A method and device for detecting tensile strength of a heat-conducting silica gel sheet

Abstract

The application relates to the technical field of tensile strength detection, in particular to a heat-conducting silica gel sheet tensile strength detection method and device, which comprises the following steps: collecting load data, deformation images and infrared thermal images of a heat-conducting silica gel sheet sample after spraying a speckle pattern at each moment in the whole tensile process; defining a surface image of the sample before tensile as a reference image, dividing the reference image into multiple local areas, and extracting corresponding matching areas of each local area in the reference image in each moment deformation image; calculating the strain abnormality of each local area at each moment; obtaining thermal force abnormality, determining the coupling damage degree of each local area at each moment; obtaining the overall instability coefficient at each moment to determine the damage strength at each moment, and correcting the tensile strength of the heat-conducting silica gel sheet sample. The application significantly improves the scientificity and reliability of the heat-conducting silica gel sheet tensile strength detection result.

Description

Technical Field

[0001] This application relates to the field of tensile strength testing technology, specifically to a method and apparatus for testing the tensile strength of a thermally conductive silicone sheet. Background Technology

[0002] Thermally conductive silicone pads are widely used interfacial thermal conductive materials in electronic devices. Their function is to fill the tiny gaps between heat-generating components and heat dissipation structures, establishing low thermal resistance heat dissipation channels. In actual service, thermally conductive silicone pads not only need to meet thermal conductivity requirements but also need to maintain structural stability under long-term compression, tension, and thermal cycling conditions. Therefore, tensile strength, as a key indicator of its mechanical load-bearing capacity, is directly related to the long-term reliability of electronic device heat dissipation systems.

[0003] Currently, the tensile strength testing of thermally conductive silicone sheets mainly relies on universal testing machines. The macroscopic ultimate load at which the thermally conductive silicone sheet sample fractures under tension is used as the tensile strength index. This testing method is based on the assumption of macroscopic homogeneity of the material, focusing only on the overall macroscopic mechanical response of the sample. It ignores the inherent local instability characteristics of thermally conductive silicone sheets as heterogeneous materials composed of a polymer matrix and a high volume fraction of ceramic filler. In actual tensile testing, due to defects such as filler agglomeration, local micropores, or uneven interface bonding, the material often suffers irreversible damage such as strain localization, microcrack initiation and propagation before the overall load reaches its peak value. If the tensile strength is determined solely based on the macroscopic ultimate load, it will mask the premature failure of local areas. This leads to a serious overestimation of the actual safe load-bearing margin of the material by the tensile strength measured based on the homogeneity assumption, affecting the accuracy and reliability of the tensile performance test results. At the same time, if such an overestimated design index is used, it is very easy for the silicone sheet to experience sudden brittle fracture or interface peeling, causing serious thermal failure problems in electronic devices. Summary of the Invention

[0004] To address the aforementioned technical problems, a method and apparatus for testing the tensile strength of thermally conductive silicone sheets are provided.

[0005] The solution to the technical problem of this application is to provide a method and apparatus for testing the tensile strength of thermally conductive silicone sheets, including the following steps: In a first aspect, embodiments of this application provide a method for testing the tensile strength of a thermally conductive silicone sheet, the method comprising the following steps: Load data, deformation images, and infrared thermal images of the thermally conductive silicone sheet sample after the speckle pattern was sprayed were collected at various moments during the entire tensile process. The surface image of the specimen before tension is defined as the reference image, which is divided into multiple local regions. An image matching algorithm is used to extract the matching region corresponding to each local region in the deformation image at each time step in the reference image. The strain of each local region at each time step is quantified by the spatial distance variation characteristics between the local region and its matching region and the adjacent region in the image. Based on the degree of deviation of the strain of the local region and its rate of change at the same time step, the strain anomaly of each local region at each time step is calculated to characterize the degree of mechanical damage caused by strain localization and its evolution trend. For each local region, the temperature deviation level of the matching region at the corresponding region in the infrared thermal image is used to reflect the degree of abnormal temperature rise caused by local damage, thereby obtaining the thermal anomaly degree. Combined with the strain anomaly degree, the coupling damage degree of each local region at each time is determined. By analyzing the abrupt bending of the sample edge in the deformation image and the characteristics of the load data deviating from the smooth trend, the overall instability coefficient at each moment is obtained. Combined with the coupling damage degree, the damage intensity at each moment is determined, and the tensile strength of the thermally conductive silicone sheet sample is corrected accordingly.

[0006] Preferably, the step of extracting the corresponding matching region in the deformed image at each time step of each local region in the reference image using an image matching algorithm includes: obtaining the centroid of each local region in the reference image, taking its corresponding pixel in the deformed image as the search starting point, using an image matching algorithm, taking each local region in the reference image as the matching template, and performing a search and matching around the search starting point to obtain the corresponding matching region.

