermal resistance fluctuation deviation caused by changes in airflow velocity to ensure that the trajectory accurately reflects the thermal resistance variation law under natural convection conditions. In step S44, an interactive feedback channel is established between the unsteady-state interface thermal resistance simulation model and the heterogeneous interface thermal resistance dynamic evolution model. The corrected thermal resistance transfer trajectory is fed back to the heterogeneous interface thermal resistance dynamic evolution model. The model compares the consistency between the trajectory data and the thermal resistance evolution trend data, calculates the deviation rate, and when the deviation rate is greater than 3%, the simulation parameters are adjusted and the simulation is repeated until the deviation rate meets the requirements, ultimately forming the unsteady-state thermal resistance time-series change trajectory. This abstract improves the accuracy and dynamic adaptability of thermal resistance transfer simulation by refining the simulation process and feedback mechanism, providing reliable dynamic thermal resistance data for subsequent feature fusion.

[0039] Preferred, such as Figure 4 As shown, S5 includes the following steps: S51, the intelligent analysis gateway for multi-dimensional thermal resistance data receives the non-steady-state thermal resistance time-series change trajectory and the corresponding relationship data of gap thermal resistance; S52, the two types of data are decomposed into dimensions, and the correlation features of time dimension, spatial dimension and characteristic dimension are extracted; S53, the multi-dimensional features are integrated using feature fusion rules to eliminate data redundancy; S54, a thermal resistance-joint state correlation matrix is ​​constructed based on the integrated features to ensure that each element in the matrix corresponds to a unique combination relationship between thermal resistance and joint state.

[0040] Specifically, step S5 achieves efficient integration of multi-dimensional data and accurate construction of the thermal resistance-joint state correlation matrix through four sub-steps, ensuring that the matrix can comprehensively reflect the correspondence between thermal resistance and joint state. During sub-step S51, the intelligent analysis gateway for multi-dimensional thermal resistance data simultaneously receives the non-steady-state thermal resistance time-series change trajectory data and the gap thermal resistance correspondence data through a high-speed data receiving module. During data reception, a verification algorithm is used to verify data integrity, ensuring no data loss or errors. After reception, both types of data are stored in a dedicated database. The database adopts a partitioned storage strategy, classifying and managing data according to data type and time dimension. In sub-step S52, the data processing module is invoked to decompose the two types of data dimensionally. From the time dimension, features such as the occurrence time of thermal resistance peak, the duration of thermal resistance stability, and the thermal resistance change cycle are extracted; from the spatial dimension, features such as the distribution density of thermal resistance at the joint and the transmission path length are extracted; from the characteristic dimension, features such as thermal resistance fluctuation amplitude, change rate, and sensitivity coefficient are extracted, resulting in a total of 25 multi-dimensional correlation features. Each feature is quantified according to a unified standard. In step S53, based on preset feature fusion rules, a weighted fusion algorithm is used to integrate multi-dimensional features. The fusion weight is determined according to the importance of the features, with a value range of 0.02 to 0.15. During the integration process, a redundancy analysis algorithm is used to identify and eliminate duplicate or overly correlated features. The correlation threshold is set to 0.9 to ensure that the fused features are both comprehensive and concise. In step S54, based on the integrated multi-dimensional features, a thermal resistance-bonding state correlation matrix is ​​constructed. The matrix row dimension corresponds to the thermal resistance feature combination, and the column dimension corresponds to the component bonding state level. The matrix dimension is set to 100×100. Each matrix element represents the matching degree between a specific thermal resistance feature combination and the corresponding bonding state. The matching degree value ranges from 0 to 1, accurate to four decimal places. Through strict matrix construction rules, it is ensured that each element corresponds to a unique thermal resistance and bonding state combination relationship. This weighting improves the effectiveness of feature fusion and the accuracy of the correlation matrix by refining the data processing and matrix construction process, providing a solid foundation for subsequent thermal resistance quantitative characterization.

