Automated production line control system and method

Abstract

The present invention relates to the field of welding control, and in particular to a control system and method for an automated production line, the method comprising: obtaining the temperature distribution of a welding area and generating a temperature heat map, and extracting a set of isotherms. A temperature anomaly value is calculated based on the curvature change of a pixel point on the isotherm, and a weighted sum is performed using the distance from the weld point to the isotherm as a weight to obtain a comprehensive anomaly value. At the same time, a sequence of the intersection distances of the weld point and the isotherm in different directions is obtained, the uniformity of the isotherm intervals is analyzed, the coefficient of variation and the mean similarity of the distance sequences in adjacent directions are calculated, and a temperature gradient anomaly value is constructed. Finally, the product of the curvature anomaly value and the comprehensive anomaly value is processed through an activation function to obtain an anomaly degree value, which is compared with a preset threshold value to determine whether there is an anomaly in the welding area, and the welding operation is adjusted accordingly, thereby optimizing the welding process and improving welding quality and production efficiency.

Description

Technical Field

[0001] The present invention relates to the field of welding control, and in particular to an automated production line control system and method. Background Art

[0002] As a high-security and decorative product, armored doors require extremely high precision during the welding process. During the cooling process, areas with large thermal expansion are constrained by the surrounding material and cannot shrink freely, resulting in weld deformation, which affects the subsequent drilling and positioning steps and compromises product quality.

[0003] In the past, welding quality inspection for armored doors relied primarily on manual labor, which was inefficient, highly subjective, and difficult to monitor in real time. This led to inconsistent inspection results and a high incidence of welding defects. With the development of automated production technology, the introduction of infrared thermal imagers, automated equipment, and intelligent control systems has enabled real-time monitoring and precise control of the welding process, improving inspection accuracy and efficiency, reducing human interference, and providing a strong guarantee for the high-quality production of armored doors.

[0004] While automated production has made significant progress in armored door weld inspection, some limitations remain. While monitoring the curvature consistency of isotherms can reflect the uniformity of temperature distribution, relying solely on curvature consistency cannot fully capture temperature anomalies in the weld area. In particular, in areas with relatively gentle temperature gradients, curvature changes may not be noticeable, leading to local anomalies being overlooked. This limits the ability to accurately detect and provide timely feedback on potential issues during the welding process, hindering further improvements in welding quality. Summary of the Invention

[0005] In order to solve the problem that it is difficult to fully capture local anomalies in areas where the temperature gradient changes gently, which is not conducive to adjusting the welding process, the present invention provides solutions in the following aspects.

[0006] In a first aspect, a method for controlling an automated production line includes: obtaining a temperature distribution in a welding area to generate a temperature heat map, preprocessing the temperature heat map, and extracting a set of isotherms from the preprocessed temperature heat map; determining a temperature anomaly value of the isotherm according to a change in curvature of a pixel point on each isotherm in the isotherm set, using the distance between the weld point and the isotherm as a weight of the temperature anomaly value, and performing weighted summation to obtain a comprehensive anomaly value; obtaining a sequence of intersection distances between the weld point and the isotherm in different directions, analyzing the uniformity of the isotherm intervals based on the intersection distance sequence, and obtaining a coefficient of variation of the intersection distance sequence; calculating a mean similarity between distance sequences in all adjacent directions, and constructing a temperature gradient anomaly value based on the coefficient of variation and the mean similarity; using an activation function to multiply the curvature anomaly value and the comprehensive anomaly value to obtain an abnormality degree value of the welding area, judging whether the welding area has an abnormality according to the abnormality degree value and a preset abnormality threshold, and adjusting the welding operation.

