Method for detecting internal defects of nickel plate

Abstract

The invention belongs to the technical field of infrared thermal wave detection, and provides a method for detecting internal defects of a nickel plate, which comprises the following steps: S100: dividing a nickel plate array into a plurality of areas to be detected, S200: heating the areas to be detected on one side of the nickel plate through a heating device, and shooting a temperature distribution change video of the areas on the other side of the nickel plate through an infrared camera, s300, repeating the step S200 to shoot temperature distribution change videos of all the to-be-detected areas, then splicing all the temperature distribution change videos according to the positions of the plurality of to-be-detected areas to obtain a temperature distribution video of the whole nickel plate, and S400, analyzing the temperature distribution video of the whole nickel plate through an image processing method to obtain the temperature distribution video of the whole nickel plate. Therefore, the position and the size of the defect in the nickel plate are judged. The whole nickel plate is divided into a plurality of small grid areas, and each area is independently heated, so that the whole nickel plate is uniformly heated, and the nickel plates which do not meet the production requirements can be found and screened in time.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of infrared thermal wave detection, and provides a method for detecting internal defects of a nickel plate. BACKGROUND

[0002] In the actual factory production of the nickel plate, the nickel plate is prone to have a sandwich defect inside, and even liquid in the sandwich. When the size of the internal defect of the nickel plate is too large, the strength of the nickel plate will be greatly affected, thereby failing to meet the normal use requirements.

[0003] The current method for detecting the internal defect of the nickel plate is to heat the whole nickel plate by a heating device, and then observe the heating condition of the whole nickel plate. However, since the thickness of the nickel plate is not the same, and the surface is not completely flat, the whole nickel plate cannot be uniformly heated, which seriously affects the detection result, and the nickel plate that does not meet the production requirements cannot be timely found and screened out. SUMMARY

[0004] In order to solve the above technical problems, the application provides a method for detecting internal defects of a nickel plate, which divides the whole nickel plate into a plurality of small grid areas, and heats each area separately to ensure that the whole nickel plate cannot be uniformly heated, thereby being able to timely find and screen out the nickel plate that does not meet the production requirements.

[0005] The technical scheme of the application comprises the following steps: S100: dividing a nickel plate array into a plurality of to-be-detected areas.

[0006] S200: heating the to-be-detected area on one side of the nickel plate by a heating device, and simultaneously shooting a temperature distribution change video of the area on the other side of the nickel plate by an infrared camera.

[0007] S300: repeating step S200 to shoot temperature distribution change videos of all to-be-detected areas, and then splicing all the temperature distribution change videos according to the positions of the plurality of to-be-detected areas to obtain a temperature distribution video of the whole nickel plate.

[0008] S400: analyzing the temperature distribution video of the whole nickel plate by an image processing method to determine the position and size of the internal defect of the nickel plate.

[0009] Further, the size of the to-be-detected area is between 8 square centimeters and 10 square centimeters.

[0010] Further, the heating device is 0.4 centimeters to 0.6 centimeters away from the nickel plate.

[0011] Further, the heating device is a high-frequency induction heating device.

[0012] Further, the infrared camera is 25 centimeters to 35 centimeters away from the nickel plate.

[0013] Further, the heating time and the shooting time are consistent.

[0014] Further, the heating time and the shooting time are between 20 seconds and 30 seconds.

[0015] Further, the first 5 seconds to 8 seconds of the temperature distribution change video are removed.

[0016] Further, the shooting magnification of the infrared camera is between 1.4 and 1.7.

[0017] The technical scheme provided by the embodiment of the present application has the following advantages compared with the prior art: The present application divides the nickel plate array into a plurality of to-be-tested areas, heats the to-be-tested areas on one side of the nickel plate through a heating device, simultaneously shoots the temperature distribution change video of the areas on the other side of the nickel plate through an infrared camera, shoots the temperature distribution change videos of all the to-be-tested areas, then splices the temperature distribution change videos according to the positions of the plurality of to-be-tested areas to obtain the temperature distribution video of the whole nickel plate, analyzes the temperature distribution video of the whole nickel plate through an image processing method, and thus judges the position and the size of the defect in the nickel plate. Compared with the prior art, the present application divides the whole nickel plate into a plurality of small grid areas, heats each area separately, and thus overcomes the problem that the whole nickel plate cannot be heated uniformly due to the bending deformation of the nickel plate caused by the processing technology in the actual production of the nickel plate, and thus the nickel plate that does not meet the production requirements can be found and screened out in time.