[0007] Preferably, the process for obtaining the strain of each local region at each time point is as follows: The remaining local regions in the reference image that are adjacent to each local region are defined as neighboring regions; the mean distance between the centroid of each local region in the reference image and the centroids of all its neighboring regions is calculated as the average reference distance. For each local region in the reference image, the corresponding matching region in the deformation image is defined as the neighboring matching region; for each deformation image at each time point, the average distance between the centroid of the corresponding matching region of each local region in the deformation image and the centroid of all neighboring matching regions is calculated as the deformation distance. The difference between the deformation distance and the average reference distance is calculated as the deformation of each local region at each time. The ratio of the deformation distance to the average reference distance is used as the strain of each local region at each time.

[0008] Preferably, the calculation of the strain anomaly degree of each local region at each time step includes: calculating the rate of change of strain of each local region at each time step relative to its previous preset time interval, as the strain rate; calculating the mean of the strain rate and the mean of the strain of all local regions at the same time step, as the average rate of change and the average strain, respectively; calculating the ratio of the strain of each local region at each time step to the average strain, and calculating the ratio of the strain rate of each local region at each time step to the average rate of change, and multiplying the two ratios as the strain anomaly degree of each local region at each time step.

[0009] Preferably, the calculation process for the thermal anomaly degree is as follows: For each local area, the corresponding matching region in the deformation image at each time step is mapped onto the synchronously acquired infrared thermal image. The mean value of the temperature data at all pixels in the mapped region is calculated as the average temperature. The mean value of the temperature data at all pixels in the infrared thermal image at each time step is calculated as the overall temperature level. The difference between the average temperature of each local area at each time and the overall temperature level is calculated as the temperature deviation. If the temperature deviation is less than 0, the thermal anomaly of each local area at each time is assigned a value of 0. Otherwise, the thermal anomaly is positively correlated with the temperature deviation.

[0010] Preferably, the coupling damage degree is positively correlated with both the strain anomaly degree and the thermodynamic anomaly degree.

[0011] Preferably, obtaining the overall instability coefficient at each time point includes: Curve fitting is performed on the load data at each time point and multiple time points prior to it, and the fitting error of the fitted curve is calculated as the load deviation at each time point. For the deformation images at each time point, the edge contour line of the thermally conductive silicone sheet sample in the deformation image is extracted; multiple edge pixels are uniformly selected on the edge contour line and defined as feature points, and the curvature at each feature point is calculated; the mean value of the difference in curvature between any two feature points is calculated and positively mapped to it as the edge anomaly degree at each time point. The overall instability coefficient is positively correlated with both load deviation and edge anomaly.

[0012] Preferably, determining the damage intensity at each moment includes: selecting the maximum coupling damage degree of all local regions at each moment; the damage intensity is positively correlated with the maximum coupling damage degree and the overall instability coefficient.

[0013] Preferably, the method for correcting the tensile strength of the thermally conductive silicone sheet sample includes: smoothing the damage strength at all times and using the smoothing results at each time as a spatiotemporal trend factor; extracting the spatiotemporal trend factor at the time when the load data reaches its peak during the entire tensile process, negatively mapping it as a correction factor; obtaining the tensile strength of the thermally conductive silicone sheet sample measured by a universal testing machine, and multiplying it by the correction factor as the final tensile strength of the thermally conductive silicone sheet sample.

[0014] Secondly, embodiments of this application also provide a device for testing the tensile strength of a thermally conductive silicone sheet, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any of the above-described methods for testing the tensile strength of a thermally conductive silicone sheet.

[0015] This application has at least the following beneficial effects: This application employs an image matching algorithm to extract the corresponding matching regions in the deformation images of each local region in the reference image at each time step, in order to quantify the strain of each local region at each time step. Its advantages lie in the fact that by dividing the reference image and accurately matching the regions, continuous tracking of the physical coordinates of the specimen throughout the tensile process is achieved, laying a spatial alignment foundation for subsequent accurate quantification of local deformation. Furthermore, by calculating the strain through the spatial position changes of the matching regions, the interference of rigid body displacement on deformation measurement can be effectively eliminated, obtaining local strain that only reflects the true tensile deformation of the material, providing a high-precision data foundation for subsequent strain anomaly identification. Calculating the strain anomaly degree of each local region at each time step has the advantage of accurately capturing minute strain concentration phenomena by comparing the deviation of local strain and strain rate from the overall average level, thereby locating structural weak points induced by internal material defects and their deterioration trends in advance. Obtaining the thermal anomaly degree has the advantage of utilizing the physical law that abnormal frictional heat energy is released when microcracks or interface slip occur locally in the material, and by quantifying the abnormal temperature rise of the local temperature relative to the global temperature, revealing the energy dissipation state of micro-damage evolution from a thermodynamic perspective, providing another dimension independent of image vision. Damage assessment criteria: Determining the coupled damage degree of each local region at each time step. Its beneficial effect lies in integrating the strain anomaly characterizing mechanical anomalies with the thermodynamic anomaly characterizing heat dissipation anomalies, comprehensively evaluating the degree of local damage from the perspective of synergistic force-thermal dual-physics field evaluation, significantly improving the accuracy and robustness of identifying and judging hidden micro-damage. Obtaining the overall instability coefficient at each time step. Its beneficial effect lies in combining the curvature abrupt changes of the edge geometry with the high-frequency oscillation characteristics of the macroscopic load curve, comprehensively characterizing the overall instability state of the specimen from two dimensions: macroscopic geometry and macroscopic mechanical response, realizing the transformation from local damage to macroscopic... The quantitative characterization of instability transformation; determining the damage intensity at each moment, and correcting the tensile strength of the thermally conductive silicone sheet sample accordingly. Its beneficial effect lies in the fusion of macroscopic overall instability and local microscopic damage, accurately capturing the dynamic cumulative effect of global structural degradation caused by the evolution of local microcracks, forcibly stripping away the false load-bearing capacity caused by premature local failure of the sample, so that the final output tensile strength index can truly reflect the material's safety physical bottom line, effectively avoiding the risk of sudden fracture caused by overestimating the strength margin in practical applications, and significantly improving the scientificity and reliability of the tensile strength test results of the thermally conductive silicone sheet. Attached Figure Description