[0007] The results are as follows: by obtaining the temperature distribution of the welding area and generating a temperature heat map, the heat map is preprocessed to extract a set of isotherms, and the temperature anomaly value is calculated based on the curvature change of the pixel points on the isotherms. The weighted summation is combined with the distance between the weld point and the isotherms to obtain a comprehensive anomaly value. Furthermore, by analyzing the distance sequence of the intersection of the weld point and the isotherms in different directions, the uniformity of the isotherm intervals is evaluated and the coefficient of variation is calculated. At the same time, the mean similarity of the distance sequence in adjacent directions is calculated to construct the temperature gradient anomaly value. Finally, the product of the curvature anomaly value and the comprehensive anomaly value is mapped to an anomaly degree value through an activation function. This is then compared with a preset threshold to determine whether the welding area is abnormal. This method can comprehensively and accurately detect temperature distribution anomalies during welding, provide timely feedback and adjust welding operations, thereby effectively improving welding quality, reducing welding defects, and enhancing production efficiency and product quality.

[0008] Preferably, the extraction isotherm set includes:

[0009] Correct the temperature range in the temperature heat map and use a linear transformation to map the temperature values ​​to The pixel value range is set, and the isotherms are extracted at each preset pixel value interval. The pixels with temperature values ​​within a specific range in the temperature heat map are set to 255, and the other pixels are set to 0. The K3M algorithm is used to extract the skeleton of the pixel area with a value of 255, and it is mapped back to the temperature heat map to obtain the isotherm set.

[0010] The effect is that by correcting and processing the temperature heat map, the temperature value is linearly mapped to The system uses a pixel value range and extracts isotherms at preset pixel value intervals, effectively standardizing temperature data and highlighting key features of the temperature distribution. Binarization and the K3M algorithm extract the isotherm skeleton, further simplifying the isotherm shape and providing a clear isotherm set for subsequent analysis. This accurately reflects the temperature distribution in the weld area, improving the accuracy of anomaly detection and providing a key basis for welding process monitoring and quality control, thereby optimizing welding parameters, reducing welding defects, and improving welding quality and production efficiency.

[0011] Preferably, the temperature anomaly value of the isotherm includes:

[0012] Taking any isotherm in the isotherm set as the marking line, calculate the sum of the squares of the differences between the curvature of each pixel on the marking line and the mean curvature of all pixels, and take the square root of the mean sum of the squares of the differences as the temperature anomaly value of the marking line.

[0013] The effect is that by calculating the sum of the squared differences between the curvature of each pixel on any isotherm and the mean curvature of all pixels in the set, and taking the square root of this mean as the temperature anomaly value for the marker line, the degree of curvature change on the isotherm can be effectively quantified. This facilitates accurate assessment of the uniformity of the temperature distribution along the isotherm; larger temperature anomaly values ​​indicate a more uneven temperature distribution. This allows for more accurate detection of temperature anomalies during welding, enabling timely adjustment of welding parameters and optimization of temperature distribution, thereby reducing welding defects and improving welding quality and production efficiency.

[0014] Preferably, the temperature anomaly value of the isotherm further includes:

[0015] Taking any isotherm in the isotherm set as the marking line, calculate the ratio of the curvature of each pixel on the marking line to the mean curvature of all pixels, sum the absolute values ​​of the differences between the ratio and 1, and divide the sum by the number of pixels to obtain the temperature anomaly value of the marking line.

[0016] Preferably, obtaining the comprehensive abnormal value includes:

[0017] The position coordinates of the solder joint are determined, and the shortest distance from all pixel points on the marking line in the isothermal line set to the solder joint is calculated. The shortest distance is used as the weight of the temperature anomaly value of the marking line for weighted summation to obtain the comprehensive anomaly value of the soldering area corresponding to the isothermal line set.

[0018] The effect is that by determining the coordinates of the solder joint and calculating the shortest distance from all pixels on the marking line in the isothermal set to the solder joint, these distances are used as the weights of the marking line temperature anomaly values ​​for weighted summation to obtain the comprehensive anomaly value of the corresponding weld area in the isothermal set. This can more accurately reflect the temperature distribution uniformity of the weld area, especially in high-temperature areas near the solder joint, improve the accuracy of anomaly detection, and promptly identify problems such as local overheating or uneven cooling. Based on the feedback of the comprehensive anomaly value, it can optimize welding parameters, reduce welding defects, and improve welding quality and production efficiency.

[0019] Preferably, the intersection distance sequence is obtained by taking the welding point as the center, selecting the intersections of the ray in the preset direction and each isotherm, calculating the distance from each intersection to the welding point, and obtaining the distance sequence in the preset direction.

