Radiating fin near-surface defect detection device and method based on active heating

Abstract

The invention discloses a radiating fin near surface defect detection device and method based on active heating, and the device comprises an angle adjustment module which comprises a rotating table which is used for bearing a corrugated radiating fin and adjusting and controlling the spatial attitude of the corrugated radiating fin; the semiconductor laser thermal excitation module is used for applying Gaussian beam thermal excitation to a to-be-detected area on the surface of the corrugated cooling fin; the infrared thermal imaging thermodetector is used for collecting a thermal infrared image sequence of a to-be-detected area on the surface of the corrugated cooling fin in a thermal excitation period and a subsequent cooling process in real time; the computer module is used for receiving the thermal infrared image sequence and carrying out preprocessing and time sequence alignment on the thermal infrared image sequence, and the computer module is further used for extracting heat conduction characteristic parameters of the to-be-detected area of the surface of the corrugated cooling fin in the thermal infrared image sequence. By applying the device and the method, the 50mu m-level microcracks can be detected, the single detection period is less than 30 seconds, and the efficiency is improved by 10 times compared with that of manual detection.

Description

Technical Field

[0001] This invention relates to the field of nuclear power technology, and in particular to a device and method for detecting near-surface defects in heat sinks based on active heating. Background Technology

[0002] Heat sinks are indispensable key components in power electronic equipment (electrical control systems, power modules, etc.) and high-power LED lighting, and their heat dissipation performance directly affects the reliability and lifespan of the equipment. During the manufacturing process of metal heat sinks (such as die casting, extrusion, brazing, CNC machining, folding, etc.), defects such as micro-cracks, pores, inclusions, delamination, or separation can easily occur on or near the surface, leading to increased local thermal resistance and significantly reduced heat dissipation efficiency. Studies have shown that microcracks with a depth of 50μm can reduce the thermal conductivity of heat sinks by 12%-18%, ultimately causing overheating failure of the equipment. Although these near-surface defects may not be directly exposed on the surface, they are extremely harmful: they reduce the structural strength and rigidity of the heat sink; they become initiation points for corrosion; they obstruct effective heat conduction paths, creating local thermal resistance at the defect sites, reducing the overall thermal conductivity of the heat sink, leading to hot spots, and ultimately affecting the heat dissipation efficiency and reliability of the equipment. Changes in defect size, depth, and thermal conductivity have a very sensitive impact on heat dissipation performance.

[0003] Currently, traditional methods for quality inspection of heat sinks include: Visual inspection: Relies on manual labor, is inefficient, cannot detect internal or minute defects, has a missed detection rate of >25% (actual data), and the corrugated structure can increase the blind spot.

[0004] Liquid penetrant testing: can only detect defects that open on the surface and can only detect through cracks; it is ineffective for closed delamination. Micropores on the surface of metal heat sinks are prone to reagent residue and require subsequent cleaning.

[0005] X-ray or industrial CT: suitable for internal cavities and inclusions, but the equipment is expensive and bulky, and has poor contrast for delamination defects parallel to the X-ray (sensitivity <200μm), and cannot detect planar defects such as debonding; there are radiation safety protection requirements, the detection speed is relatively slow, the operating cost is high, and it is difficult to use for large-scale rapid detection on the production line.

[0006] Ultrasonic testing: suitable for detecting internal volumetric defects, but difficult to couple to complex-shaped heat sinks. Summary of the Invention

[0007] The technical problem to be solved by the present invention is to provide a device and method for detecting near-surface defects of heat sinks based on active heating, in order to address the shortcomings of the prior art.

[0008] The technical solution adopted by this invention to solve its technical problem is: constructing a near-surface defect detection device for a heat sink based on active heating, comprising: An angle adjustment module includes a rotary table, which is used to support the corrugated heat sink and adjust the spatial orientation of the corrugated heat sink; The semiconductor laser thermal excitation module includes a fiber-coupled semiconductor laser and a laser detection head that cooperates with the fiber-coupled semiconductor laser. The semiconductor laser thermal excitation module is used to apply Gaussian beam thermal excitation to the area to be inspected on the surface of the corrugated heat sink. An infrared thermal imaging thermometer is used to acquire in real time a sequence of thermal infrared images of the area to be inspected on the surface of the corrugated heat sink during thermal excitation and subsequent cooling process. A computer module is connected to the semiconductor laser thermal excitation module and the infrared thermal imaging thermometer. The computer module is used to receive the thermal infrared image sequence and perform preprocessing and timing alignment on the thermal infrared image sequence. The computer module is also used to extract the thermal conduction characteristic parameters of the area to be inspected on the surface of the corrugated heat sink in the thermal infrared image sequence.

[0009] In some embodiments, the rotary table is used to adjust the pitch angle of the corrugated heat sink. So that the angle between the optical axis of the infrared thermal imaging thermometer and the normal of the area to be measured on the surface of the corrugated heat sink is less than 30°.

[0010] In some embodiments, the power range of the fiber-coupled semiconductor laser is 70W-640W.

[0011] In some embodiments, the infrared thermal imaging thermometer includes a 640×512 pixel long-wave infrared camera with a sampling rate of 100Hz and a thermal sensitivity of 40mK.

