A machine vision-based industrial detection method and system for a fully-welded heat exchanger

Abstract

The application relates to the technical field of machine vision industrial detection, and discloses a machine vision-based full-weld heat exchanger industrial detection method and system, which comprises the following steps: acquiring a weld area image of a heat exchanger to be detected, extracting a boundary contour of the weld area through an image preprocessing algorithm; calculating a weld center line according to the boundary contour of the weld area, and dividing the weld center line into a plurality of equal-length detection units; in each detection unit, linear light scanning data of a weld surface and ultrasonic echo signals of the weld area are collected respectively; a three-dimensional topographic model of the weld surface is generated according to the linear light scanning data, and acoustic characteristic parameters in the weld are extracted; after the three-dimensional topographic model and the acoustic characteristic parameters are preprocessed, the pre-trained defect evaluation model is input, and a comprehensive defect index of the weld area is obtained; according to a comparison result of the comprehensive defect index and a specified threshold group, whether the weld area has obvious defects is judged, and comprehensive evaluation of the weld quality is completed.

Description

Technical Field

[0001] The present application relates to the technical field of machine vision industrial inspection, and in particular to a machine vision-based industrial inspection method and system for fully welded heat exchangers. Background Art

[0002] In modern industrial equipment, fully welded plate heat exchangers are widely used due to their efficient heat transfer performance and compact structure. However, their complex weld structure and small gaps between plates make quality inspection a major challenge.

[0003] Traditional manual inspection methods rely primarily on the senses of experienced inspectors and simple tools. These inspectors use powerful flashlights, magnifying glasses, and even endoscopes (for hard-to-see locations) to observe the weld surface, checking for surface cracks, undercuts, weld beads / spatter pores (visible on the surface), lack of fusion / insufficient penetration (surface signs), significant depressions / insufficient height, and severe surface contamination. While these methods offer low initial investment costs, they exhibit low efficiency and a high rate of missed detections in mass production, making inspection consistency difficult to ensure.

[0004] Existing machine vision solutions mostly focus on two-dimensional image analysis. Although they can identify some surface defects, the surface reflection characteristics of the weld area are complex, and the internal structure is invisible. Therefore, there are certain limitations in the quantitative evaluation of three-dimensional weld morphological defects (such as edge depth and weld bead height) and internal leakage.

[0005] Therefore, there is an urgent need to improve the coverage and accuracy of defect identification to meet the needs of industrial production for high-quality inspection. Summary of the Invention

[0006] In order to improve the quality inspection efficiency and accuracy of all-welded heat exchangers, the present application provides an industrial inspection method and system for all-welded heat exchangers based on machine vision.

[0007] In a first aspect, the present application provides a machine vision-based industrial inspection method for all-welded heat exchangers, which employs the following technical solutions:

[0008] A machine vision-based industrial inspection method for fully welded heat exchangers, comprising:

[0009] S1. Obtain an image of the weld area of ​​the heat exchanger to be inspected, and extract the boundary contour of the weld area using an image preprocessing algorithm;

[0010] S2. Calculate the weld centerline according to the boundary contour of the weld area, and divide the weld centerline into a plurality of detection units of equal length;

[0011] S3, linear light scanning data of the weld surface and ultrasonic echo signals of the weld area are respectively collected in each detection unit;

[0012] S4, a three-dimensional reconstruction is performed according to the linear light scanning data to generate a three-dimensional topographic model of the weld surface;

[0013] S5, a spectrum analysis is performed on the ultrasonic echo signals to extract acoustic characteristic parameters of the weld interior;

[0014] S6, the three-dimensional topographic model and the acoustic characteristic parameters are input into a pre-trained defect evaluation model after preprocessing to obtain a comprehensive defect index of the weld area;

[0015] S7, whether the weld area has obvious defects is determined according to a comparison result of the comprehensive defect index and a specified threshold group.

[0016] By adopting the above technical solution, the three-dimensional topographic information of the weld surface is obtained by performing a three-dimensional reconstruction on the weld surface by using the linear light scanning data, so that the surface defects such as the undercut depth and the weld bead height can be quantified; meanwhile, the acoustic characteristic parameters of the weld interior are extracted by performing a spectrum analysis on the ultrasonic echo signals, which can be used to evaluate the potential defects such as cracks or cavities in the weld interior; then, the three-dimensional topographic model and the acoustic characteristic parameters are combined and input into a pre-trained defect evaluation model to form a comprehensive defect index, thereby completing a comprehensive evaluation of the weld quality, so that not only the multi-dimensional detection requirements of the surface and internal defects can be covered, but also the detection efficiency and accuracy are significantly improved.