[0016] The following is a detailed description of a method for testing the tensile strength of a thermally conductive silicone sheet according to the present application, with reference to the accompanying drawings.

[0017] Figure 1 A flowchart illustrating the steps of a method for testing the tensile strength of a thermally conductive silicone sheet, as provided in this application embodiment; Figure 2A flowchart illustrating the steps of the method for obtaining the overall instability coefficient provided in this application embodiment. Detailed Implementation

[0018] The following description, in conjunction with the accompanying drawings and embodiments, provides a more detailed explanation of the method and apparatus for testing the tensile strength of thermally conductive silicone sheets proposed in this application.

[0019] Please see Figure 1 The diagram illustrates a flowchart of a method for testing the tensile strength of a thermally conductive silicone sheet according to an embodiment of this application. The method includes the following steps: Step 1: Collect load data, deformation images, and infrared thermal images of the thermally conductive silicone sheet sample after the speckle pattern has been sprayed throughout the tensile process.

[0020] During the operation of electronic devices, thermally conductive silicone pads are subjected to tensile stresses caused by thermal expansion and contraction and assembly deformation over extended periods. If the material cracks or peels off at the interface under stress, it will interrupt the heat conduction path, leading to localized overheating or even equipment damage. Tensile strength is a fundamental indicator for assessing mechanical reliability. Therefore, accurately evaluating the tensile strength of thermally conductive silicone pads is of significant engineering importance for material selection, structural design, and lifespan prediction.

[0021] Based on the above analysis, a thermally conductive silicone sheet sample was selected for testing. Surface treatment was performed, with a high-contrast random speckle pattern uniformly sprayed onto the test area surface, followed by drying. Next, the sample with the prepared speckle pattern was clamped between the upper and lower clamps of a universal testing machine. The clamp position was adjusted to ensure the sample was vertically aligned, and a small preload was applied to eliminate initial wrinkles. Simultaneously, the longitudinal axis of the sample was ensured to be aligned with the tensile direction. The clamp was then tightened to complete the clamping process. A high-precision load sensor connected in series on the loading shaft was used to collect real-time load data of the thermally conductive silicone sheet sample at different moments throughout the tensile process. After the sample is clamped, the industrial camera is fixed in front of the sample. The lens aperture and focal length are adjusted to make the speckle pattern on the sample surface clearly imaged in the field of view. The camera optical axis is perpendicular to the sample surface. The deformation images of the thermally conductive silicone sheet sample after the speckle pattern is sprayed on the surface are continuously acquired at different times during the entire tensile process. An industrial-grade infrared thermal imager is placed in the coaxial or near-coaxial optical path of an industrial camera. The position of the thermal imager is adjusted so that its field of view completely coincides with that of the camera. Non-uniformity correction and focusing are completed, and infrared thermal images of the thermally conductive silicone sheet sample at different times during the entire stretching process are acquired simultaneously. In this embodiment, the acquisition frequency of load data and infrared thermal image is 100Hz, and the acquisition frame rate of deformation image is 100fps. As for other implementation methods, the implementer can set them according to the actual situation.

[0022] It should be noted that, since the infrared thermal imager and the industrial camera are arranged in a coaxial or near-coaxial manner, through pixel registration, the temperature data corresponding to each pixel of the infrared thermal image can be accurately mapped to the position of the silicone sheet sample surface in the same spatial coordinate system as the deformation image, thereby realizing the joint characterization of the temperature field and the strain field in the same physical region.

[0023] Thus, the load data, deformation images, and infrared thermal images of the thermally conductive silicone sheet sample with the sprayed speckle pattern were obtained at each moment during the entire tensile process.

[0024] Step 2: Define the surface image of the specimen before tension as the reference image, divide it into multiple local regions, and use an image matching algorithm to extract the corresponding matching regions in the deformation images of each local region in the reference image at each time. Quantify the strain of each local region at each time by using the spatial distance change characteristics between the local region and its matching region and the adjacent regions in the image. Calculate the strain anomaly degree of each local region at each time based on the deviation of the strain and its rate of change of the local region at the same time.