[0020] Its effect is that it is helpful to accurately reflect the relative position of the weld point and the isotherm in different directions, intuitively display the temperature distribution characteristics, detect abnormal changes in temperature gradients, improve the accuracy of abnormality detection, and provide data support for the optimization of welding parameters, thereby reducing welding defects and improving welding quality and production efficiency.

[0021] Preferably, the coefficient of variation of the intersection distance sequence includes:

[0022] The distance sequence in each direction is differentiated to obtain a differential sequence, the standard deviation and mean of the differential sequence are calculated, and the ratio between the standard deviation and the mean is taken as the coefficient of variation of the differential sequence.

[0023] Preferably, the temperature gradient abnormal value includes:

[0024] Taking the isotherm corresponding to any pixel value as the marker line and any direction as the target direction, the average value of the sum of the ratios of the curvature of all marker lines in the isotherm set in the target direction to the average curvature is calculated to obtain the average rate of change of the marker line. The mean function is used to calculate the absolute value of the difference between 1 and the average rate of change multiplied by the coefficient of variation of the isotherm interval sequence in the target direction to obtain the degree of temperature anomaly in the target direction of the welding area.

[0025] A negative exponential function is used to map the mean similarity between the isothermal line distance sequences in the target direction centered on the solder point to obtain the mean similarity. The product of the temperature anomaly degree and the mean similarity is taken as the total temperature gradient anomaly value of the soldering area.

[0026] Preferably, the step of determining whether there is an abnormality in the welding area based on the abnormality degree value and a preset abnormality threshold, and adjusting the welding operation, includes:

[0027] In response to the abnormality level value being less than the preset abnormality threshold, the welding area is normal; otherwise, if it is greater than or equal to the preset abnormality threshold, the welding area is abnormal, and the welding parameters or welding path are adjusted, where the welding parameters include: welding current, welding voltage and welding speed.

[0028] In a second aspect, an automated production line control system includes: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the automated production line control method described above is implemented.

[0029] The present invention has the following effects:

[0030] 1. By comprehensively analyzing the curvature changes of isotherms, the uniformity of isothermal intervals, and the differences in temperature distribution in different directions, the present invention can more comprehensively and accurately detect temperature anomalies in the welding area. This not only makes up for the shortcomings of relying solely on curvature consistency detection, but also can effectively identify local anomalies in areas with gentle temperature gradient changes, thereby reducing welding defects and improving welding quality and accuracy.

[0031] 2. The present invention combines curvature anomalies, comprehensive anomalies, and temperature gradient anomalies, activates function mapping, and compares with preset thresholds to quickly and accurately determine whether there are anomalies in the welding area. In response to detected anomalies, the system can provide timely feedback and adjust welding parameters (such as current, voltage, speed) or optimize the welding path, thereby achieving real-time monitoring and dynamic adjustment of the welding process, improving production efficiency and product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 This is a method flow chart of steps S1 to S4 in an automated production line control method according to an embodiment of the present invention.

[0033] Figure 2 This is a structural block diagram of an automated production line control system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0034] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0035] Reference Figure 1 , an automated production line control method includes steps S1 to S4, specifically as follows:

[0036] It should be noted that the specific analysis scenario below is: the weld point in each welding image is used as the analysis object. During the welding process, ideally, heat radiation is evenly distributed within the weld area centered on the weld point. Therefore, all isotherms are concentric rings centered on the weld point. Therefore, the curvature of the isotherms can be used to determine whether anomalies have occurred. In abnormal situations, the temperature distribution in the area around the weld is uneven, resulting in significant differences in the curvature of the isotherms. The consistency of the isotherm curvature is then established to determine anomalies.

[0037] S1: Obtain the temperature distribution of the welding area to generate a temperature heat map, preprocess the temperature heat map, and extract a set of isotherms from the preprocessed temperature heat map.