[0017] Optionally, the method further includes the following steps:

[0018] A comprehensive defect index sequence is formed by acquiring the comprehensive defect indexes of a plurality of weld areas in the recent period;

[0019] The comprehensive defect index sequence is fitted into a comprehensive defect change curve, and a growth trend of the comprehensive defect change curve is calculated;

[0020] The ultrasonic emission frequency is adjusted according to the growth trend value, that is, the greater the growth trend value, the lower the ultrasonic emission frequency, and the smaller the growth trend value, the higher the ultrasonic emission frequency.

[0021] By adopting the above technical solution, the change trend of the comprehensive defect index reflects the distribution law and evolution characteristics of the weld defects. According to the dynamic adjustment of the ultrasonic emission frequency according to the change trend, the balance between the sound wave penetration depth and the resolution can be optimized.

[0022] Optionally, the method further includes the following steps:

[0023] SZ1, for a detection unit with obvious defects, at least three sections along the center of the weld, to obtain a first subdivision unit, the second subdivision unit and the third subdivision unit with less than a specified percentage of each other overlap;

[0024] SZ2, obtain the corresponding comprehensive defect index of each subdivision unit, select the subdivision unit corresponding to the highest comprehensive defect index;

[0025] Determine whether the subdivision unit is centered, if not, reselect the detection unit centered on the subdivision unit, return to step SZ1, otherwise go to step SZ3;

[0026] SZ3, obtain the corresponding comprehensive defect index and unit area of the first non-overlapping unit, the first overlapping subdivision unit, the second non-overlapping unit and the second overlapping subdivision unit, and the third non-overlapping unit contained in the current detection unit;

[0027] SZ4, compare the comprehensive defect index of the non-overlapping unit and the overlapping unit two by two without repetition, and accumulate abnormal values according to the comparison;

[0028] SZ5, if the top two abnormal values are located at the two ends of the detection unit at the beginning of the current round, form a key detection unit centered on the two units respectively, and then go to step SZ3, otherwise, take the two units as the two ends of the next round of detection unit, and then go to step SZ3;

[0029] SZ6, when the area of the detection unit is less than or equal to a specified area threshold, mark as having potential defects, and the loop ends.

[0030] By using the above technical scheme, the position of the potential defect can be located relatively more accurately through the gradual subdivision of the detection unit and the accumulation of abnormal values. The method is especially suitable for local defect identification in complex weld structures, and avoids the missed detection problem caused by the large detection range of the traditional detection method.

[0031] Optionally, the process of accumulating abnormal values according to the comparison includes:

[0032] Configure an abnormal value accumulation coefficient and a cumulative base value of the current round, and the abnormal value is the product of the accumulation coefficient and the cumulative base value; the abnormal value accumulation coefficient is inversely proportional to the unit area, and the cumulative base value is related to whether it is an overlapping area;

[0033] In the two-by-two comparison without repetition, the larger the comprehensive defect index, the more abnormal values are accumulated, and the smaller the comprehensive defect index, the less abnormal values are accumulated.

[0034] By adopting the technical scheme, the calculation manner of the abnormal value combines the influences of the unit area and the detection round, so that the distribution of the abnormal value is more consistent with the spatial characteristics of the actual defect and the dynamic changes of the detection process, and the accuracy of potential defect identification can be effectively improved.

[0035] Optionally, the method further comprises the following steps:

[0036] For the weld area where the potential defect exists, a focused heat source is used to apply a thermal excitation in a direction perpendicular to the center line of the weld;

[0037] During the process before and after the application of the thermal excitation, a time sequence of infrared thermal images is collected;

[0038] The infrared spatio-temporal feature parameters of the time sequence of infrared thermal images are extracted;

[0039] The infrared spatio-temporal feature parameters are input into a pre-trained infrared defect evaluation model after being preprocessed to form an infrared parameter matrix;

[0040] The authenticity of the potential defect is determined according to the output of the infrared defect evaluation model.