[0025] The tensile strength testing of thermally conductive silicone pads is primarily based on the assumption of macroscopic homogeneity, relying on the peak value of the overall macroscopic load throughout the tensile process. However, as a heterogeneous material composed of a silicone rubber matrix and thermally conductive fillers, thermally conductive silicone pads inevitably contain microscopic defects such as filler agglomeration, localized micropores, or uneven interfacial bonding. During tensile testing, these weak areas preferentially experience localized stress concentration and strain localization, causing microcracks to initiate and propagate long before the overall load reaches its peak. At this point, the material's macroscopic mechanical response may still show an upward trend, giving the illusion of continuously increasing load-bearing capacity, but irreversible damage has already occurred to the internal local structure. Judging tensile strength solely based on the overall load will severely overestimate the material's actual safety margin. In actual operating conditions, thermally conductive silicone pads often need to withstand complex loads such as thermal cycling, assembly stress, and vibration shock simultaneously. Designing based on an overestimated tensile strength value can lead to sudden brittle fracture or interfacial delamination within the expected lifespan of the material, directly causing the failure of electronic device cooling systems, and even causing core components to be damaged due to overheating. Therefore, it is necessary to introduce quantitative analysis of local damage characteristics into tensile strength testing, and to correct the tensile strength test results to eliminate the artificially high value caused by local instability, so that the measured tensile strength can truly reflect the reliable load-bearing capacity of the material under service conditions.

[0026] First, under normal conditions, there are no obvious weak areas inside the material. During tensile testing, due to the stable stress transfer between the internal filler and the matrix, the local strain field exhibits a uniformly diffused distribution, with small strain differences between different local areas. When local stiffness decreases due to factors such as filler agglomeration and voids, the low-stiffness areas will preferentially undergo larger deformations and form strain localization, manifested as significantly higher strain and its growth rate in this area compared to the surrounding areas. Therefore, by analyzing the strain change trend of local areas on the surface of the thermally conductive silicone sheet sample during tensile testing, the strain anomaly degree is calculated to measure the abrupt changes in strain concentration within the material during tensile testing. Specifically: Acquire a surface image of the thermally conductive silicone sheet sample after spraying a speckle pattern before the start of tensile testing, and define it as the reference image; then divide the reference image into multiple local regions. In this embodiment, using The moving window divides the local area, wherein there is at least 20% overlap between adjacent local areas, and each local area contains at least 3 speckles; The value is set to 20 pixels. As another implementation method, the implementer can set it according to the actual situation.

[0027] The centroids of each local region in the reference image are obtained. Their corresponding pixels in the deformed image are used as the search starting point. An image matching algorithm is used, with each local region in the reference image as the matching template, to search and match around the search starting point to obtain the corresponding matching region. In this embodiment, the zero-mean normalized cross-correlation matching algorithm is used for matching. The zero-mean normalized cross-correlation matching algorithm is a well-known technology and will not be described in detail here.

[0028] The remaining local regions in the reference image that are adjacent to each local region are defined as neighboring regions; the mean distance between the centroid of each local region in the reference image and the centroids of all its neighboring regions is calculated as the average reference distance. For each local region in the reference image, the corresponding matching region in the deformed image of its neighboring region is defined as the neighboring matching region; For each time step of the deformation image, the mean distance between the centroid of the corresponding matching region of each local region in the deformation image and the centroids of all neighboring matching regions is calculated as the deformation distance. The difference between the deformation distance and the average reference distance is calculated as the deformation of each local area at each time, and the ratio of the deformation distance to the average reference distance is used as the strain of each local area at each time. It should be noted that the average reference distance reflects the spatial relationship between the local area and its surrounding adjacent areas in the initial undeformed state; while the deformation distance reflects the actual spatial relationship between the local area and its surrounding adjacent areas at that moment. The difference between the two eliminates the influence of rigid body displacement and only retains the actual tensile deformation of the local area. The larger the deformation, the longer the local area is stretched relative to the surrounding areas, that is, the greater the actual degree of deformation that occurs at that point; correspondingly, the larger the strain, the more severe the tensile deformation that the local area is subjected to, and the higher the risk of strain localization or damage.

[0029] Calculate the rate of change of strain in each local region at each time relative to its previous preset time interval, and use it as the strain rate; In this embodiment, the preset interval duration is 0.5s. As for other implementation methods, the implementer can set it according to the actual situation. Secondly, the formula for calculating the rate of change is a well-known technique and will not be elaborated here. The specific formula is as follows: in, For the first A local area in strain rate at any given moment For the first A local area in The dependent variable at time, For the first A local area in The dependent variable at time, The preset interval duration; It should be noted that, if The duration of all moments preceding a given moment is less than [a certain value]. If the strain rate is not calculated at that moment, then the strain rate at that moment is not calculated.