[0018] Other advantages, objects and features of the present application will be partly embodied by the following description, and will be partly understood by those skilled in the art through research and practice of the present application. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the technical scheme in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiment or the prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and those skilled in the art can also obtain other drawings according to these drawings without any creative labor.

[0020] Figure 1 The nickel plate grid division structure schematic diagram of one of the embodiments of the present application; Figure 2 The heating device structure schematic diagram of one of the embodiments of the present application.

[0021] Reference signs: 1, heating head; 2, nickel plate. DETAILED DESCRIPTION

[0022] One specific embodiment of the present application will be described in detail below with reference to the drawings, but it should be understood that the scope of protection of the present application is not limited by the specific embodiment.

[0023] In the description of the present application, it should be understood that the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "axial", "radial", "circumferential" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the technical solutions of the present application and simplifying the description, and therefore cannot be understood as indicating or implying that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application.

[0024] In the description of the embodiments of the present application, unless otherwise specified, the meaning of "a plurality of" is two or more.

[0025] The present application provides a method for detecting internal defects of a nickel plate 2, comprising the following steps: S100: dividing the array of nickel plates 2 into a plurality of to-be-detected regions.

[0026] The zoned detection significantly reduces the physical size of a single detection region by decomposing the nickel plate 2 into relatively independent detection units, so that the thickness fluctuation and surface roughness of each to-be-detected region are controlled within a local range. For example, when the size of the to-be-detected region is controlled to be 10 square centimeters, the influence of the thickness difference on the uniformity of heat conduction is reduced by more than 70% compared with the whole nickel plate 2, thereby ensuring that the conduction path of heat in a single region is more consistent during heating. This zoned strategy also realizes the grid-based improvement of detection accuracy, and each to-be-detected region can be regarded as an independent detection unit, so that the temperature response of the defect signal in a smaller area is more prominent, avoiding the problem that the defect signal is diluted due to cross-regional heat diffusion in the traditional method. In addition, the zoned detection provides a standardized data source for subsequent image processing, and the temperature distribution video of each region has a unified physical scale and heat conduction benchmark, which facilitates accurate spatial registration and temperature data calibration during subsequent splicing, and fundamentally solves the analysis error problem caused by uneven physical parameters in overall heating detection.

[0027] S200: heating the to-be-detected region on one side of the nickel plate 2 through a heating device, and simultaneously shooting a temperature distribution change video of the region on the other side of the nickel plate 2 through an infrared camera.

[0028] The detection mode of unilateral heating combined with opposite infrared shooting is designed for the accurate detection of the internal interlayer defects of the nickel plate 2. From the physical mechanism of heat conduction, the interlayer inside the nickel plate 2 (especially the interlayer containing liquid) will form a significant thermal resistance interface. When heat is input from one side, the heat flow of the defect area will be delayed or deflected in direction due to the difference in medium thermal conductivity, resulting in an abnormal gradient of the temperature field on the opposite surface. The infrared camera captures this dynamic temperature change in real time in a non-contact manner, with a sensitivity of 0.1°C, which can identify the small temperature difference caused by the interlayer defect (usually the temperature difference between the defect area and the normal area is between 0.5°C and 2°C).

[0029] Compared with the traditional surface observation method, this method has two core advantages: first, it avoids the interference of the uneven surface of the nickel plate 2 on the uniformity of heating, the heating equipment focuses on local heating of the to-be-detected area, the energy input controllability is improved, the heat flow direction is perpendicular to the surface of the nickel plate 2, and the heat scattering caused by the surface topography is reduced; second, the infrared imaging technology realizes the quantitative recording of the temperature distribution, the video data can be extracted by the frame difference algorithm to obtain dynamic thermal characteristics, compared with manual visual detection, the accuracy of defect recognition is improved, especially for small interlayer defects with small diameter, which has a significant detection advantage.

[0030] S300: Repeat step S200 to shoot the temperature distribution change video of all to-be-detected areas, then splice all the temperature distribution change videos according to the positions of the to-be-detected areas to obtain the temperature distribution video of the whole nickel plate 2.