[0012] This invention provides a method for detecting near-surface defects in heat sinks based on active heating, used in any of the above embodiments of the device for detecting near-surface defects in heat sinks based on active heating, comprising the following steps: S10: The angle adjustment module adjusts the spatial orientation of the corrugated heat sink so that the angle between the optical axis of the infrared thermal imaging thermometer and the normal of the area to be measured on the surface of the corrugated heat sink is less than 30°. S20: The fiber-coupled semiconductor laser applies a Gaussian beam thermal excitation to the area to be inspected on the surface of the corrugated heat sink through a laser detection head; S30: The infrared thermal imaging thermometer acquires in real time the thermal infrared image sequence of the area to be inspected on the surface of the corrugated heat sink during thermal excitation and subsequent cooling process, and transmits the thermal infrared image sequence to the computer module. S40: The computer module receives the thermal infrared image sequence and performs the following steps on the thermal infrared image sequence: S41: Preprocess and time-align the thermal infrared image sequence; S42: Extract the thermal conductivity characteristic parameters of the area to be inspected on the surface of the corrugated heat sink in the thermal infrared image sequence; S43: Compare the thermal conductivity characteristic parameters with the preset defect-free benchmark model and output the comparison results; S44: Determine whether there are near-surface defects based on the comparison results; if so, output the location, size estimate, and type information of the near-surface defects.

[0013] In some embodiments, in step S42, extracting the thermal conductivity characteristic parameters of the inspected area of ​​the corrugated heat sink surface in the thermal infrared image sequence includes: collecting and calculating the characteristic parameters of the corrugated top area, corrugated valley area and corrugated sloping wall area; and analyzing the difference in characteristic parameters between the same type of corrugated areas under different defect conditions.

[0014] In some embodiments, step S43 involves comparing and analyzing the thermal conductivity characteristic parameters with a preset defect-free benchmark model, including: S431: Perform time-domain relative thermal contrast (RTC) analysis: calculate the relative temperature difference between the area under inspection and the reference area at the same time point. ;in, Temperature of the area to be inspected. Temperature for reference area; And / or, S432: Perform thermal diffusion rate difference analysis: calculate the standard deviation of thermal diffusion rate between the area under test and the reference model at the same location; And / or, S433: Perform peak arrival time offset analysis: calculate the offset between the peak arrival time of the region under inspection and the corresponding time in the reference model. ; And / or, S434: Perform machine learning model discrimination processing: input the heat conduction feature parameters into the neural network classification prediction model to determine the probability of the existence of defects.

[0015] In some embodiments, the reference area is: a defect-free symmetrical corrugated area on the same part; or, corrugated area data obtained from standard parts in the same batch under the same testing conditions.

[0016] In some embodiments, the benchmark model is: a standard part from the same batch or a symmetrical, defect-free area.

[0017] In some embodiments, in step S431, when If so, it is considered abnormal.

[0018] The present invention offers the following advantages: The active heating-based near-surface defect detection device and method for heat sinks can detect microcracks down to 50μm (traditional X-ray sensitivity ≥200μm), with a single detection cycle of less than 30 seconds, improving efficiency by 10 times compared to manual inspection (traditional visual inspection has a false negative rate greater than 25%), thus significantly enhancing detection efficiency. Furthermore, it is compatible with low-emissivity metals such as titanium alloys and stainless steel, and supports the inspection of curved and thin-walled (≥1mm) heat sinks, offering improved practicality and applicability. Attached Figure Description

[0019] To more clearly illustrate the technical solution of the present invention, the present invention will be further described below in conjunction with the accompanying drawings and embodiments. It should be understood that the following drawings only show some embodiments of the present invention and should not be considered as a limitation of the scope. For those skilled in the art, other related drawings can be obtained from these drawings without creative effort. In the drawings: Figure 1 This is a schematic diagram of a near-surface defect detection device for a heat sink based on active heating in some embodiments of the present invention. Detailed Implementation

[0020] To provide a clearer understanding of the technical features, objectives, and effects of this invention, specific embodiments are now described in detail with reference to the accompanying drawings. In the following description, it should be understood that the orientations or positional relationships indicated by terms such as "front," "rear," "upper," "lower," "left," "right," "longitudinal," "horizontal," "vertical," "horizontal," "top," "bottom," "inner," "outer," "head," and "tail" are based on the orientations or positional relationships shown in the accompanying drawings, and are constructed and operated in a specific orientation. They are only for the convenience of describing this technical solution and do not indicate that the device or element referred to must have a specific orientation; therefore, they should not be construed as limitations on this invention.

[0021] It should also be noted that, unless otherwise explicitly specified and limited, terms such as "installation," "connection," "linking," "fixing," and "setting" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; 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; they can refer to the internal communication of two components or the interaction between two components. When an component is referred to as being "on" or "below" another component, the component can be located "directly" or "indirectly" on the other component, or there may be one or more intermediary components. The terms "first," "second," "third," etc., are only for the convenience of describing this technical solution and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, features defined with "first," "second," "third," etc., may explicitly or implicitly include one or more of that feature. For those skilled in the art, the specific meaning of the above terms in this invention can be understood according to the specific circumstances.