[0030] Calculate the mean strain rate of all local regions at the same time, and use it as the average rate of change; Calculate the mean of the strain in all local regions at the same time point, and use it as the average strain. Calculate the ratio of strain to average strain for each local region at each time step, and calculate the ratio of strain rate to average rate of change for each local region at each time step. The product of the two ratios is taken as the strain anomaly degree of each local region at each time step. It should be noted that, when calculating the two ratios, to avoid the denominator being 0, a parameter tuning factor is added to the denominator. In this embodiment, the parameter tuning factor is set to [value missing]. As another implementation method, the implementer can set it according to the actual situation; secondly, the larger the strain rate, the faster the deformation rate of the local area, that is, the material in the area is undergoing rapid shape change. During the stretching process of the thermally conductive silicone sheet, an increase in strain rate usually means that damage events such as increased strain localization, microcrack initiation, or interface debonding have occurred in the area, causing the material to deteriorate rapidly at that point; the larger the ratio of strain to average strain, the more significantly the current deformation degree of the local area is higher than the average level of the whole field, that is, the area has become a weak link with concentrated deformation; the larger the ratio of strain rate to average rate of change, the more significantly the deformation deterioration rate of the local area is faster than the average rate of the whole field, that is, the damage in the area is developing rapidly; therefore, the larger the obtained strain anomaly, the higher the degree of deformation concentration in the local area and the faster the deterioration rate, that is, the area has become the most dangerous weak link, and the more serious the local damage.

[0031] Thus, the strain anomaly degree of each local region at each time moment is obtained.

[0032] Step 3: For each local area, the temperature deviation level of the matching area at the corresponding area in the infrared thermal image is used to reflect the degree of abnormal temperature rise caused by local damage, thereby obtaining the thermal anomaly degree. Combined with the strain anomaly degree, the coupling damage degree of each local area at each time is determined.

[0033] Secondly, under normal tensile conditions, the thermally conductive silicone sheet primarily undergoes elastic deformation, with weak frictional energy dissipation and interfacial slippage. The overall temperature rise is slow and uniform only due to the viscoelastic hysteresis effect, without any localized high-temperature anomalies. However, when local strain becomes sustained and leads to microcracks or interfacial debonding, the unsteady propagation of cracks and severe interfacial slippage cause a sharp increase in localized energy dissipation, resulting in significant localized hot spots in infrared thermal images. To capture this thermal effect anomaly induced by mechanical damage, the overall level of temperature data within the local area is analyzed, and the degree of thermal anomaly is calculated to measure the localized thermodynamic anomaly caused by frictional work due to damage. Specifically: For each local area, the corresponding matching region in the deformation image at each time moment is mapped onto the synchronously acquired infrared thermal image, and the mean value of the temperature data of all pixels in the mapped region is calculated as the average temperature. Calculate the average temperature data of all pixels in the infrared thermal image at each time point, and use it as the overall temperature level; The difference between the average temperature of each local area at each time point and the overall temperature level is calculated as the temperature deviation. If the temperature deviation is less than 0, the thermal anomaly degree of each local area at each time is assigned to 0; otherwise, the thermal anomaly degree is positively correlated with the temperature deviation. It should be noted that a positive correlation means that the dependent variable increases as the independent variable increases and decreases as the independent variable decreases.

[0034] In this embodiment, the temperature deviation is normalized and used as the thermal anomaly degree. The normalization process is as follows: the maximum temperature deviation of all local areas at each time point is obtained, and the ratio of the temperature deviation to the maximum temperature deviation is used as the normalization result.

[0035] It should be noted that during normal stretching, the thermally conductive silicone sheet experiences a slow and uniform temperature rise due to viscoelastic hysteresis. The local temperature is basically the same as the overall temperature, with the temperature deviation fluctuating around zero. When microcracks propagate, interfaces debond, or internal friction intensifies in a localized area, the energy consumption in that area increases sharply, forming a local hot spot significantly higher than the overall temperature. In this case, the temperature deviation is positive and relatively large. When the temperature deviation is negative, the thermal anomaly is forcibly set to zero because such situations do not represent thermally induced damage and should not be included in the anomaly evaluation. The larger the obtained thermal anomaly, the more significant the abnormal temperature rise in that localized area during stretching, reflecting the more severe the energy dissipation caused by damage behaviors such as localized microcrack propagation, interface debonding, or internal friction, and the more serious the damage.

[0036] Furthermore, based on the strain anomaly degree and the thermodynamic anomaly degree, the coupled damage degree is determined, specifically as follows: The degree of coupled damage, strain anomaly, and thermodynamic anomaly in each local region at each time point are all positively correlated. In this embodiment, the sum of the strain anomaly and the thermodynamic anomaly is used as the coupling damage degree.

[0037] It should be noted that the coupled damage degree reflects the degree of local damage in the region from two dimensions: mechanical performance and thermal performance. The larger the value, the more significant the strain concentration and accelerated deterioration trend in the region, as well as the abnormal heat release. The more severe the local damage and the faster the damage evolution, the greater the negative impact on the true value of tensile strength in the region.

[0038] Thus, the coupling impairment degree of each local region in the reference image at each time step is obtained.