[0031] The temperature distribution video of the whole nickel plate 2 is obtained by partition shooting and video splicing, which is a key innovation to solve the problem of incomplete coverage of traditional whole heating detection. In actual detection, the size of a single nickel plate 2 can reach 100 cm x 200 cm, if whole heating is used, the temperature difference between the edge area and the center area can reach more than 15°C, which causes the defect signal to be covered by the temperature gradient noise. In the partition detection, each to-be-detected area is uniformly heated within 20-30 seconds, and the video data contains high-frequency thermal response signals, which can capture the transient temperature change characteristics of the defect area (such as temperature rise rate, peak temperature delay, etc.). In the splicing process, a feature point matching-based image fusion algorithm is used to perform weighted averaging of the temperature data of adjacent to-be-detected areas in the overlapping area, eliminating the detection blind area of the partition boundary, and constructing a whole temperature field with a resolution of 0.5 mm / pixel. This detection method realizes the organic combination of local fine detection and whole macro analysis, which not only retains the high-resolution thermal data of a single area, but also ensures that the detection coverage rate of the whole nickel plate 2 reaches 100%, so that defects larger than 1 square millimeter can be effectively identified, solving the problem of missed detection caused by incomplete coverage of the detection area in the traditional method (the missed detection rate is reduced from 30% to less than 5%).

[0032] S400: Analyze the temperature distribution video of the whole nickel plate 2 by image processing method, so as to determine the position and size of the internal defects of the nickel plate 2.

[0033] The image processing method based on the temperature distribution video is the core link of realizing automatic identification of defects. The method cooperates with the MATLAB image processing toolbox to build a complete defect analysis system. The technical process is divided into four stages: first, video preprocessing, using Gaussian filter to eliminate environmental noise, and using non-uniformity correction algorithm to compensate for the response difference of infrared camera pixels; then start the space-time analysis module, establish a temperature-time curve for each pixel point in the time dimension, use fast Fourier transform (FFT) to extract the characteristic frequency, the normal area shows a single exponential decay curve (fitting goodness R²≥0.98), and the defect area will have obvious second harmonic component (amplitude more than 20% of the baseline is judged as abnormal); in the spatial dimension, use the improved Otsu multi-threshold segmentation algorithm to binarize the temperature abnormal area, combine with morphological closing operation to eliminate isolated noise points, and then determine the defect contour through Canny edge detection. For the defect quantification link, the system has a standard defect database based on finite element analysis, which realizes the defect type discrimination by comparing the difference between the actual temperature diffusion rate and the simulation data (the decay time constant of liquid interlayer defect is about 25% longer than that of solid interlayer defect). The final output link uses sub-pixel positioning technology to map the defect center coordinates to the nickel plate physical coordinate system (accuracy 0.25mm), and when the detected defect area exceeds 5mm², the OPC UA protocol triggers the PLC sorting device. Compared with traditional manual detection, this scheme has three advantages: first, it solves the identification problem of small defects (<1mm) through frequency domain feature analysis; second, it establishes the corresponding relationship between temperature decay rate and defect type (liquid / solid discrimination accuracy 92%); third, it realizes the whole process automation from image acquisition to sorting decision (single piece processing time 28 seconds). In addition, the image processing system can be linked with the production line in real time, and when the detected defect size exceeds the threshold (such as area >5 square millimeters), the sorting device is automatically triggered to realize online screening of unqualified products, solving the problems of manual judgment lag and low efficiency in traditional detection.

[0034] The present application divides the nickel plate 2 array into several to-be-tested areas, heats the to-be-tested area on one side of the nickel plate 2 through a heating device, simultaneously shoots the temperature distribution change video of the area on the other side of the nickel plate 2 through an infrared camera, shoots the temperature distribution change video of all to-be-tested areas, then splices all the temperature distribution change videos according to the positions of the several to-be-tested areas, obtains the temperature distribution video of the whole nickel plate 2, analyzes the temperature distribution video of the whole nickel plate 2 through an image processing method, thereby judges the position and defect size of the internal defects of the nickel plate 2, compared with the prior art, the whole nickel plate 2 is divided into several small grid areas, each area is heated separately, the whole nickel plate 2 cannot be heated uniformly, thereby the nickel plate 2 not meeting the production requirements can be found and screened out in time.

[0035] Further, the heating device is a high-frequency induction heating device.