[0022] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of the invention. However, those skilled in the art will understand that the invention can be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods are omitted so as not to obscure the description of the invention with unnecessary detail.

[0023] See Figure 1 This invention discloses a near-surface defect detection device for heat sinks based on active heating, suitable for rapid, non-contact, and highly sensitive automated detection of minute near-surface defects in metal heat sinks with corrugated structures. The active heating-based near-surface defect detection device for heat sinks includes: Angle adjustment module 4 includes a rotary table, which is used to support the corrugated heat sink 3 and adjust the spatial orientation of the corrugated heat sink 3.

[0024] The semiconductor laser thermal excitation module includes a fiber-coupled semiconductor laser 1 and a laser detection head 2 that cooperates with the fiber-coupled semiconductor laser 1. The semiconductor laser thermal excitation module is used to apply Gaussian beam thermal excitation to the area to be inspected on the surface of the corrugated heat sink 3. Specifically, the fiber-coupled semiconductor laser 1 applies a continuous thermal pulse Gaussian beam with controllable power, spot shape, and size to the area to be inspected on the corrugated heat sink 3 through the laser detection head 2.

[0025] The infrared thermal imaging thermometer 5 is used to acquire in real time the thermal infrared image sequence of the area to be inspected on the surface of the corrugated heat sink 3 during thermal excitation and subsequent cooling process.

[0026] Computer module 6, connected to the semiconductor laser thermal excitation module and the infrared thermal imaging thermometer 5, is used to receive the thermal infrared image sequence and perform preprocessing and timing alignment on the thermal infrared image sequence. Computer module 6 is also used to extract the thermal conductivity characteristic parameters of the inspected area on the surface of the corrugated heat sink 3 in the thermal infrared image sequence. Computer module 6 is further used to perform the following steps on the thermal infrared image sequence: preprocessing and timing alignment of the thermal infrared image sequence; extracting the thermal conductivity characteristic parameters of the inspected area on the surface of the corrugated heat sink in the thermal infrared image sequence; comparing the thermal conductivity characteristic parameters with a preset defect-free benchmark model and outputting the comparison result; determining whether near-surface defects exist based on the comparison result; and if so, outputting the location, size estimate, and type information of the near-surface defects.

[0027] The extraction of thermal conduction characteristic parameters of the inspected area on the surface of the corrugated heat sink in the thermal infrared image sequence includes: acquiring and calculating characteristic parameters of the corrugated top area, corrugated valley area, and corrugated sloping wall area; and obtaining the degree of difference in characteristic parameters between the same type of corrugated areas under different defect conditions. Specifically, the complex surface of the corrugated heat sink can lead to uneven heat distribution in normal areas (e.g., rapid heat dissipation in the corrugated top area and slow heat dissipation in the valley area), which can easily mask defect signals. This invention achieves the separation of heat sink structural interference and defect signals by performing feature analysis on the corrugated areas: processing thermal infrared images under the three regions of corrugated top, valley, and sloping wall, and extracting thermal conduction characteristic parameters (e.g., thermal diffusion rate in the top area and peak temperature in the valley area) respectively. The key parameters are calculated as follows: thermal diffusion rate: By calculating the corrugated top, valley, and sloping wall The values ​​are analyzed to determine the degree of difference in similar areas (difference >10% indicates a suspected defect). Then, the thermal conduction characteristic parameters of each area (such as the standard deviation of thermal diffusion rate and the peak arrival time offset) are compared with the preset defect-free benchmark model (standard parts in the same batch or symmetrical defect-free areas).

[0028] In some embodiments, the thermal conductivity characteristic parameters are compared and analyzed with a preset defect-free benchmark model, including: Perform time-domain relative thermal contrast (RTC) analysis: calculate the relative temperature difference between the area under inspection and the reference area at the same time point. ;in, Temperature of the area to be inspected. For the reference region temperature; wherein, in step S431, when If so, it is considered abnormal.

[0029] And / or, perform thermal diffusion rate difference analysis (reflecting thermal conductivity uniformity): calculate the standard deviation of thermal diffusion rate between the area under test and the reference model at the same location; And / or, perform peak arrival time offset analysis: calculate the offset between the peak arrival time of the region under inspection and the corresponding time in the baseline model. ; And / or, perform machine learning model discrimination processing: input the heat conduction feature parameters into the neural network classification and prediction model to determine the probability of the existence of defects.

[0030] In some embodiments, the reference area is: a defect-free symmetrical corrugated area on the same sample; or, corrugated area data obtained from standard samples in the same batch under the same testing conditions.

[0031] In some embodiments, the benchmark model is: a batch of standard parts or a symmetrical, defect-free area.

[0032] In some embodiments, extracting the thermal conductivity characteristic parameters of the inspected area on the surface of the corrugated heat sink in the thermal infrared image sequence includes: acquiring and calculating the thermal conductivity characteristic parameters of the corrugated top region, corrugated valley region, and corrugated sloping wall region; and analyzing the degree of difference in characteristic parameters between corrugated regions of the same type under different defect conditions. Different defects refer to defects such as scratches, cracks, and thinning. Same type refers to the same region of the corrugation, such as the corrugated top region.