[0039] Step 4: Analyze the abrupt bending of the sample edge in the deformation image and the characteristics of the load data deviating from the smooth trend to obtain the overall instability coefficient at each time. Combine the coupling damage degree to determine the damage intensity at each time and correct the tensile strength of the thermally conductive silicone sheet sample accordingly.

[0040] Furthermore, the flowchart of the method for obtaining the overall instability coefficient provided in this application embodiment is as follows: Figure 2 As shown.

[0041] First, under normal conditions, the stress transfer between the matrix and the filler is stable. During the normal and uniform stretching of the thermally conductive silicone sheet, the load shows a continuous and stable increasing trend, and the load curve is smooth overall with only minor random fluctuations. However, with the occurrence of local damage, especially the unsteady propagation of microcracks, severe interface slippage, and sudden rearrangement of the internal structure, the material's resistance to tension will experience instantaneous and discontinuous jumps, which are reflected in the macroscopic load curve as local high-frequency oscillations. To extract this abnormal fluctuation reflecting internal damage from the macroscopic mechanical response, the degree of deviation of the load data's changing trend is analyzed, and the load deviation is calculated, specifically: Curve fitting is performed on the load data at each time point and multiple time points prior to it, and the fitting error of the fitted curve is calculated as the load deviation at each time point. In this embodiment, a quadratic polynomial fitting algorithm is used to perform curve fitting on the load data at each time point and the 20 time points prior. As for other implementation methods, the implementer can set them according to the actual situation. Secondly, the fitting error is measured by calculating the root mean square error. The quadratic polynomial fitting algorithm and the calculation of the root mean square error are well-known techniques and will not be described in detail here.

[0042] It should be noted that the greater the load deviation, the more significant the deviation of the actual load data from the fitted trend curve, that is, the more severe the local fluctuations or oscillations of the load curve, reflecting the more severe the local damage events that occurred nearby and the more serious the degradation of the effective load-bearing structure inside the material.

[0043] Secondly, under normal tensile conditions, the thermally conductive silicone sheet exhibits uniform overall deformation, with the edge contour typically remaining continuous and smooth, displaying only uniform shrinkage characteristics. However, when localized tears or cracks propagate and extend to the edge, the material's effective load-bearing structure severely degrades. The appearance of cracks directly disrupts the geometric continuity of the edge, forming local curvature abrupt changes or micro-notches. Therefore, to quantify this edge geometric anomaly caused by damage, the differences in edge contour curvature distribution in the deformation image are analyzed, and the edge anomaly degree is calculated to measure the degree of damage to the integrity and continuity of the macroscopic geometric structure of the sample edge. Specifically: For the deformation images at each time point, the edge contour lines of the thermally conductive silicone sheet sample in the deformation images are extracted; In this embodiment, the Canny edge detection algorithm is used for edge detection. The Canny edge detection algorithm is a well-known technology and will not be described in detail here.

[0044] Multiple edge pixels are uniformly selected along the edge contour line and defined as feature points. The curvature at each feature point is then calculated. In this embodiment, 10% of the total number of edge pixels on the edge contour line are selected as feature points. In other implementation methods, the implementer can set it according to the actual situation. Secondly, the calculation of curvature is a well-known technique and will not be described in detail here.

[0045] Calculate the mean of the difference in curvature between any two feature points, and perform a positive mapping on it to obtain the edge anomaly degree at each time step; In this embodiment, the mean of the absolute values ​​of the difference in curvature between any two feature points is positively mapped to the edge anomaly degree at each time step. The specific process of positive mapping is as follows: the sum of the mean and the value 1 is used as the result of positive mapping. Through the process of positive mapping, the edge anomaly degree is made to be greater than 0.

[0046] It should be noted that the greater the edge anomaly, the less smooth the edge contour of the thermally conductive silicone sheet sample at that moment, and the more significant the local curvature abrupt change or geometric discontinuity features. This reflects that the sample edge has already developed or is developing local damage such as cracks, notches or tears, and the structural integrity of the material is being destroyed.

[0047] Furthermore, based on the load deviation and edge anomaly, the overall instability coefficient is determined as follows: The overall instability coefficient at each time point is positively correlated with the load deviation and the edge anomaly. In this embodiment, the normalized result of the product of load deviation and edge anomaly is used as the overall instability coefficient. Secondly, the normalization process is as follows: the product at all times is normalized using the maximum-minimum normalization method. The maximum-minimum normalization method is a well-known technique and will not be described in detail here.

[0048] It should be noted that the larger the overall instability coefficient, the more macroscopic anomalies the thermally conductive silicone sheet sample exhibits at that moment. These anomalies are manifested as severe load fluctuations and obvious curvature abrupt changes in the edge contour, indicating that the sample as a whole has entered an unstable state and macroscopic damage is developing.

[0049] Furthermore, based on the overall instability coefficient and coupled damage degree, the damage intensity is determined, specifically as follows: Select the maximum coupling damage degree of all local regions at each time point; The damage intensity at each time point is positively correlated with the maximum coupled damage degree and the overall instability coefficient; In this embodiment, the product of the maximum coupling damage degree and the overall instability coefficient is used as the damage intensity at each time point.