[0041] By adopting the technical scheme, the thermal excitation of the focused heat source can make the potential defect of the weld area appear in the infrared thermal image, so as to realize further verification of the authenticity of the defect, and is suitable for confirmation of small defects or hidden defects.

[0042] Optionally, the method further comprises the following steps:

[0043] The spatial position sequence of all potential defects is obtained;

[0044] Based on the spatial position sequence, the intermediate positions between any two adjacent potential defect positions in space are calculated to form an intermediate position sequence;

[0045] The dispersion degree of the intermediate position sequence is calculated;

[0046] If the dispersion degree is less than a preset dispersion threshold, the average value of all intermediate positions in the intermediate position sequence is calculated as a re-inspection recommendation point.

[0047] By adopting the technical scheme, the statistical analysis of the spatial positions of the potential defects can determine the key area for re-inspection, so as to reduce unnecessary repeated detection workload and improve the overall detection efficiency.

[0048] Optionally, the ultrasonic wave emission frequency adopts different first and second frequencies for detection respectively:

[0049] First, a preset first frequency is used to collect a first ultrasonic echo signal of the weld area;

[0050] acquire a second ultrasonic echo signal of the weld area using a preset second frequency;

[0051] extract a first acoustic sub-feature parameter set based on the first ultrasonic echo signal respectively;

[0052] extract a second acoustic sub-feature parameter set based on the second ultrasonic echo signal respectively;

[0053] In the step of calculating the comprehensive defect index of the weld area, the first acoustic sub-feature parameter set and the second acoustic sub-feature parameter set are both input into the pre-trained defect evaluation model.

[0054] By using the above technical solution, the response characteristics of ultrasonic signals of different frequencies to internal defects of the weld are different, so that more comprehensive acoustic feature information can be obtained through multi-frequency detection. After inputting these feature parameter sets into the defect evaluation model, the reliability of the comprehensive defect index can be further improved.

[0055] Optionally, the method further comprises the following steps:

[0056] If it is determined that there is a suspected feature of a specific type of defect based on the first acoustic sub-feature parameter set and / or the second acoustic sub-feature parameter set, and the confidence level is lower than a preset threshold, then a third ultrasonic echo signal of the weld area is acquired using a preset third frequency;

[0057] extract a third acoustic sub-feature parameter set based on the third ultrasonic echo signal;

[0058] In the step of calculating the comprehensive defect index of the weld area, the third acoustic sub-feature parameter set is also input into the pre-trained defect evaluation model.

[0059] By using the above technical solution, when the preliminary detection result is uncertain, the introduction of ultrasonic signals of a third frequency for supplementary detection can further improve the confidence level and accuracy of defect recognition.

[0060] In a second aspect, the application provides a machine vision-based full-weld heat exchanger industrial detection system, which adopts the following technical solution:

[0061] A machine vision-based full-weld heat exchanger industrial detection system comprises a processor, and the processor executes the program of the machine vision-based full-weld heat exchanger industrial detection method according to any one of the above.

[0062] Optionally, the system further comprises an image acquisition module, an ultrasonic detection module, an infrared thermal imaging module, a linear light scanning module, a defect evaluation model, and an infrared defect evaluation model, and the modules work cooperatively through signal transmission and data processing to complete comprehensive detection of the weld area.

[0063] In summary, this application includes at least one of the following beneficial technical effects:

[0064] This application uses linear light scanning data to perform three-dimensional reconstruction of the weld surface to obtain three-dimensional morphological information of the weld surface, so as to quantify surface defects such as edge depth and weld bead height; at the same time, the acoustic characteristic parameters inside the weld are extracted through spectral analysis of the ultrasonic echo signal, which can be used to evaluate potential defects such as cracks or voids inside the weld; the three-dimensional morphological model and the acoustic characteristic parameters are then combined and input into a pre-trained defect assessment model to form a comprehensive defect index, completing a comprehensive assessment of the weld quality. This not only covers the multi-dimensional detection needs of surface and internal defects, but also significantly improves the detection efficiency and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 This is a module connection diagram of the fully welded heat exchanger industrial detection system of this application.

[0066] Figure 2 It is a schematic diagram of the processing of the detection unit in the industrial detection method of the fully welded heat exchanger of this application. DETAILED DESCRIPTION

[0067] Embodiments of the present application are described in detail below, examples of which are illustrated in the accompanying drawings.