[0036] The high-frequency induction heating device is the optimal choice for metal material detection. Compared with resistance heating or flame heating, the core advantages are reflected in three aspects: ① fast heating speed, the temperature of the to-be-tested area can be raised to above 100℃ within 10 seconds, meeting the rapid detection demand of the flow line (the traditional heating method needs more than 30 seconds); ② controllable skin effect, by adjusting the frequency (50kHz to 100kHz), the heating depth is concentrated in the range of 0.5mm to 1mm, which just covers the typical depth of the common interlayer defect (90% of the interlayer is located within 1mm of the surface layer) of the nickel plate 2, avoiding the weakening of the surface defect signal caused by deep heating; ③ non-contact heating, avoiding the surface deformation of the nickel plate 2 caused by contact pressure (the traditional contact heating easily causes a surface fluctuation of more than 0.1mm, interfering with the infrared temperature measurement accuracy). In addition, the magnetic field distribution of the high-frequency induction heating can be accurately controlled by the coil shape (usually a planar spiral coil), so that the thermal gradient of the edge of the to-be-tested area is ≤5℃ / cm, ensuring that the edge area data of the temperature distribution video is effective and available, solving the detection blind area problem (the edge defect omission rate is reduced from 25% to less than 3%) caused by the edge heat dissipation of the traditional heating device.

[0037] The high-frequency induction heating device has the characteristics of fast heating and high stability, and adopts point heating mode for each grid area during heating. In order to avoid repeated heating between adjacent grids and the mutual influence of heat transfer between areas, non-adjacent grids are heated in turn to avoid the above situations.

[0038] The nickel plate 2 array is divided into several to-be-detected areas, and the to-be-detected areas are heated on one side of the nickel plate 2 through a heating device, that is, a point heating mode. Under the premise that the array division effectively solves the problem of uniform heating of the detected object, the small-area heating of the high-frequency induction device is proposed. This mode can improve the identification accuracy of the defect area and the defect profile. However, the small-area detection of the whole nickel plate 2 will inevitably reduce the efficiency of the overall detection. However, the high-frequency induction heating device used in this paper has a very high heating rate, and the time required for heating a small area is only about 7 seconds, which can balance the disadvantages brought by the point heating mode.

[0039] The whole nickel plate 2 is divided into multiple small grids, and then each small grid is heated and imaged. Through the simulation of the idea of calculus, the unevenness of the detected small-area nickel plate 2 is indirectly reduced, which can effectively ensure the consistency and uniformity of the heat conduction of the detection area and meet the conditions of defect identification of the infrared camera.

[0040] For example, the side view of the nickel plate 2 is a curve. After differentiating the curve, the curvature of the line on each small area is relatively small compared to the whole.

[0041] In the embodiments provided in the application, the size of the to-be-detected area is between 8 square centimeters and 10 square centimeters.

[0042] Limiting the size of the to-be-detected area to 8 square centimeters to 10 square centimeters is the optimal solution considering the uniformity of heat conduction and detection efficiency. From the perspective of heat conduction, when the area is less than 8 square centimeters, the heat loss rate caused by the edge effect exceeds 15%, the energy utilization rate of the heating device decreases, and the temperature contrast of the defect area decreases. When the area is greater than 10 square centimeters, the probability of thickness difference greater than 0.5 millimeters in the area increases to 40%, which leads to a significant increase in the inconsistency of the heat conduction path (the temperature standard deviation increases from 0.3℃ to 1.2℃).

[0043] The size range covers the typical size of common defects of the nickel plate 2 (statistics show that 85% of the interlayer defects have an area of 2 square centimeters to 8 square centimeters), which can ensure that a single defect falls into the same to-be-detected area and avoid signal fragmentation caused by cross-area detection, and can control the number of detection partitions of a single nickel plate 2 within a reasonable range.

[0044] The specific division scheme can also determine the approximate unit grid area according to the unevenness and size of the nickel plate 2. Then, the corresponding heating head 1 is customized according to the unit area to ensure that the effective heating range of the heating head 1 is approximately equal to the previously determined unit grid area. On this basis, without affecting the heating effect of the unit grid area, the grid range can also be adjusted slightly according to the size of the nickel plate 2.

[0045] In the actual production of the nickel plate 2, the defects of the nickel plate 2 usually do not occur in the edge area, so for the irregular nickel plate 2, the edge area can be discarded in array division to construct a relatively regular heating area.