[0041] like Figure 5As shown, a component joint thermal resistance detection system based on natural convection heat dissipation is used in a component joint thermal resistance detection method based on natural convection heat dissipation. The system includes: a natural convection environment parameter acquisition unit, used to capture the heat flux density, interface temperature gradient, and gap characteristic correlation parameters at the component joint, and transmit the acquired data to a heterogeneous interface thermal resistance dynamic evolution processing unit; a heterogeneous interface thermal resistance dynamic evolution processing unit, which receives the acquired data and performs dynamic correlation mapping through a heterogeneous interface thermal resistance dynamic evolution model, outputting a thermal resistance evolution trend correlation dataset to a natural convection gap thermal resistance identification unit; and a natural convection gap thermal resistance identification unit, which uses a natural convection joint gap thermal resistance identification calculation... The method extracts the thermal resistance correlation features of the gap, establishes the corresponding relationship, and transmits it to the unsteady-state thermal resistance simulation unit. The unsteady-state thermal resistance simulation unit simulates the thermal resistance transfer process through the unsteady-state bonding thermal resistance simulation model, generates a time-series change trajectory, and sends it to the multi-dimensional data intelligent analysis gateway unit. The multi-dimensional data intelligent analysis gateway unit performs multi-dimensional feature fusion on the received time-series change trajectory, constructs a thermal resistance-bonding state correlation matrix, and transmits the matrix to the thermal resistance quantification characterization output unit. The thermal resistance quantification characterization output unit completes the thermal resistance quantification of the component bonding degree based on the correlation matrix, and outputs the detection results through the interactive verification between the heterogeneous interface thermal resistance dynamic evolution model and the unsteady-state bonding thermal resistance simulation model.

[0042] The formulas in this invention integrate different scalar and vector parameters for calculation, establishing an adaptation logic through physical meaning association and dimensional consistency processing: First, all parameters revolve around the formation, evolution, and transfer laws of component joint thermal resistance under natural convection heat dissipation scenarios, possessing clear thermal or structural correlation attributes. For example, in the formulas related to the dynamic thermal resistance of heterogeneous interfaces, the scalar interface contact area ratio and the vector heterogeneous material thermal conductivity parameter, the former reflecting the interface contact state and the latter characterizing the material's heat transfer direction characteristics, form an intrinsic correlation through the physical mechanism of heat conduction, jointly supporting the evolutionary description of dynamic thermal resistance; Second, dimensional consistency is achieved through coefficient correction and mathematical form transformation during formula construction. First, in the natural convection joint gap thermal resistance identification algorithm, the vector natural convection temperature difference (reflecting the direction of the temperature gradient) and the scalar gap geometric characteristic parameters (reflecting the size of the gap structure) are used to eliminate computational conflicts caused by dimensional differences through the dimensional adaptation of the identification weight coefficients and the mathematical transformation of integral operations. Furthermore, the directional characteristics of the vector parameters are transformed into a quantitative impact on the results through mathematical operations. For example, in the unsteady-state joint thermal resistance simulation model, the second derivative of the temperature gradient at the joint is multiplied by the scalar simulated thermal conductivity, transforming the directional information of thermal resistance transmission into a quantitative contribution of the simulated value. Simultaneously, combined with the synergistic effect of other scalar parameters, this comprehensively reflects the complex characteristics of thermal resistance transmission. This integrated approach, based on the physical correlation and mathematical adaptability between parameters, preserves the essential characteristics of each parameter while achieving the unity of computational logic, ensuring that the formula can accurately describe the complex variation law of component joint thermal resistance under natural convection conditions.

[0043] A method and system for detecting the thermal resistance of component joints based on natural convection heat dissipation addresses the problem that traditional detection methods cannot adapt to the dynamic characteristics of natural convection. This solution fully captures the intrinsic relationship between changes in the environmental flow field and the thermal resistance of component joints through multi-dimensional parameter acquisition and dynamic correlation processing. It completely restores the temporal variation law of thermal resistance under unsteady conditions, making the detection results highly consistent with actual application scenarios and effectively solving the pain point of disconnect between static detection and dynamic conditions. To address the lack of a precise correlation mechanism between thermal resistance and joint degree in existing technologies, a complete technical link covering parameter acquisition, feature processing, and simulation verification is established through gap feature extraction, thermal resistance transfer simulation, and multi-dimensional data fusion. This effectively removes complex factors such as interface characteristics and gap morphology, significantly improving the accuracy of thermal resistance quantitative characterization.