[0068] Throughout this specification, reference to the terms "certain embodiments," "one embodiment," "some embodiments," "illustrative embodiments," "examples," "specific examples," or "some examples" means that a specific feature, structure, material, or characteristic described in conjunction with the embodiment or example is included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0069] The embodiment of the present application discloses a method and system for industrial detection of fully welded heat exchangers based on machine vision. Figure 1 For detailed description. Figure 1 As shown in Figure 1, the system includes a processor, an image acquisition module, a linear light scanning module, an ultrasonic detection module, an infrared thermal excitation module, and a defect assessment model. These modules work together through physical connections and data transmission to complete a comprehensive inspection of the weld area of ​​a fully welded plate heat exchanger.

[0070] In actual operation, first, the image acquisition module acquires the image of the weld area of the core or plate bundle of the full-weld plate heat exchanger to be detected. The image acquisition module usually includes an industrial camera and a light source assembly, wherein the industrial camera is installed above the weld and perpendicular to the weld surface, and the light source assembly is arranged on both sides of the weld to ensure uniform illumination of the light.

[0071] The image acquisition module transmits the acquired original image data to the processor, and the processor extracts the boundary contour of the weld area through the built-in image preprocessing algorithm. The extraction process of the boundary contour is based on gray threshold segmentation and edge detection technology, and finally generates accurate boundary information of the weld area by analyzing the pixel difference between the weld area and the background area. On this basis, the processor further calculates the weld center line and divides the weld center line into multiple equal-length detection units. The length of each detection unit can be pre-set according to the actual size of the weld and the detection accuracy requirement.

[0072] Next, the linear light scanning module and the ultrasonic detection module respectively acquire data of the weld area in each detection unit.

[0073] The linear light scanning module adopts the configuration of camera + linear laser + motion mechanism / mirror, and the motion structure can be equipped with a stable linear motor. A laser line can obtain the height information of a cross section, and can quickly complete the scanning of a long weld by cooperating with a motion platform or a mirror, which is very suitable for the rhythm of automatic online detection. The reflected light signal of the weld surface is captured and transmitted to the processor, and the processor performs three-dimensional reconstruction using the received linear light scanning data to generate a three-dimensional topographic model of the weld surface. The construction of the three-dimensional topographic model can adopt the principle of triangulation, analyze the reflected light intensity and angle of different points on the weld surface, and calculate the height variation and geometric characteristics (such as height difference standard deviation, maximum concave depth, local curvature entropy, and wave index) of the weld surface.

[0074] At the same time, the ultrasonic detection module emits ultrasonic signals through an ultrasonic probe and receives echo signals of the weld area. The echo signals are transmitted to the processor after analog-to-digital conversion. The processor performs frequency spectrum analysis on the received ultrasonic echo signals to extract acoustic characteristic parameters inside the weld. The acoustic characteristic parameters include frequency components, amplitude distribution, and phase changes of the echo signals, which are used to evaluate potential defects inside the weld.

[0075] After completing data acquisition, the processor inputs the generated three-dimensional topographic model of the weld surface and the acoustic characteristic parameters inside the weld into a pre-trained defect evaluation model. The defect evaluation model is pre-trained based on the existing Dual-Stream DNN architecture and embedded in the processor, which comprehensively analyzes the input data and outputs a comprehensive defect index of the weld area.

[0076] In the process of obtaining the comprehensive defect index, the geometric features such as height difference and curvature change in the three-dimensional topographic model and the acoustic features such as frequency deviation and amplitude anomaly in the acoustic feature parameters need to be preprocessed, and the geometric feature normalization (which can be based on the plate thickness) and the acoustic feature normalization (which can be based on the defect-free area) are realized.

[0077] Then, according to the size of the comprehensive defect index, the defect level of the weld area is determined. If the comprehensive defect index is less than the first threshold value, it is determined that the weld area has no obvious defects.

[0078] If the comprehensive defect index is between the first threshold value and the second threshold value, it is determined that the weld area has slight defects.

[0079] If the comprehensive defect index is between the second threshold value and the third threshold value, it is determined that the weld area has moderate defects.