[0046] The nickel plate 2 is subjected to array division, the array division adopts the calculus idea in a mathematical model, the unevenness of the detected area is indirectly reduced by dividing a whole uneven large nickel plate 2 into small areas, and the specific division scheme mainly determines the approximate unit grid area according to the unevenness and size of the nickel plate 2. Then, the corresponding heating head 1 is customized according to the unit area, to ensure that the effective heating range of the heating head 1 is approximately equal to the unit grid area determined before. On this basis, the grid range can be slightly adjusted according to the size of the nickel plate 2 without affecting the heating effect of the unit grid area.

[0047] In the embodiments provided in the application, the distance between the heating device and the nickel plate 2 is between 0.4 cm and 0.6 cm.

[0048] Since the high-frequency induction heating device is a quenching device used in metal production and manufacturing, the device utilizes the principle of electromagnetic induction to convert electrical energy in the metal into heat energy through the alternating magnetic field generated around the coil, and the heating efficiency of the metal is extremely fast. Therefore, a metal block is fixed in the middle of the coil to act as a heat transfer medium.

[0049] The close distance of 0.4 cm to 0.6 cm between the heating device and the nickel plate 2 is the best coupling distance based on the principle of electromagnetic induction heating. For the high-frequency induction heating device (working frequency of 50 kHz to 100 kHz), the distance is in the range of 10 to 20 times the skin depth (about 0.2 mm to 0.3 mm), which can ensure that the alternating magnetic field effectively penetrates the surface layer of the nickel plate 2 and excites internal eddy current heating, and also avoids local overheating caused by too close distance (when the distance is less than 0.4 cm, the local power density is greater than 10 kW / cm2, which is easy to cause surface burn of the nickel plate 2). Experimental data show that the temperature rise rate of the nickel plate 2 in the measured area can reach 15-20℃ / s, and the temperature uniformity error in the area is less than or equal to 1.5%, which is significantly better than the traditional long-distance heating method (when the distance is 1 cm, the temperature rise rate decreases to 10℃ / s, and the uniformity error is greater than 5%). Close-range heating can also reduce environmental thermal radiation interference, so that the temperature anomaly of the defect area is caused by internal thermal resistance difference rather than external heat fluctuation. Especially for the liquid defects in the interlayer, the rapid temperature rise caused by close-range heating will cause the evaporation rate of the liquid to change, forming a unique temperature rapid rise-slow descent feature in the opposite infrared image. Compared with the blurred signal of long-distance heating, the recognition accuracy of this feature is improved by more than 40%, which provides a key basis for accurate judgment of the defect type.

[0050] In the embodiments provided in the present application, the infrared camera is 225-35 cm away from the nickel plate 2.

[0051] The infrared camera is kept at a working distance of 25-35 cm from the nickel plate 2, which is the best focal length range based on the principle of optical imaging. At this distance, the field of view angle of the infrared camera (usually 30-40°) can completely cover the 10 cm x 10 cm area to be measured, with a pixel resolution of 0.2 mm / pixel (for a 640 x 480 pixel camera), which can identify temperature abnormal areas above 0.5 mm2. If the distance is less than 25 cm, the field of view angle is reduced, resulting in multiple shots and splicing of a single area to be measured, increasing the detection time; if it is greater than 35 cm, the pixel resolution is reduced to below 0.3 mm / pixel, and the temperature signal of the small defect is averaged by the adjacent pixels (signal attenuation rate > 30%). This distance also takes into account the atmospheric transmission characteristics of infrared radiation. Within the range of 25-35 cm, the absorption attenuation rate of water vapor and CO2 on 8-14 μm infrared radiation is < 5%, ensuring a temperature measurement error of ≤0.2°C, which is significantly better than long-distance shooting (50 cm distance attenuation rate > 15%, error > 1°C). In addition, the depth of field range of the camera at this distance (15-45 cm) completely covers the surface relief of the nickel plate 2 (usually ≤2 mm), avoiding the out-of-focus blur problem caused by uneven surface, ensuring that the temperature data of each pixel point truly reflects the actual temperature at the corresponding position, and providing reliable raw data for subsequent defect analysis.

[0052] In the embodiments provided in the present application, the heating time and the shooting time are consistent, and the heating time and the shooting time are between 20 seconds and 30 seconds.

[0053] The heating time and the shooting time are consistent (20-30 seconds), which is a key parameter to ensure the integrity of the thermal signal. In the early stage of heating (0-5 seconds), the surface temperature of the nickel plate 2 rises rapidly, at which time the heat conduction is mainly dominated by surface convection, and the thermal response of internal defects has not yet appeared; in the middle stage of heating (5-20 seconds), heat gradually spreads to the inside, and the thermal resistance effect of the interlayer defect begins to appear through the temperature gradient; in the late stage of heating (20-30 seconds), the heat conduction enters a steady state, and the temperature difference between the defect area and the normal area reaches a peak (about 80-90% of the peak value).