[0044] This method and system enhance the stability and reliability of detection results through interactive verification of dynamic evolution analysis and unsteady-state simulation, avoiding the limitations of a single model or algorithm. By integrating multi-dimensional information through a specialized data processing gateway, it breaks through the bottlenecks of isolated parameters and low data utilization in traditional detection, accurately mapping the correspondence between component bonding degree and thermal resistance parameters. Through targeted technical design, it overcomes the shortcomings of existing technologies in adapting to natural convection environments and solves the deficiency in thermal resistance detection accuracy, providing reliable technical support for evaluating the heat dissipation performance of highly integrated, heterogeneous material bonding components, and possessing broad scenario adaptability and practical value.

[0045] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," "link," and "fix" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0046] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various equivalent changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for detecting the thermal resistance of component bonding based on natural convection heat dissipation, characterized in that, The steps include: S1, collecting heat flux density, interface temperature gradient and gap characteristic correlation parameters at the component joint under natural convection environment through a multi-dimensional thermal resistance data intelligent analysis gateway; S2, The collected parameters are dynamically correlated and mapped using a heterogeneous interface thermal resistance dynamic evolution model to form a thermal resistance evolution trend correlation dataset; S3, The natural convection junction gap thermal resistance identification algorithm is used to extract gap thermal resistance correlation features from the dataset to establish the correspondence between gap parameters and thermal resistance response. S4, The extracted features are simulated using an unsteady-state bonding thermal resistance simulation model to generate an unsteady-state thermal resistance time-series change trajectory; S5, The time-series change trajectory is fed back to the thermal resistance multi-dimensional data intelligent analysis gateway for multi-dimensional feature fusion to construct a thermal resistance-bonding state correlation matrix; S6, Based on the correlation matrix, the thermal resistance of the component bonding degree is quantitatively characterized, and the detection results are output through the interactive verification between the heterogeneous interface thermal resistance dynamic evolution model and the unsteady-state bonding thermal resistance simulation model.

2. The method for detecting the thermal resistance of component bonding based on natural convection heat dissipation according to claim 1, characterized in that, The expression for the dynamic evolution model of heterogeneous interface thermal resistance in S2 is as follows: ， in, For the dynamic thermal resistance of heterogeneous interfaces, For interface adaptation coefficients, This represents the percentage of the interface contact area. These are the thermal conductivity parameters of heterogeneous materials. For time variables, The evolution decay coefficient, For gap morphology parameters, This is a reference thermal resistance value.

3. The method for detecting the thermal resistance of component bonding based on natural convection heat dissipation according to claim 1, characterized in that, The expression for the natural convection joint gap thermal resistance identification algorithm in S3 is as follows: ， in, To identify the obtained gap thermal resistance, For natural convection temperature difference, These are parameters representing convective flow characteristics. These are the geometric characteristic parameters of the gap. To identify the weighting coefficients, This is a parameter for heat transfer efficiency. This is the initial value of the gap thermal resistance. It is a time variable.

4. The method for detecting the thermal resistance of component bonding based on natural convection heat dissipation according to claim 1, characterized in that, The expression for the unsteady-state bonding thermal resistance simulation model in S4 is: ， in, These are unsteady-state thermal resistance simulation values. To simulate thermal conductivity, The second derivative of the temperature gradient at the joint is given. For simulation correction coefficients, For the bonding tightness parameter, For the dynamic thermal resistance of heterogeneous interfaces, For time variables, For the simulated attenuation coefficient, The obtained gap thermal resistance is used for identification.

5. The method for detecting the thermal resistance of component bonding based on natural convection heat dissipation according to claim 1, characterized in that, The multi-dimensional feature fusion expression used by the thermal resistance multi-dimensional data intelligent analysis gateway in S5 is: ， in, To fuse feature values, These are the fusion weights for the unsteady-state thermal resistance simulation value, the identified gap thermal resistance, and the dynamic thermal resistance, respectively. These are unsteady-state thermal resistance simulation values. To identify the obtained gap thermal resistance, For the dynamic thermal resistance of heterogeneous interfaces, These are the correlation coefficients for multi-dimensional parameters.