[0033] In some embodiments, the near-surface defect size estimation method includes: Preprocessing of thermal infrared image sequences is performed based on image preprocessing and defect segmentation (fuzzy C-means clustering).

[0034] Based on the pixel area and imaging resolution of thermal anomaly regions in thermal infrared image sequences.

[0035] Spatial calibration was performed by combining the laser spot size and the thermal diffusion model. The thermal diffusion model here is the thermal diffusion model on the corrugated heat sink obtained from the calibration results of the experiment.

[0036] A deep prediction model based on convolutional neural networks is employed. This model identifies and distinguishes between "normal differences caused by structure" and "abnormal differences caused by defects." A defect is identified when the probability of its presence, P_defect, is ≥ 95%. Finally, defect information is generated based on the results of the deep prediction model and a heatmap is overlaid. The defect type, location, and size estimate are then labeled on the heatmap, enabling quantitative detection of near-surface micro-defects in corrugated heat sinks.

[0037] Specifically, the angle adjustment module 4 is used to customize the position and angle of the corrugated heat sink 3 under different detection requirements.

[0038] The semiconductor laser thermal excitation module is used to apply a continuously variable thermal pulse signal to the area to be inspected on the corrugated heat sink 3.

[0039] The infrared thermal imaging thermometer 5 is used to collect and record the thermal response process of the surface of the corrugated heat sink 3 under inspection in real time, and obtain a thermal infrared image sequence.

[0040] Computer module 6 can precisely synchronize the timing of laser thermal excitation and thermal infrared image acquisition. It extracts the thermal infrared image sequence acquired by infrared thermal imaging thermometer 5 and uses image processing algorithms to extract the thermal conduction characteristic parameters of the area to be inspected on the surface of the corrugated heat sink 3. Then, it can quantitatively analyze the changes in thermal conduction characteristic parameters caused by near-surface micro-defects of the corrugated heat sink 3 and compare them with a preset defect-free benchmark model. Based on the results of image processing and quantitative analysis, it can determine whether there are near-surface defects or other micro-defects in the corrugated heat sink 3. If there are near-surface defects, it can mark their estimated location and size.

[0041] In some embodiments, the rotary table is used to adjust the pitch angle of the corrugated heat sink 3. So that the angle between the optical axis of the infrared thermal imaging thermometer 5 and the normal of the area to be measured on the surface of the corrugated heat sink 3 is less than 30°.

[0042] In some embodiments, the rotation angle θ of the corrugated heat sink 3 of the rotary table can be adjusted by a motor drive device or manually. The purpose of the adjustment is to ensure that the angle between the optical axis of the infrared thermal imaging thermometer 5 and the normal of the surface of the corrugated heat sink 3 to be measured is <30°, thereby avoiding temperature measurement deviation caused by surface reflection. Understandably, the rotary table can be a multi-angle rotary table of the prior art, and no specific limitation is made here.

[0043] In some embodiments, the semiconductor laser thermal excitation module is configured to generate a Gaussian beam with power, shape, size, and allowing for a custom scanning path; the semiconductor laser thermal excitation module is equipped with a water-cooled temperature control system (temperature control accuracy ±0.1℃) and real-time power monitoring to accurately manage the heat load and ensure the temperature stability of the system (especially the laser thermal excitation source).

[0044] The power range of the fiber-coupled semiconductor laser 1 is 70W-640W. Specifically, the power control of the fiber-coupled semiconductor laser 1 is as follows: the semiconductor laser power can be freely adjusted within the range of 70W to 640W; scanning control allows the laser spot to optionally perform linear scanning motion along or perpendicular to the ripple direction (speed 40mm / s); pulse control allows the thermal pulse duration to be adjusted within any time range of more than 10ms.

[0045] The laser detection head 2 is used for emitting and shaping the Gaussian beam, enabling the laser beam to form a circular or elliptical Gaussian spot with a diameter of 50 mm on the surface of the corrugated heat sink 3. Furthermore, the laser detection head 2 includes a DOE diffraction optical element. By modulating the wavefront of the laser beam, the laser detection head 2 shapes the Gaussian beam into a 50 mm circular / elliptical spot, making the energy distribution of the Gaussian beam more uniform and ensuring that the surface of the corrugated heat sink 3 is uniformly heated during active laser thermal excitation.

[0046] In some embodiments, when a Gaussian-distributed thermal pulse is applied to the surface of the corrugated heat sink 3 by a fiber-coupled semiconductor laser 1 (70–640W adjustable), heat diffuses inward according to an unsteady-state heat conduction equation: Where α is the thermal diffusivity of the material (unit: m) 2 / s)(Titanium alloy ≈ 7.5 × 10 -6 m 2 / s, 304 stainless steel ≈ 4.2 × 10 -6 m 2 / s). Heat transfer within an object follows the law of thermal conductivity, but near-surface defects (such as cracks, inclusions, debonding, etc.) disrupt the continuity of the material, causing abrupt changes in local thermal conductivity characteristics. If the defect is an air gap (such as a crack), due to the air thermal diffusivity (≈2×10), -5 m 2 The / s is much lower than that of metals (titanium alloy ≈ 9×10). -6 m 2 / s, 304 stainless steel ≈ 5×10 -6 m 2 The heat rises rapidly at the defect front, causing heat to accumulate and creating hot spots due to localized temperature anomalies. If the defect is a low-thermal-conductivity inclusion (such as a non-metallic impurity), it will hinder heat diffusion, forming localized cold spots. These thermal anomalies exhibit specific evolutionary patterns over time (such as differences in heating rate, peak temperature, and cooling rate), serving as the physical basis for defect identification. The temperature difference between the abnormal and normal regions... It can be quantified by time-domain relative thermal contrast (RTC) and become a direct basis for defect identification.