[0038] For example, the infrared thermal imager is positioned to clearly observe the weld, ensuring that its field of view covers the entire weld area and that its line of sight is perpendicular to the weld surface. This ensures that the collected temperature data is accurate and complete. The thermal imager collects infrared radiation signals from the weld area in real time and converts them into temperature data, generating a temperature heat map. Each pixel in the heat map corresponds to a temperature value, reflecting the real-time temperature distribution of the weld area.

[0039] Correct the temperature range in the temperature heat map and use a linear transformation to map the temperature values ​​to The pixel value range is set, and isotherms are extracted at preset pixel value intervals. The pixels with temperature values ​​within a specific range in the temperature heat map are set to 255, and the other pixels are set to 0. The K3M algorithm is used to extract the skeleton of the pixel area with a value of 255, and it is mapped back to the original temperature heat map to obtain a set of isotherms.

[0040] For example, the lowest temperature point corresponds to pixel value 0, the highest temperature point corresponds to pixel value 255, and the intermediate temperature values ​​are mapped proportionally to the pixel value 0. The preset pixel value is 5 pixel values, which can be adjusted according to the specific situation. For example, the isotherms corresponding to pixel values ​​of 5, 10, 15, ..., 250, and 255 are selected. For each isotherm, the point in the heat map with a pixel value equal to the current isotherm is set to 255, and the rest are set to 0. For example, when extracting the isotherm with a pixel value of 250, the point in the heat map with a pixel value of 250 is set to 255, and the rest are set to 0 for binarization.

[0041] In other words, if the isotherms are evenly spaced and continuous, the temperature distribution in the weld area is relatively uniform. If the isotherms are broken, clustered, or discontinuous, it indicates possible welding defects (such as localized overheating or rapid cooling). The density of the isotherms is used to analyze the temperature gradient. Areas with dense isotherms indicate rapid temperature changes, and welding parameters may need to be adjusted to optimize the temperature distribution. The isotherm analysis results are fed back to the welding control system, allowing real-time adjustments to the welding current, voltage, or speed to reduce weld distortion and improve weld quality.

[0042] S2: Determine the temperature anomaly value of the isotherm according to the curvature change of the pixel points on each isotherm in the isotherm set, use the distance between the solder point and the isotherm as the weight of the temperature anomaly value, and perform weighted summation to obtain a comprehensive anomaly value.

[0043] Temperature anomalies of the isotherm, including:

[0044] Taking any isotherm in the isotherm set as the marking line, calculate the sum of the squares of the differences between the curvature of each pixel on the marking line and the mean curvature of all pixels, and take the square root of the mean sum of the squares of the differences as the temperature anomaly value of the marking line.

[0045] Specifically, the temperature anomaly value satisfies the following relationship:

[0046] ;

[0047] Where, Indicates that the pixel value is Isotherms The temperature anomaly value, Indicates that the pixel value is Isotherms Previous The curvature of a pixel, Indicates that the pixel value is Isotherms The mean curvature of all pixels on Indicates that the pixel value is Isotherms The number of pixels on the image.

[0048] That is to say, Isotherm The standard deviation of the curvature of all pixels on the The measure of curvature consistency is A smaller value indicates The curvature consistency of the upper point is high, and the possibility of abnormality is small; on the contrary, if A larger value indicates that the isotherm The curvature consistency of the upper point is poor, that is, the isotherm The curvature of each point is quite different, indicating that the temperature around the solder joint is not uniform and the possibility of abnormality is high.

[0049] In addition, another embodiment further includes:

[0050] Taking any isotherm in the isotherm set as the marking line, calculate the ratio of the curvature of each pixel on the marking line to the mean curvature of all pixels, sum the absolute values ​​of the differences between the ratio and 1, and divide the sum by the number of pixels to obtain the temperature anomaly value of the marking line.

[0051] Specifically, the temperature anomaly value satisfies the following relationship:

[0052] ;

[0053] Where, Indicates that the pixel value is Isotherms The temperature anomaly value, Indicates that the pixel value is Isotherms Previous The curvature of a pixel, Indicates that the pixel value is Isotherms The mean curvature of all pixels on Indicates that the pixel value is Isotherms The number of pixels on the image.