[0050] It should be noted that the greater the damage intensity, the more severe the damage to the weakest local area of ​​the sample at that moment, and the more significant the overall macroscopic instability of the sample. This reflects that irreversible local damage has occurred inside the material, and the more serious the overestimation of the traditional tensile strength test results. Therefore, a greater reduction should be given in subsequent corrections.

[0051] In the tensile strength testing of thermally conductive silicone sheets, the evolution of internal damage is not an instantaneous event, but a dynamic cumulative process from local initiation to global expansion. The damage intensity at a single moment cannot depict the trajectory from quantitative to qualitative change. Therefore, by introducing an exponential smoothing algorithm to calculate the spatiotemporal trend factor, the irreversible damage accumulation of the material is dynamically integrated with the abnormal deterioration at the current moment. This avoids unreasonable jumps in the correction results with instantaneous data, and can reflect the irreversible evolution process of internal damage in a true and stable manner, thereby more scientifically quantifying the risk of reduction in overall tensile strength.

[0052] The damage intensity at all times is smoothed, and the smoothing results at each time are used as spatiotemporal trend factors. In this embodiment, an exponentially weighted smoothing algorithm is used for smoothing. This algorithm is a well-known technique and will not be described in detail here. The specific formula is as follows: in, for Spatiotemporal trend factors at any given moment for Damage intensity at any given moment The preset smoothing coefficient, for The spatiotemporal trend factor at each moment is determined first, and then the spatiotemporal trend factor at the first moment is set to 0; because when If the value is too small, the algorithm's historical memory becomes too heavy, resulting in a slow response speed. This can lead to problems when the material experiences rapid local instability. There is a significant lag, making it impossible to reflect the true evolutionary process of the damage in a timely manner; when When the value is too large, the weight of recent data is too high, making it susceptible to transient disturbances. This can lead to spurious fluctuations in the calculation results of subsequent correction factors. Therefore, the preset smoothing coefficient... The range of values ​​is In this embodiment, the value is set to 0.4. In other implementation methods, the implementer can set it according to the actual situation.

[0053] It should be noted that the larger the spatiotemporal trend factor, the more severe the local defects become during the stretching stage, reflecting a serious irreversible degradation of the material's internal structure. By using exponential smoothing to eliminate random noise from a single acquisition, the dynamic accumulation process of defects from their inception to global expansion is accurately recorded.

[0054] Extract the spatiotemporal trend factor at the moment when the load data reaches its peak during the entire tensile process, and perform a negative mapping on it as a correction factor. In this embodiment, the specific process of negative mapping is as follows: the arctangent function is used to map the spatiotemporal trend factor, and the compression coefficient 2 / π is multiplied to control its output range to not exceed 1, thereby completing the normalization process. On this basis, the difference between the value 1 and the normalized result is used as the correction factor. The arctangent function is a well-known technique and will not be described in detail here.

[0055] The tensile strength of the thermally conductive silicone sheet sample measured by the universal testing machine is obtained, and its product with the correction factor is taken as the final tensile strength of the thermally conductive silicone sheet sample. It should be noted that extracting the moment when the load reaches its peak is to accurately pinpoint the most extreme and dangerous cumulative damage state that the material experiences before macroscopic failure. Therefore, a larger correction factor indicates that there is no obvious premature damage inside, reflecting that the tensile strength measured by the testing machine is true and reliable, and no reduction is needed. A smaller correction factor indicates that severe local damage has occurred inside the material before reaching the maximum load-bearing capacity, reflecting that the tensile strength measured by the testing machine is artificially high. The more severe the damage, the more severe the reduction. By using mathematical rules to forcibly remove the false load-bearing capacity caused by local instability, the final output tensile strength can truly reflect the material's safe load-bearing limit. This completely eliminates the hidden danger of overestimation of strength caused by the assumption of macroscopic homogeneity, effectively avoiding sudden brittle fracture or interface peeling caused by premature local instability in actual working conditions, and effectively ensuring the design reliability and operational safety of the heat dissipation system of electronic equipment.

[0056] Based on the same inventive concept as the above method, this application embodiment also provides a thermally conductive silicone sheet tensile strength testing device, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of the above-described methods for testing the tensile strength of a thermally conductive silicone sheet.

Claims

1. A method for testing the tensile strength of a thermally conductive silicone sheet, characterized in that, The method includes the following steps: Load data, deformation images, and infrared thermal images of the thermally conductive silicone sheet sample after the speckle pattern was sprayed were collected at various moments during the entire tensile process. The surface image of the specimen before tension is defined as the reference image, which is divided into multiple local regions. An image matching algorithm is used to extract the matching region corresponding to each local region in the deformation image at each time step in the reference image. The strain of each local region at each time step is quantified by the spatial distance variation characteristics between the local region and its matching region and the adjacent region in the image. Based on the degree of deviation of the strain of the local region and its rate of change at the same time step, the strain anomaly of each local region at each time step is calculated to characterize the degree of mechanical damage caused by strain localization and its evolution trend. For each local region, the temperature deviation level of the matching region at the corresponding region in the infrared thermal image is used to reflect the degree of abnormal temperature rise caused by local damage, thereby obtaining the thermal anomaly degree. Combined with the strain anomaly degree, the coupling damage degree of each local region at each time is determined. By analyzing the abrupt bending of the sample edge in the deformation image and the characteristics of the load data deviating from the smooth trend, the overall instability coefficient at each moment is obtained. Combined with the coupling damage degree, the damage intensity at each moment is determined, and the tensile strength of the thermally conductive silicone sheet sample is corrected accordingly.