[0047] To amplify the difference in thermal response between defective and normal regions, this invention employs active thermal excitation using a semiconductor laser. Its core principle is to achieve targeted energy excitation of the defective region through precise control of laser parameters. Where Q is the heat absorbed by the workpiece (J), P is the laser power (W), t is the irradiation time (s), and η is the material absorptivity (titanium alloy ≈ 0.3, 304 stainless steel ≈ 0.25). The power is dynamically adjusted within the range of 70W-640W according to the heat sink material (titanium alloy / 304 stainless steel) to ensure that the thermal anomaly signal at the defect reaches the detectable threshold (≥0.5℃ temperature difference) without damaging the workpiece. Then, the laser is set to a Gaussian spot (circular or elliptical) with a diameter of 50mm using a DOE diffraction optical element, and can scan along the ripple direction to ensure that the heat is evenly distributed along the heat sink texture, avoiding excitation blind spots caused by structural shadows. This is particularly relevant for metals with high thermal conductivity (thermal diffusivity ≈ 10). -5 m 2 The rapid decay of thermal response caused by the / s signal is addressed by using millisecond-level pulse control synchronized with the infrared acquisition timing to ensure the optimal time window for capturing defect thermal anomalies.

[0048] In some embodiments, the infrared thermal imaging thermometer 5 includes a 640×512 pixel long-wave infrared camera (8–14 μm band) with a sampling rate of 100 Hz and a thermal sensitivity of 40 mK (NETD). The infrared thermal imaging thermometer 5 has a sampling rate of ≥50 Hz and a thermal sensitivity of <50 mK; the temperature measurement accuracy is ±2℃ or ±2%. To address the low emissivity characteristics of titanium alloys (ε≈0.15), a dynamic emissivity calibration algorithm can be used to reduce interference from reflected light.

[0049] In some embodiments, the infrared thermal imaging thermometer 5 converts thermal anomalies into quantifiable image signals, and the key parameters are calculated as follows: Thermal sensitivity (NETD): in, To ensure the minimum resolvable temperature difference (<50mK) and signal-to-noise ratio (SNR) of ≥30dB, the system captures changes down to the 0.1℃ level. The rotation angle θ of the corrugated heat sink is adjusted via an angle adjustment module, ensuring the angle between the optical axis of the infrared thermal imaging thermometer and the normal to the inspected area of ​​the corrugated heat sink is <30°. This minimizes temperature measurement errors caused by surface reflection (reflection error can be reduced to within ±0.2℃), enabling precise acquisition of thermal conduction characteristics in the corrugated top, valley, and sloping wall regions. An infrared camera with a thermal sensitivity of <50mK and a sampling rate of ≥50Hz is used to capture minute temperature changes on the metal surface down to the 0.1℃ level. The long-wave infrared band (8-14μm) is adapted to the low emissivity of the metal, improving the signal-to-noise ratio. Finally, the system simultaneously records thermal infrared image sequences during thermal excitation (heating stage) and cooling stage, fully capturing the dynamic process of defects from "heat accumulation" to "heat dissipation," providing data support for multi-dimensional feature analysis.

[0050] In some embodiments, the material of the corrugated heat sink 3 includes, but is not limited to, titanium alloy or 304 stainless steel, and the thickness of the corrugated heat sink 3 is approximately 2 mm.

[0051] This invention also provides a method for detecting near-surface defects of heat sinks based on active heating, applicable to the near-surface defect detection device for heat sinks based on active heating in any of the above embodiments. The core working principle of the method for detecting near-surface defects of heat sinks based on active heating lies in a three-level collaborative mechanism of "laser active thermal excitation - thermal conduction anomaly capture - infrared imaging collaborative detection technology". By controlling the laser thermal energy input, thermal conduction anomalies of near-surface defects are excited. High-precision infrared imaging captures thermal response signals with high sensitivity. Combined with the extraction and quantitative analysis of heat sink ripple structure features, quantitative detection of small near-surface defects of heat sinks can be achieved.

[0052] The method for detecting near-surface defects in heat sinks based on active heating includes the following steps: S10: The angle adjustment module adjusts the spatial orientation of the corrugated heat sink so that the angle between the optical axis of the infrared thermal imaging thermometer and the normal of the area to be measured on the surface of the corrugated heat sink is less than 30°. This can eliminate imaging errors caused by surface reflection and achieve accurate acquisition of the changes in heat conduction characteristics in the three regions of corrugated top, corrugated valley, and corrugated sloping wall.