6. The method for detecting the thermal resistance of component bonding based on natural convection heat dissipation according to claim 1, characterized in that, The thermal resistance quantification expression in S6 is as follows: ， in, The results are for the quantitative characterization of thermal resistance. For the dynamic thermal resistance of heterogeneous interfaces, To identify the obtained gap thermal resistance, These are unsteady-state thermal resistance simulation values. For time variables, For the bonding tightness parameter, These are the correlation coefficients for multi-dimensional parameters.

7. The method for detecting the thermal resistance of component bonding based on natural convection heat dissipation according to claim 1, characterized in that, S3 includes the following steps: S31, synchronously capturing the heat flux density and interface temperature gradient at the joint using a natural convection joint gap thermal resistance identification algorithm to establish a heat flux-temperature correlation dataset; S32, extracting thermal resistance response features corresponding to gap width and contact flatness based on the dataset to form a feature vector sequence; S33, associating and matching the feature vector sequence with the evolution trend data output by the heterogeneous interface thermal resistance dynamic evolution model to filter gap thermal resistance sensitive features; S34, strengthening the filtered features through feature sensitivity weight allocation to construct a one-to-one correspondence between gap parameters and thermal resistance response.

8. The method for detecting the thermal resistance of component bonding based on natural convection heat dissipation according to claim 1, characterized in that, S4 includes the following steps: S41, the unsteady-state junction thermal resistance simulation model receives the gap thermal resistance sensitive characteristic data and sets the boundary conditions corresponding to the natural convection environment; S42, based on the boundary conditions, the thermal resistance transmission path and energy loss process between heterogeneous interfaces are simulated to generate instantaneous thermal resistance data; S43, the instantaneous thermal resistance data are arranged in time series and the thermal resistance transmission trajectory is corrected by combining the natural convection heat dissipation characteristics. S44 optimizes the dynamic adaptability of the simulation trajectory through interactive feedback with the dynamic evolution model of thermal resistance at heterogeneous interfaces, thus forming the final unsteady-state thermal resistance time-series change trajectory.

9. The method for detecting the thermal resistance of component bonding based on natural convection heat dissipation according to claim 1, characterized in that, S5 includes the following steps: S51, the intelligent analysis gateway for multi-dimensional thermal resistance data receives the non-steady-state thermal resistance time-series change trajectory and the corresponding relationship data of gap thermal resistance; S52, the two types of data are decomposed into dimensions, and the correlation features of time dimension, spatial dimension and characteristic dimension are extracted; S53, the multi-dimensional features are integrated using feature fusion rules to eliminate data redundancy; S54, a thermal resistance-joint state correlation matrix is ​​constructed based on the integrated features to ensure that each element in the matrix corresponds to a unique combination relationship between thermal resistance and joint state.

10. A component bonding thermal resistance detection system based on natural convection heat dissipation, characterized in that, This system is applied to the component joint thermal resistance detection method based on natural convection heat dissipation as described in claim 1, comprising: a natural convection environment parameter acquisition unit, used to capture the heat flux density, interface temperature gradient, and gap characteristic correlation parameters at the component joint, and transmit the acquired data to a heterogeneous interface thermal resistance dynamic evolution processing unit; a heterogeneous interface thermal resistance dynamic evolution processing unit, which receives the acquired data and performs dynamic correlation mapping through a heterogeneous interface thermal resistance dynamic evolution model, and outputs a thermal resistance evolution trend correlation dataset to a natural convection gap thermal resistance identification unit; and a natural convection gap thermal resistance identification unit, which extracts the gap thermal resistance using a natural convection joint gap thermal resistance identification algorithm. The system establishes a correspondence between the characteristics and transmits the data to the unsteady-state thermal resistance simulation unit. The unsteady-state thermal resistance simulation unit simulates the thermal resistance transfer process using the unsteady-state bonding thermal resistance simulation model, generates a time-series change trajectory, and sends it to the multi-dimensional data intelligent analysis gateway unit. The multi-dimensional data intelligent analysis gateway unit performs multi-dimensional feature fusion on the received time-series change trajectory, constructs a thermal resistance-bonding state correlation matrix, and transmits the matrix to the thermal resistance quantification characterization output unit. The thermal resistance quantification characterization output unit quantifies the thermal resistance of the component bonding degree based on the correlation matrix and outputs the detection results through interactive verification between the heterogeneous interface thermal resistance dynamic evolution model and the unsteady-state bonding thermal resistance simulation model.

Images