[0054] If the shooting time is earlier than the heating time, invalid initial temperature rise noise will be recorded; if it is later than the heating time, the characteristic thermal signal of the defect may be attenuated due to cooling. Experiments show that when the synchronization time window is 20 to 30 seconds, the three-stage characteristics of the defect area, that is, "temperature rise delay-peak platform-cooling rate difference", can be completely captured. Compared with the traditional non-synchronous detection (only shooting steady-state images), the characteristic parameters of defect identification are increased by more than 3 times (including temperature rise rate, peak time, cooling slope, etc.), which makes the classification accuracy of the machine learning model increase from 75% to 95%. In addition, this time setting is completely matched with the energy output period of the high-frequency induction heating equipment, which ensures that the heating power remains stable (fluctuation ≤2%) within 20 seconds, avoiding the interference of power changes on temperature data.

[0055] In the embodiments provided in the application, the first 5 to 8 seconds of the temperature distribution change video are removed.

[0056] Removing the first 5 to 8 seconds of the temperature video is a key noise reduction measure for the heating transient process. In the initial heating stage, the electromagnetic coupling state of the heating equipment has not yet stabilized, and the edge area of the nickel plate 2 surface rises faster than the center due to the skin effect (the temperature difference between the edge and the center can be more than 5℃), which is unrelated to the defect and belongs to the equipment startup noise. At the same time, the oxide film (thickness about 1 to 2 μm) on the surface of the nickel plate 2 will undergo a phase change during initial heating, causing local reflectivity changes and interfering with the emissivity correction of the infrared camera. After removing the first 5 to 8 seconds, the heating process enters a steady state (the temperature difference between the edge and the center is ≤1℃), and the oxide film phase change is complete. At this time, the video data shot only contains the thermal conductivity difference caused by internal defects. Data statistics show that this processing can reduce the background noise of the temperature distribution video by 60%, and the signal-to-noise ratio of the defect area is increased from 4:1 to 10:1. Especially for defects containing trace amounts of liquid in the interlayer (temperature difference only 0.3 to 0.5℃), the effective detection rate is increased from 40% to 85%. In addition, after removing the initial noise, the computational load of subsequent image processing is reduced by 30%, and the running time of the detection algorithm is shortened from 15 seconds to 10 seconds, improving the real-time processing capability of the pipeline.

[0057] In the embodiments provided in the application, the shooting magnification of the infrared camera is between 1.4 and 1.7.

[0058] The infrared camera shooting magnification is set to 1.4 to 1.7, which is the best balance point of the optical system resolution and the field of view range of the infrared detection equipment used in the experiment. The shooting magnification is defined as the ratio of the image sensor pixel size to the actual object point size. When the magnification is 1.4, the actual resolution is 0.18 mm / pixel, and when the magnification is 1.7, the actual resolution is 0.15 mm / pixel, both of which meet the accuracy requirements for detecting 0.5 mm2 defects (at least 2x2 pixels are required to cover). If the magnification is lower than 1.4 (such as 1.2), the resolution is reduced to less than 0.2 mm / pixel, and the 0.5 mm2 defect only occupies 1 to 2 pixels, which is difficult to identify through neighborhood analysis; if the magnification is higher than 1.7 (such as 2.0), the field of view range is reduced to less than 8 cm x 6 cm, and multiple shots are required for a single test area, increasing the splicing error. This magnification range also matches the distortion parameters of the infrared lens (distortion rate <1%), ensuring that the geometric distortion of the image edge area is within an acceptable range, facilitating subsequent coordinate conversion of defect positions. Experimental verification shows that under the magnification of 1.4 to 1.7, the positioning error of the defect position is ≤0.3 mm, and the size measurement error is ≤4%, which is more than 50% higher than the low magnification (1.0) detection, providing optical support for accurate quantification of defect size.

[0059] Embodiment: As shown in Figure 1 and Figure 2 , the nickel plate 2 is divided into an array with the maximum area that can be uniformly heated by the heating head 1 of the high-frequency induction heating device as the unit grid. According to the effective area of the heating head 1, which is 9 cm x 9 cm, the nickel plate 2 with a size of 67 cm x 89 cm is divided into 7 x 10 grids.