[0080] If the comprehensive defect index is greater than the third threshold value, it is determined that the weld area has serious defects.

[0081] In terms of multi-frequency detection, the ultrasonic detection module uses different first and second frequencies for detection. The processor controls the ultrasonic probe to use a preset first frequency to collect the first ultrasonic echo signal of the weld area, and then uses a preset second frequency to collect the second ultrasonic echo signal of the weld area. The echo signals of the two frequencies are respectively subjected to spectral analysis to extract the first and second sets of acoustic sub-feature parameters, and these parameter sets are input into the defect evaluation model to further improve the reliability of the comprehensive defect index. If the preliminary detection result has uncertainty and the confidence is lower than the preset threshold, the processor controls the ultrasonic probe to use a preset third frequency to collect the third ultrasonic echo signal of the weld area, and extracts the third set of acoustic sub-feature parameters as supplementary data.

[0082] Among them, the first and second frequencies and the third frequency can be selected within 20-100 kHz in low frequency, which can penetrate thick plates, and high frequency can be selected within 4.6-5.2 MHz, which can grab surface micro-cracks.

[0083] In order to further optimize the detection effect, the processor can also statistically analyze the comprehensive defect indexes of multiple continuous weld areas to form a comprehensive defect index sequence, and the comprehensive defect index sequence is a comprehensive defect change curve generated by a fitting algorithm, and the processor dynamically adjusts the ultrasonic emission frequency according to the growth trend of the curve.

[0084] It is a common sense in the welding field that it is better to accept visible shallow defects than to miss hidden deep defects; however, the core or plate bundle targeted in the application is an important component of the full-welded plate heat exchanger, and since the heat exchanger may often operate in an environment with corrosive media, surface porosities generated from shallow defects may become the starting point of stress corrosion cracking, and moreover, such welds are usually thin (4-12 mm), at which time shallow undercutting is more dangerous than deep slag inclusion because it directly reduces the pressure-bearing wall thickness. If such defects can be detected at a high coverage rate during the manufacturing stage, the scrap cost can be much lower than that of post-sale failure.

[0085] Therefore, in the present embodiment, the weight of shallow defects on the comprehensive defect index is designed to be higher than that of deep defects. When the growth trend value is large, it indicates that shallow defects dominate and surface defects are likely to be obvious, at which time the ultrasonic emission frequency can be reduced to enhance the detection ability of deep defects and improve the rationality of the comprehensive defect index; when the growth trend value is small, it indicates that shallow defects are not obvious, at which time the ultrasonic emission frequency needs to be increased to further improve the resolution of shallow defects; for deep defects, low-frequency ultrasonic waves can be used for inspection by subsequent sampling inspection.

[0086] In addition, for the detection unit containing at least slight defects, to further determine the defect area, it is divided into three subdivided units with an overlap degree less than 50% along the center line direction of the weld, and the comprehensive defect index of each subdivided unit is recalculated. The processor determines whether the detection unit needs to be reselected according to the position of the subdivided unit corresponding to the highest comprehensive defect index, and calculates the cumulative abnormal value by comparing the comprehensive defect indexes of the non-overlapping unit and the overlapping unit, so as to accurately locate the position of the potential defect.

[0087] In the present embodiment, as shown in Figure 2 Taking a long rectangular detection unit as an example, it can be divided into three subdivided units with an overlap degree of 45% between each other, i.e., the first subdivided unit K1, the second subdivided unit K2, and the third subdivided unit K3, and then the respective comprehensive defect indexes of the three subdivided units are obtained, i.e., KC1, KC2, and KC3.

[0088] If KC1 or KC3 is the largest, a new long rectangular detection unit is reselected with the first subdivided unit K1 or the third subdivided unit K3 as the center, and then the three subdivided units are divided again;

[0089] If KC2 is the largest, it indicates that the probability of defects being concentrated in the middle of the inspection unit is relatively high. Based on the overlap, the current inspection unit can be further divided from left to right into the first non-overlapping unit Q1, the first overlapping subdivision unit A1, the second non-overlapping unit Q2, the second overlapping subdivision unit A2, and the third non-overlapping unit Q3. The first-round comprehensive defect index of each overlapping or non-overlapping unit can be obtained.