2. The method for testing the tensile strength of a thermally conductive silicone sheet as described in claim 1, characterized in that, The step of extracting the corresponding matching regions in the deformed image at each time step using an image matching algorithm includes: obtaining the centroid of each local region in the reference image, using its corresponding pixel in the deformed image as the search starting point, using an image matching algorithm, using each local region in the reference image as the matching template, performing a search and matching around the search starting point to obtain the corresponding matching region.

3. The method for testing the tensile strength of a thermally conductive silicone sheet as described in claim 2, characterized in that, The process of obtaining the strain of each local region at each time point is as follows: The remaining local regions in the reference image that are adjacent to each local region are defined as neighboring regions; the mean distance between the centroid of each local region in the reference image and the centroids of all its neighboring regions is calculated as the average reference distance. For each local region in the reference image, the corresponding matching region in the deformation image is defined as the neighboring matching region; for each deformation image at each time point, the average distance between the centroid of the corresponding matching region of each local region in the deformation image and the centroid of all neighboring matching regions is calculated as the deformation distance. The difference between the deformation distance and the average reference distance is calculated as the deformation of each local region at each time. The ratio of the deformation distance to the average reference distance is used as the strain of each local region at each time.

4. The method for testing the tensile strength of a thermally conductive silicone sheet as described in claim 1, characterized in that, The calculation of the strain anomaly degree of each local region at each time step includes: calculating the rate of change of strain of each local region at each time step relative to its previous preset time interval, as the strain rate; calculating the mean of the strain rate and the mean of the strain of all local regions at the same time step, as the average rate of change and the average strain, respectively; calculating the ratio of the strain of each local region at each time step to the average strain, and calculating the ratio of the strain rate of each local region at each time step to the average rate of change, and multiplying the two ratios as the strain anomaly degree of each local region at each time step.

5. The method for testing the tensile strength of a thermally conductive silicone sheet as described in claim 1, characterized in that, The calculation process for the thermal anomaly is as follows: For each local area, the corresponding matching region in the deformation image at each time step is mapped onto the synchronously acquired infrared thermal image. The mean value of the temperature data at all pixels in the mapped region is calculated as the average temperature. The mean value of the temperature data at all pixels in the infrared thermal image at each time step is calculated as the overall temperature level. The difference between the average temperature of each local area at each time and the overall temperature level is calculated as the temperature deviation. If the temperature deviation is less than 0, the thermal anomaly of each local area at each time is assigned a value of 0. Otherwise, the thermal anomaly is positively correlated with the temperature deviation.

6. The method for testing the tensile strength of a thermally conductive silicone sheet as described in claim 1, characterized in that, The coupling damage degree is positively correlated with both the strain anomaly degree and the thermodynamic anomaly degree.

7. The method for testing the tensile strength of a thermally conductive silicone sheet as described in claim 1, characterized in that, The obtained overall instability coefficients at each time point include: Curve fitting is performed on the load data at each time point and multiple time points prior to it, and the fitting error of the fitted curve is calculated as the load deviation at each time point. For the deformation images at each time point, the edge contour line of the thermally conductive silicone sheet sample in the deformation image is extracted; multiple edge pixels are uniformly selected on the edge contour line and defined as feature points, and the curvature at each feature point is calculated; the mean value of the difference in curvature between any two feature points is calculated and positively mapped to it as the edge anomaly degree at each time point. The overall instability coefficient is positively correlated with both load deviation and edge anomaly.

8. The method for testing the tensile strength of a thermally conductive silicone sheet as described in claim 1, characterized in that, Determining the damage intensity at each moment includes: selecting the maximum coupling damage degree of all local regions at each moment; the damage intensity is positively correlated with the maximum coupling damage degree and the overall instability coefficient.

9. The method for testing the tensile strength of a thermally conductive silicone sheet as described in claim 1, characterized in that, The process of correcting the tensile strength of the thermally conductive silicone sheet sample includes: smoothing the damage strength at all times and using the smoothing results at each time as a spatiotemporal trend factor; extracting the spatiotemporal trend factor at the time when the load data reaches its peak during the entire tensile process, negatively mapping it as a correction factor; obtaining the tensile strength of the thermally conductive silicone sheet sample measured by a universal testing machine, and multiplying it by the correction factor as the final tensile strength of the thermally conductive silicone sheet sample.

10. A device for testing the tensile strength of a thermally conductive silicone sheet, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method for testing the tensile strength of a thermally conductive silicone sheet as described in any one of claims 1-9.

Images