[0053] S20: The fiber-coupled semiconductor laser applies a Gaussian beam thermal excitation to the area to be inspected on the surface of the corrugated heat sink through a laser detection head.

[0054] S30: The infrared thermal imaging thermometer acquires in real time the thermal infrared image sequence of the area to be inspected on the surface of the corrugated heat sink during thermal excitation and subsequent cooling process, and transmits the thermal infrared image sequence to the computer module.

[0055] S40: The computer module receives the thermal infrared image sequence and performs the following steps on the thermal infrared image sequence: S41: Preprocess and time-align the thermal infrared image sequence.

[0056] S42: Extract the thermal conductivity characteristic parameters of the area to be inspected on the surface of the corrugated heat sink in the thermal infrared image sequence.

[0057] S43: Compare the thermal conductivity characteristic parameters with the preset defect-free benchmark model and output the comparison results.

[0058] S44: Determine whether there are near-surface defects based on the comparison results; if so, output the location, size estimate, and type information of the near-surface defects.

[0059] In step S20, when a Gaussian-distributed thermal pulse is applied to the surface of the corrugated heat sink by a fiber-coupled semiconductor laser (70–640W adjustable), heat diffuses inward according to an unsteady-state heat conduction equation: Where α is the thermal diffusivity of the material (unit: m) 2 / s)(Titanium alloy ≈ 7.5 × 10 -6 m 2 / s, 304 stainless steel ≈ 4.2 × 10 -6 m 2 / s). Heat transfer within an object follows the law of thermal conductivity, but near-surface defects (such as cracks, inclusions, debonding, etc.) disrupt the continuity of the material, causing abrupt changes in local thermal conductivity characteristics. If the defect is an air gap (such as a crack), due to the air thermal diffusivity (≈2×10), -5 m 2 The / s is much lower than that of metals (titanium alloy ≈ 9×10). -6 m 2 / s, 304 stainless steel ≈ 5×10 -6 m 2 The heat rises rapidly at the defect front, causing heat to accumulate and creating hot spots due to localized temperature anomalies. If the defect is a low-thermal-conductivity inclusion (such as a non-metallic impurity), it will hinder heat diffusion, forming localized cold spots. These thermal anomalies exhibit specific evolutionary patterns over time (such as differences in heating rate, peak temperature, and cooling rate), serving as the physical basis for defect identification. The temperature difference between the abnormal and normal regions... It can be quantified by time-domain relative thermal contrast (RTC) and become a direct basis for defect identification.

[0060] To amplify the difference in thermal response between defective and normal regions, this invention employs active thermal excitation using a semiconductor laser. Its core principle is to achieve targeted energy excitation of the defective region through precise control of laser parameters. Where Q is the heat absorbed by the workpiece (J), P is the laser power (W), t is the irradiation time (s), and η is the material absorptivity (titanium alloy ≈ 0.3, 304 stainless steel ≈ 0.25). The power is dynamically adjusted within the range of 70W-640W according to the heat sink material (titanium alloy / 304 stainless steel) to ensure that the thermal anomaly signal at the defect reaches the detectable threshold (≥0.5℃ temperature difference) without damaging the workpiece. Then, the laser is set to a Gaussian spot (circular or elliptical) with a diameter of 50mm using a DOE diffraction optical element, and can scan along the ripple direction to ensure that the heat is evenly distributed along the heat sink texture, avoiding excitation blind spots caused by structural shadows. This is particularly relevant for metals with high thermal conductivity (thermal diffusivity ≈ 10). -5 m 2 The rapid decay of thermal response caused by the / s signal is addressed by using millisecond-level pulse control synchronized with the infrared acquisition timing to ensure the optimal time window for capturing defect thermal anomalies.

[0061] In some embodiments, in step S30, the infrared thermal imaging thermometer converts thermal anomalies into quantifiable image signals, and the key parameters are calculated as follows: Thermal sensitivity (NETD): in, To ensure the minimum resolvable temperature difference (<50mK) and signal-to-noise ratio (SNR) of ≥30dB, the system captures changes down to the 0.1℃ level. The rotation angle θ of the corrugated heat sink is adjusted via an angle adjustment module, ensuring the angle between the optical axis of the infrared thermal imaging thermometer and the normal to the inspected area of ​​the corrugated heat sink is <30°. This minimizes temperature measurement errors caused by surface reflection (reflection error can be reduced to within ±0.2℃), enabling precise acquisition of thermal conduction characteristics in the corrugated top, valley, and sloping wall regions. An infrared camera with a thermal sensitivity of <50mK and a sampling rate of ≥50Hz is used to capture minute temperature changes on the metal surface down to the 0.1℃ level. The long-wave infrared band (8-14μm) is adapted to the low emissivity of the metal, improving the signal-to-noise ratio. Finally, the system simultaneously records thermal infrared image sequences during thermal excitation (heating stage) and cooling stage, fully capturing the dynamic process of defects from "heat accumulation" to "heat dissipation," providing data support for multi-dimensional feature analysis.

[0062] In some embodiments, in step S42, extracting the thermal conductivity characteristic parameters of the inspected area of ​​the corrugated heat sink surface in the thermal infrared image sequence includes: collecting and calculating the characteristic parameters of the corrugated top area, corrugated valley area and corrugated sloping wall area; and obtaining the degree of difference of characteristic parameters between the same type of corrugated areas under different defect conditions.