[0090] Taking the first non-overlapping unit Q1 and the first overlapping subdivision unit A1 as an example, QC1 represents the comprehensive defect index of the first non-overlapping unit, AC1 represents the comprehensive defect index of the first overlapping unit, SQ1 represents the unit area of ​​the first non-overlapping unit, SA1 represents the unit area of ​​the first overlapping unit, and the same applies to the others.

[0091] Perform a pairwise, non-repetitive comparison of the comprehensive defect indexes of non-overlapping units and overlapping units. If QC1 is less than AC1, AC1 accumulates an outlier value once. If QC2 is greater than AC2, QC2 accumulates an outlier value once. After obtaining the final outlier value accumulation result, sort the outlier value sequence in descending order and obtain the top two units.

[0092] If the first two units are Q1 and Q3 (rarely in actual situations), two new non-overlapping detection units are obtained with Q1 and Q3 as the centers respectively;

[0093] If the first two units are A1 and A2 (which is often the case in practice), A1 and A2 are used as the two ends of the next round of inspection units, and then the next round of subdivision begins. If the area of ​​the inspection unit is less than or equal to the specified area threshold, it is marked as a potential defect and the cycle ends. The specified area threshold can be determined based on the historical defect size.

[0094] In the process of accumulating abnormal values ​​according to the comparison results, first configure the abnormal value accumulation coefficient and the cumulative base value of the current round. Abnormal value = accumulation coefficient * cumulative base value;

[0095] Because in the training process of the defect assessment model, samples with smaller areas are easier to obtain and more in number, and if defect features exist, they are easier to capture. However, welding defects are mostly continuous defects with a small unit area, but the comprehensive defect index can show that the defect features are more obvious. Therefore, the outlier cumulative coefficient is designed to be inversely proportional to the unit area. Since the overlapping area is located within K2, the cumulative basic value corresponding to the overlapping area is larger; in the non-repetitive comparison between the two, the one with a larger comprehensive defect index accumulates an outlier value once, and the one with a smaller comprehensive defect index does not accumulate an outlier value.

[0096] For easier understanding, let's take part of the first round of comparison as an example:

[0097] If QC1 is less than AC1, and SQ1 (55%) is greater than SA1 (45%), then the cumulative abnormal value of A1 this time is 0.8*1=0.8, and Q1 is not cumulative.

[0098] If QC2 is greater than AC2, and SQ2 (55%) is greater than SA2, then the cumulative abnormal value of Q2 this time is 1*0.9=0.9, and A2 is not cumulative.

[0099] In addition, before the start of this round of abnormal value accumulation, the variance of the comprehensive defect index sequence of the non-overlapping unit and the overlapping unit can be calculated, and if the variance value is less than the specified variance value, the overlap degree can be reduced until the variance value is greater than or equal to the specified variance value.

[0100] The infrared thermal excitation module includes a heat source generator and an infrared thermal imager. For the weld area marked with potential defects, the heat source generator uses a laser light source or a high-power density hot air gun to apply a short-time, high-energy density pulse thermal excitation vertically to the center line direction of the weld above the potential defect area. A mid-wave or long-wave infrared thermal imager continuously collects the infrared thermal image time sequence before and after the application of thermal excitation. The selection of the energy density range and the pulse width range of the pulse thermal excitation is related to the material type, thickness range or expected maximum defect depth of the heat exchanger plate being inspected, which can be determined through thermal conductivity simulation and experimental calibration to ensure that the excitation energy can effectively reach the expected defect depth and produce a detectable temperature rise, while avoiding damage to the material surface.

[0101] The infrared thermal imager can be used to continuously collect the infrared thermal image time sequence of the potential defect micro area and the surrounding appropriate expansion area (the expansion range covers at least the micro area boundary) at a sampling frequency of not less than 50Hz, starting 1 second before the application of thermal excitation and ending 5 seconds after the end of thermal excitation.

[0102] Then the infrared thermal image time sequence is processed to extract a set of infrared space-time feature parameters representing the thermal response anomaly of the potential defect micro area, including but not limited to:

[0103] The highest temperature rise value (relative to the baseline temperature before excitation) in the micro area;

[0104] The change curve of the average temperature of the micro area with time t;

[0105] The maximum temperature rise rate of the average temperature rise stage of the micro area;

[0106] The phase delay of the average temperature peak of the micro area relative to the average temperature peak of the defect-free reference area (selected near the micro area);

[0107] The average temperature difference between the micro area and its surrounding immediately adjacent defect-free reference ring (the inner diameter of the ring is equal to the diameter of the circumscribed circle of the micro area, and the outer diameter is determined according to thermal diffusion);

[0108] The pixel area in the micro area whose temperature exceeds the set threshold at a certain time (such as the peak time or the sampling end time);

[0109] The average decreasing rate of the micro area average temperature in the cooling stage within a certain time period.