[0054] That is to say, It reflects the change of the curvature of a point on the isotherm relative to the average curvature of the isotherm. A ratio close to 1 indicates that the curvature of the point is close to the average curvature and the temperature distribution is relatively uniform. It reflects the deviation of the curvature ratio from 1 and is used to measure the uniformity of temperature distribution at that point.

[0055] Get comprehensive outlier values, including:

[0056] The position coordinates of the solder joint are determined, and the shortest distance from all pixels on the marking line in the isothermal line set to the solder joint is calculated. The shortest distance is used as the weight of the temperature anomaly value of the marking line for weighted summation to obtain the comprehensive anomaly value of the corresponding soldering area of ​​the isothermal line set.

[0057] Specifically, the comprehensive outlier value satisfies the following relationship:

[0058] ;

[0059] Where, Indicates the comprehensive abnormal value of the welding area corresponding to the isothermal line set, Indicates that the pixel value is Isotherms The temperature anomaly value, Indicates that the pixel value is Isotherms The shortest distance to the soldering point, Represented by natural numbers An exponential function with base .

[0060] That is to say, the larger the temperature anomaly value, the The greater the curvature difference of the pixel points on the isothermal line, the more uneven the temperature distribution in the area is, and the greater the possibility of abnormal deformation; the smaller the temperature anomaly value is, the greater the possibility of abnormal deformation. The higher the curvature consistency of the pixels on an isotherm, the smaller the possibility of anomalies.

[0061] The smaller the shortest distance, that is, the closer the isotherm is to the weld point, the more dramatic the temperature change of the isotherm, the greater the impact of the high welding temperature, and the greater the contribution to the entire area. If the curvature consistency of the isotherm is low, the area around the weld is more likely to have abnormal deformation. Therefore, by using distance weighting to give a higher weight to the area near the weld point, the sensitivity of anomaly detection is improved.

[0062] It should also be noted that the consistency of the curvature of each pixel point on the isotherm is used to reflect whether the temperature distribution is uniform, and the isotherm is weighted by the shortest distance between the weld point to obtain the outlier value of the entire weld area. However, this cannot accurately detect welding anomalies. Inconsistent curvature (large curvature changes) indicates that there are local fluctuations in the temperature distribution. It is also necessary to consider whether the intervals between the isotherms are uniform. The intervals between the isotherms can reflect the temperature gradient and supplement the gentle anomalies that the curvature cannot capture. The specific steps are as follows:

[0063] S3: Obtain the intersection distance sequence between the solder point and the isothermal line in different directions. Based on the intersection distance sequence, analyze the uniformity of the isothermal line interval and obtain the coefficient of variation of the intersection distance sequence. Calculate the mean similarity between the distance sequences in all adjacent directions and construct the temperature gradient anomaly value based on the coefficient of variation and the mean similarity.

[0064] The method for obtaining the intersection distance sequence is as follows: with the weld point as the center, select the intersection points of the ray in the preset direction and each isotherm, calculate the distance from each intersection point to the weld point, and obtain the distance sequence in the preset direction.

[0065] The coefficient of variation of the intersection distance sequence, including:

[0066] The distance sequence in each direction is differentiated to obtain a differential sequence, the standard deviation and mean of the differential sequence are calculated, and the ratio between the standard deviation and the mean is taken as the coefficient of variation of the differential sequence.

[0067] It should be noted that the larger the coefficient of variation, the more uneven the isotherms are in that direction, that is, the temperature gradient is uneven, and abnormal conditions may occur in that direction.

[0068] Temperature gradient anomalies, including:

[0069] Taking the isotherm corresponding to any pixel value as the marker line and any direction as the target direction, the average value of the sum of the ratios of the curvature of all marker lines in the isotherm set in the target direction to the average curvature is calculated to obtain the average rate of change of the marker line. The mean function is used to calculate the absolute value of the difference between 1 and the average rate of change multiplied by the coefficient of variation of the isotherm interval sequence in the target direction to obtain the degree of temperature anomaly in the target direction of the welding area.