[0063] Specifically, the complex surface of a corrugated heatsink can lead to uneven heat distribution in normal areas (e.g., rapid heat dissipation at the top of the corrugation and slow heat dissipation in the valleys), easily masking defect signals. This invention separates heatsink structural interference from defect signals by performing feature analysis on the corrugated areas: processing thermal infrared images of the corrugated top, valley, and sloping wall regions, and extracting heat conduction characteristic parameters (e.g., thermal diffusion rate at the top and peak temperature in the valley). The key parameters are calculated as follows: Thermal diffusion rate: By calculating the corrugated top, valley, and sloping wall The values ​​are analyzed to determine the degree of difference in similar areas (difference >10% indicates a suspected defect). Then, the thermal conduction characteristic parameters of each area (such as the standard deviation of thermal diffusion rate and the peak arrival time offset) are compared with the preset defect-free benchmark model (standard parts in the same batch or symmetrical defect-free areas).

[0064] In some embodiments, step S43 involves comparing and analyzing the thermal conductivity characteristic parameters with a preset defect-free benchmark model, including: S431: Perform time-domain relative thermal contrast (RTC) analysis: calculate the relative temperature difference between the area under inspection and the reference area at the same time point. ;in, Temperature of the area to be inspected. For the reference region temperature; wherein, in step S431, when If so, it is considered abnormal.

[0065] And / or, S432: Perform thermal diffusion rate difference analysis (reflecting thermal conductivity uniformity): calculate the standard deviation of thermal diffusion rate between the area under test and the reference model at the same location; And / or, S433: Perform peak arrival time offset analysis: calculate the offset between the peak arrival time of the region under inspection and the corresponding time in the reference model. ; And / or, S434: Perform machine learning model discrimination processing: input the heat conduction feature parameters into the neural network classification prediction model to determine the probability of the existence of defects.

[0066] In some embodiments, the reference area is: a defect-free symmetrical corrugated area on the same sample; or, corrugated area data obtained from standard samples in the same batch under the same testing conditions.

[0067] In some embodiments, the benchmark model is: a batch of standard parts or a symmetrical, defect-free area.

[0068] In some embodiments, step S42, extracting the thermal conductivity characteristic parameters of the inspected area on the surface of the corrugated heat sink in the thermal infrared image sequence includes: acquiring and calculating the thermal conductivity characteristic parameters of the corrugated top area, corrugated valley area, and corrugated sloping wall area; and analyzing the difference in characteristic parameters between corrugated areas of the same type under different defect conditions. Different defects refer to defects such as scratches, cracks, and thinning. Same type refers to the same area of ​​the corrugation, such as the corrugated top area.

[0069] In some embodiments, the near-surface defect size estimation method in step S44 includes: Preprocessing of thermal infrared image sequences is performed based on image preprocessing and defect segmentation (fuzzy C-means clustering).

[0070] Based on the pixel area and imaging resolution of thermal anomaly regions in thermal infrared image sequences.

[0071] Spatial calibration was performed by combining the laser spot size and the thermal diffusion model. The thermal diffusion model here is the thermal diffusion model on the corrugated heat sink obtained from the calibration results of the experiment.

[0072] A deep prediction model based on convolutional neural networks is employed. This model identifies and distinguishes between "normal differences caused by structure" and "abnormal differences caused by defects." A defect is identified when the probability of its presence, P_defect, is ≥ 95%. Finally, defect information is generated based on the results of the deep prediction model and a heatmap is overlaid. The defect type, location, and size estimate are then labeled on the heatmap, enabling quantitative detection of near-surface micro-defects in corrugated heat sinks.

[0073] The working principle of this invention can be summarized as follows: Focusing on abnormal heat conduction caused by defects, the signal is amplified through controllable laser excitation, and dynamic thermal response is captured using high-sensitivity infrared imaging. Through corrugation feature analysis and benchmark comparison, rapid and accurate identification of near-surface micro-defects in corrugated heat sinks is achieved. Its innovation lies in the deep integration of "active thermal excitation—dynamic imaging—surface special structure analysis," specifically solving the problem of false detection of defects caused by surface contour changes in corrugated heat sinks. It also achieves precise location and determination of the size information of micro-defects on corrugated heat sinks, realizing fully automated detection from "qualitative discovery" to "quantitative characterization."

[0074] This active heating-based near-surface defect detection device and method for heat sinks can detect microcracks down to 50μm (traditional X-ray sensitivity ≥200μm), with a thermal conductivity decrease detection threshold of 12%. A single detection cycle is less than 30 seconds, representing a 10-fold increase in efficiency compared to manual inspection (traditional visual inspection has a false negative rate greater than 25%), effectively improving detection efficiency. Furthermore, it is compatible with low-emissivity metals such as titanium alloys and stainless steel, and supports the inspection of curved and thin-walled (≥1mm) heat sinks, offering better practicality and applicability.