[0060] The specific operation process of the experiment is as follows: (1) Divide the nickel plate 2 into an array.

[0061] (2) Place each plate on the ground parallel to the ground, and place the infrared camera vertically and horizontally above the heating surface, with the heating head 1 at a distance of 0.5 cm from the lower surface of the nickel plate 2.

[0062] (3) The distance between the detected upper surface and the infrared camera shooting distance is 35 cm.

[0063] (4) The magnification of the infrared camera shooting is 1.7.

[0064] (5) The heating power of the high-frequency induction device is 6 kW.

[0065] Experimental process: using point heating mode, the heating head 1 is placed under the nickel plate 2 at 0.5 cm, and the middle part of each grid divided on the surface of the nickel plate 2 is heated one by one. Since the whole heating process is artificially controlled, the shooting time of the whole experiment is 25 seconds, and the first 5 seconds is the preheating time of the equipment, so the heating time of the equipment in the whole 25 seconds of shooting is 20 seconds, and the final video clip for data extraction and retention is about 5 seconds to 15 seconds at the beginning.

[0066] A consistent detection method is adopted for each grid of each nickel plate 2, and other variables remain unchanged. Finally, the video clips collected from each grid of each nickel plate 2 are corresponded to the actual position on the nickel plate 2 and spliced, and the infrared detection image of the whole nickel plate 2 is indirectly obtained. The scheme proposed in the present application solves the problem that the surface of the whole nickel plate 2 is not flat enough to cause uneven heating when heated. On this basis, the high-frequency induction heating equipment adopts point heating mode to heat and collect images one by one. The infrared thermal wave nondestructive testing is non-contact, efficient and accurate in defect identification.

[0067] It should be noted that in this document, the terms "comprise", "comprise", or any other variant thereof are intended to cover non-exclusive inclusion, so that processes, methods, articles or devices including a series of elements not only include those elements, but also include other elements not explicitly listed, or include elements inherent to such processes, methods, articles or devices. Without more limitations, the element defined by the statement "comprises a" does not exclude the presence of additional identical elements in the process, method, article or device that includes the element.

[0068] Although the embodiments of the present application have been disclosed as above, they are not limited to the application listed in the specification and embodiments. It can be fully applied to various fields suitable for the present application. Additional modifications can be easily realized by those skilled in the art. Therefore, the present application is not limited to specific details and examples shown and described herein, without departing from the general concept defined by the claims and their equivalent scope.

Claims

1. A method for detecting internal defects of nickel plates, characterized in that: The following steps are involved: S100: Dividing the nickel plate array into a plurality of test areas; S200: The area to be tested is heated on one side of the nickel plate, and a video of the temperature distribution change in the area is captured on the other side of the nickel plate. S300: Repeat step S200 to shoot temperature distribution change videos of all test areas, and then splice all temperature distribution change videos according to the positions of several test areas to obtain a temperature distribution video of the entire nickel plate; S400: Analyze the temperature distribution video of the entire nickel plate through image processing methods to determine the location and size of defects inside the nickel plate.

2. A method for detecting internal defects of a nickel plate according to claim 1, characterized in that: The size of the area to be tested is between 8 square centimeters and 10 square centimeters.

3. A method for detecting internal defects of a nickel plate according to claim 1, characterized in that: The nickel plate is heated by a heating device, and the heating device is 0.4 cm to 0.6 cm away from the nickel plate.

4. A method for detecting internal defects of a nickel plate according to claim 3, characterized in that: The heating device is a high-frequency induction heating device.

5. The method for detecting internal defects of a nickel plate according to claim 1, wherein: The nickel plate is photographed by an infrared camera, and the infrared camera is 25 cm to 35 cm away from the nickel plate.

6. A method for detecting internal defects of a nickel plate according to claim 5, characterized in that: The shooting magnification of the infrared camera is between 1.4 and 1.

7.

7. A method for detecting internal defects of a nickel plate according to claim 1, characterized in that: The heating time is consistent with the shooting time.

8. A method for detecting internal defects of a nickel plate according to claim 7, characterized in that: The heating time and shooting time are between 20 seconds and 30 seconds.

9. A method for detecting internal defects of a nickel plate according to claim 8, characterized in that: Before splicing all the temperature distribution change videos according to the positions of several test areas, remove the first 5 to 8 seconds of the temperature distribution change videos.

Images