[0110] Then, the extracted infrared space-time feature parameter set is input into a pre-trained infrared defect evaluation model; the infrared defect evaluation model can output a confidence degree of the existence of defects; when the confidence degree is greater than 90%, it can be confirmed that the micro area truly exists defects, and the infrared evaluation result is recorded, otherwise further verification is needed; wherein the infrared defect evaluation model is trained based on an existing lightweight convolutional recurrent network (Lightweight ConvLSTM).

[0111] Finally, the processor statistically analyzes the spatial position sequence of all potential defects, because according to the above prevention, for each detection unit, considering the quality inspection cost, only one real defect micro area is finally obtained, and the welding defects are often continuous, therefore, in the present scheme, the intermediate position between any two adjacent defect positions in space can be calculated to form an intermediate position sequence, and then whether a re-inspection recommended point needs to be set is judged according to the discrete degree of the intermediate position sequence; if the discrete degree is less than a preset discrete threshold, the processor calculates the average value of all intermediate positions in the intermediate position sequence as the re-inspection recommended point, thereby reducing unnecessary repeated detection workload and improving the overall detection efficiency.

[0112] In summary, the present application realizes efficient and accurate detection of the welding area of the full-welded plate heat exchanger by combining a multi-dimensional cooperative detection method. The modules work cooperatively to ensure the automation and intelligentization of the detection process, significantly improving the detection efficiency and accuracy, and providing strong technical support for industrial production.

[0113] Although the embodiments of the present application have been shown and described above, it should be understood that the above embodiments are exemplary and should not be construed as limiting the present application, and those skilled in the art can make changes, modifications, replacements and variations to the above embodiments within the scope of the present application.

Claims

1. A machine vision-based industrial inspection method for fully welded heat exchangers, characterized in that: include: S1. Obtain an image of the weld area of ​​the heat exchanger to be inspected, and extract the boundary contour of the weld area using an image preprocessing algorithm; S2. Calculate the weld centerline according to the boundary contour of the weld area, and divide the weld centerline into a plurality of detection units of equal length; S3. In each detection unit, linear optical scanning data of the weld surface and ultrasonic echo signals of the weld area are collected respectively; S4, performing three-dimensional reconstruction based on the linear light scanning data to generate a three-dimensional topography model of the weld surface; S5. Perform spectrum analysis on the ultrasonic echo signal to extract acoustic characteristic parameters inside the weld; S6. Preprocessing the three-dimensional morphology model and the acoustic characteristic parameters and inputting them into a pre-trained defect assessment model to obtain a comprehensive defect index of the weld area; S7. judging whether there are obvious defects in the weld area based on the comparison result of the comprehensive defect index and the specified threshold group; The method further comprises the steps of: Obtain comprehensive defect indices of multiple recent weld areas to form a comprehensive defect index sequence; Fit the comprehensive defect index sequence into a comprehensive defect change curve, and calculate the growth trend of the comprehensive defect change curve; Adjust the ultrasonic emission frequency according to the growth trend value. The larger the growth trend value, the lower the ultrasonic emission frequency; the smaller the growth trend value, the higher the ultrasonic emission frequency. The following steps are also included: SZ1: For a test unit with obvious defects, divide it into at least three equal parts along the direction of the weld center to obtain the first subdivision unit, the second subdivision unit, and the third subdivision unit with a mutual overlap of less than a specified percentage; SZ2. Obtain the comprehensive defect index corresponding to each subdivision unit and select the subdivision unit corresponding to the highest comprehensive defect index; Determine whether the subdivision unit is centered. If not, reselect the detection unit with the subdivision unit as the center and return to step SZ1. Otherwise, go to step SZ3. SZ3, obtaining the comprehensive defect index and unit area corresponding to the first non-overlapping unit, the first overlapping subdivision unit, the second non-overlapping unit, the second overlapping subdivision unit, and the third non-overlapping unit contained in the current detection unit; SZ4: Perform a pairwise, non-repetitive comparison of the comprehensive defect indexes of non-overlapping units and overlapping units, and accumulate abnormal values ​​based on the comparison results; SZ5: If the two cells with the top two outlier values ​​are located at the two ends of the detection cell at the beginning of this round, these two cells are set as the center to form the key exploration detection cell and then proceed to step SZ3; otherwise, these two cells are used as the two ends of the next round of detection cells and then proceed to step SZ3; SZ6. When the area of ​​the inspection unit is less than or equal to the specified area threshold, it is marked as a potential defect and the loop ends.