[0075] It is understood that the above embodiments only illustrate preferred embodiments of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can freely combine the above technical features without departing from the concept of the present invention, and can also make several modifications and improvements, all of which fall within the protection scope of the present invention. Therefore, all equivalent transformations and modifications made with respect to the scope of the claims of the present invention should fall within the scope of the claims of the present invention.

Claims

1. An active heating based fin near surface defect detection apparatus, characterized by, The application relates to a near-surface defect detection method for a corrugated heat sink, comprising the following steps: An angle adjusting module, which comprises a rotating table for carrying and adjusting the spatial posture of the corrugated heat sink; A semiconductor laser thermal excitation module, which comprises a fiber-coupled semiconductor laser and a laser detection head matched with the fiber-coupled semiconductor laser, and is used for applying a Gaussian beam thermal excitation to a to-be-detected area on the surface of the corrugated heat sink; An infrared thermal imaging thermometer, which is used for collecting a thermal infrared image sequence of the to-be-detected area on the surface of the corrugated heat sink during thermal excitation and a subsequent cooling process in real time; A computer module, which is connected with the semiconductor laser thermal excitation module and the infrared thermal imaging thermometer, is used for receiving the thermal infrared image sequence, pre-processing and time sequence alignment of the thermal infrared image sequence, and extracting thermal conduction characteristic parameters of the to-be-detected area on the surface of the corrugated heat sink in the thermal infrared image sequence.

2. The actively-heated fin near-surface defect detection apparatus of claim 1, wherein, The rotating table is used for adjusting the pitch angle of the corrugated heat sink So that the optical axis of the infrared thermal imaging temperature measuring instrument and the normal line of the region to be measured on the surface of the corrugated heat sink are less than 30 degrees.

3. The active heating based fin near surface defect detection apparatus of claim 1, wherein, The power range of the fiber-coupled semiconductor laser is 70W-640W.

4. The actively heated fin near surface defect detection apparatus of claim 1, wherein The infrared thermal imaging thermometer comprises a 640*512 pixel long-wave infrared camera, the sampling rate of which is 100Hz, and the thermal sensitivity is 40mK.

5. An active heating-based near-surface defect detection method for a heat sink, for the active heating-based near-surface defect detection apparatus according to any one of claims 1 to 4, characterized by, The application further discloses a near-surface defect detection device for a corrugated heat sink, which comprises the following modules: S10: The angle adjusting module adjusts the spatial posture of the corrugated heat sink, so that the included angle between the optical axis of the infrared thermal imaging thermometer and the normal line of the to-be-detected area on the surface of the corrugated heat sink is less than 30 degrees; S20: The fiber-coupled semiconductor laser applies a Gaussian beam thermal excitation to the to-be-detected area on the surface of the corrugated heat sink through the laser detection head; S30: The infrared thermal imaging thermometer collects a thermal infrared image sequence of the to-be-detected area on the surface of the corrugated heat sink during thermal excitation and a subsequent cooling process in real time, and transmits the thermal infrared image sequence to the computer module; S40: The computer module receives the thermal infrared image sequence, and performs the following steps on the thermal infrared image sequence: S41: The thermal infrared image sequence is pre-processed and time sequence aligned; S42: Thermal conduction characteristic parameters of the to-be-detected area on the surface of the corrugated heat sink in the thermal infrared image sequence are extracted; S43: The thermal conduction characteristic parameters are compared with a preset defect-free reference model, and a comparison result is outputted; S44: Whether there is a near-surface defect is judged according to the comparison result; If there is, the position, size estimation and type information of the near-surface defect are outputted.

6. The actively heated fin near surface defect detection method of claim 5, wherein, In step S42, the thermal conduction characteristic parameters of the to-be-detected area on the surface of the corrugated heat sink in the thermal infrared image sequence are extracted, which comprises collecting and calculating characteristic parameters of a corrugated top area, a corrugated valley area and a corrugated inclined wall area; and analyzing the difference degree of the characteristic parameters between the same type of corrugated areas under different defect conditions.

7. The actively heated fin near surface defect detection method of claim 5, wherein, In step S43, the thermal conduction characteristic parameters are compared with the preset defect-free reference model and analyzed, which comprises: S431: Perform time domain relative thermal contrast analysis processing: calculate the relative temperature difference of the region under test and the reference region at the same time point ; wherein, is the temperature of the region under test, is the temperature of the reference region; And / or, S432: thermal diffusion rate difference analysis processing: calculating the thermal diffusion rate standard deviation of the to-be-detected area and the reference model at the same position; And / or, S433: peak arrival time offset analysis processing: calculate the offset of the peak arrival time of the region under test and the time of the corresponding position in the reference model ; And / or, S434: performing machine learning model discrimination processing: inputting the heat conduction characteristic parameter into a neural network classification prediction model to discriminate the defect existence probability.

8. The actively heated fin near surface defect detection method of claim 7, wherein, The reference region is: a symmetrical corrugated region without defects on the same piece; or, corrugated region data obtained under the same detection condition from a standard sample in the same batch.

9. The actively heated fin near surface defect detection method of claim 7, wherein, The reference model is: a standard piece in the same batch or a symmetrical region without defects.

10. The actively heated fin near surface defect detection method of claim 7, wherein, In step S431, when then it is determined that an abnormality exists.

Images