2. The machine vision-based industrial inspection method for all-welded heat exchangers according to claim 1 is characterized in that: The process of accumulating abnormal values ​​according to the comparison situation includes: Configuring an outlier cumulative coefficient and a cumulative base value of the current round, where the outlier value is the product of the cumulative coefficient and the cumulative base value; the outlier cumulative coefficient is inversely proportional to the unit area, and the cumulative base value is related to whether it is an overlapping area; In the pairwise non-repeated comparison, the one with the larger comprehensive defect index is accumulated as an outlier, while the one with the smaller comprehensive defect index is not accumulated as an outlier.

3. The machine vision-based industrial inspection method for all-welded heat exchangers according to claim 1, characterized in that: The following steps are also included: For weld areas marked as potentially defective, a focused heat source is used to apply thermal stimulation perpendicular to the weld centerline. Before and after the thermal excitation is applied, a time series of infrared thermal images is acquired; Extract infrared spatiotemporal feature parameters of infrared thermal image time series; The infrared spatiotemporal characteristic parameters are pre-processed into an infrared parameter matrix and then input into a pre-trained infrared defect assessment model; The authenticity of the potential defect is determined according to the output of the infrared defect assessment model.

4. The machine vision-based industrial inspection method for all-welded heat exchangers according to claim 3 is characterized in that: The following steps are also included: Obtain the spatial position order of all potential defects; Based on the spatial position sequence, the intermediate position between any two adjacent potential defect positions in space is calculated to form an intermediate position sequence; Calculate the degree of dispersion of the intermediate position sequence; If the degree of dispersion is less than the preset discrete threshold, the average value of all the middle positions in the middle position sequence is calculated as the re-inspection recommended point.

5. The machine vision-based industrial inspection method for all-welded heat exchangers according to claim 1, characterized in that: Ultrasonic emission frequencies use different first and second frequencies for detection: First, a preset first frequency is used to collect a first ultrasonic echo signal in the weld area; Then, a second ultrasonic echo signal from the weld area is collected using a preset second frequency; extracting first acoustic sub-feature parameter sets based on the first ultrasonic echo signals respectively; extracting second acoustic sub-feature parameter sets based on the second ultrasonic echo signals respectively; In the step of calculating the comprehensive defect index of the weld area, both the first acoustic sub-feature parameter set and the second acoustic sub-feature parameter set are input into the pre-trained defect assessment model.

6. The machine vision-based industrial inspection method for all-welded heat exchangers according to claim 5, characterized in that: The following steps are also included: If it is determined based on the first acoustic sub-feature parameter set and / or the second acoustic sub-feature parameter set that there are suspected features of a specific type of defect, and the confidence level is lower than a preset threshold, a third ultrasonic echo signal is collected from the weld area using a preset third frequency; extracting a third acoustic sub-feature parameter set based on the third ultrasonic echo signal; In the step of calculating the comprehensive defect index of the weld area, the third acoustic sub-feature parameter set is also input into the pre-trained defect assessment model.

7. A fully welded heat exchanger industrial inspection system based on machine vision, characterized in that: The method comprises a processor, wherein the processor executes the steps of the all-welded heat exchanger industrial detection method based on machine vision according to any one of claims 1 to 6.

8. The machine vision-based all-welded heat exchanger industrial inspection system according to claim 7 is characterized in that: It also includes an image acquisition module, an ultrasonic detection module, an infrared thermal imaging module, a linear light scanning module, a defect assessment model, and an infrared defect assessment model. The modules work together through signal transmission and data processing to complete comprehensive inspection of the weld area.

Images