[0070] A negative exponential function is used to map the mean similarity between the isothermal line distance sequences in the target direction centered on the solder point to obtain the mean similarity. The product of the temperature anomaly degree and the mean similarity is taken as the total temperature gradient anomaly value of the soldering area.

[0071] Specifically, the temperature gradient anomaly satisfies the following relationship:

[0072] ;

[0073] Where, Indicates the abnormal value of the comprehensive temperature gradient in the welding area, Indicates the The coefficient of variation of the isotherm interval sequence in each direction, Indicates that the pixel value is Isotherms No. The curvature in the direction, Indicates that the pixel value is Isotherms The mean curvature of all pixels on represents the total number of isotherms, Indicates the first The mean of the similarity between the isotherm distance sequences in each direction, represents the mean function, Represented by natural numbers An exponential function with base .

[0074] That is to say, For solder joint Direction and The curvature of the intersection of the first isotherm is The ratio of the curvature of each direction to the mean curvature of all pixels on the isotherm reflects the deviation between the curvature of the intersection point and the average curvature of the isotherm. The closer the curvature ratio is to 1, the more uniform the temperature distribution is. Conversely, the more the curvature ratio deviates from 1, the more uneven the temperature distribution is and the possibility of abnormal conditions may occur.

[0075] Centered on the solder joint The mean of the similarity between the isothermal distance sequences in each direction reflects the similarity of the isothermal distance sequences in different directions. The higher the similarity, the more uniform the temperature distribution in the welding area centered on the weld point, and the lower the possibility of abnormal deformation. Conversely, the lower the similarity between the isothermal distance sequences, the more uneven the temperature distribution in the welding area, and the possibility of abnormality.

[0076] S4: The product of the curvature anomaly value and the comprehensive anomaly value is used for activation function to obtain the abnormality degree value of the welding area. According to the abnormality degree value and the preset abnormality threshold, it is judged whether there is an abnormality in the welding area, and the welding operation is adjusted.

[0077] Specifically, the abnormality degree value satisfies the following relationship:

[0078] ;

[0079] Where, Indicates the abnormality value. represents the activation function, Indicates the comprehensive abnormal value of the welding area corresponding to the isothermal line set, Indicates the abnormal value of the comprehensive temperature gradient in the welding area.

[0080] The abnormality degree value is mapped to the range of 0-1 using the sigmoid function, and an abnormality threshold is preset, the preset abnormality threshold is 0.8, and in response to the abnormality degree value being less than the preset abnormality threshold, the welding area is normal, otherwise, if it is greater than or equal to the preset abnormality threshold, the welding area is abnormal and needs to be fed back to the staff to adjust the welding operation, for example, such as adjusting the welding parameters or adjusting the welding path, where the welding parameters include but are not limited to: welding current, welding voltage and welding speed.

[0081] In other words, if the abnormal information indicates that the local temperature is too high (such as a high curvature abnormal value), the welding current may be too high. Appropriately reduce the welding current to reduce heat input and avoid local overheating. If the temperature gradient is abnormally high, it may be due to unstable welding voltage, resulting in arc instability. Adjust the welding voltage to ensure arc stability and more even heat distribution. If the temperature distribution is uneven, it may be due to excessive or slow welding speed. Adjust the welding speed appropriately to ensure even heat distribution. Excessive speed may cause localized cooling, while excessive speed may cause localized overheating.

[0082] The present invention also provides an automated production line control system. Figure 2 As shown, the system includes a processor and a memory. The memory stores computer program instructions. When the computer program instructions are executed by the processor, an automated production line control method according to the first aspect of the present invention is implemented. The system also includes other components familiar to those skilled in the art, such as a communication bus and a communication interface. The configuration and functions of these components are well known in the art and are therefore not described in detail here.

[0083] It should be noted that those skilled in the art may make various modifications and improvements without departing from the scope of the present invention, and these modifications and improvements fall within the scope of protection of the present invention. Therefore, the scope of protection of the patent for this invention shall be based on the appended claims.

Claims

1. A method for controlling an automated production line, characterized in that: include: Obtain the temperature distribution of the welding area to generate a temperature heat map, pre-process the temperature heat map, and extract the isotherm set from the pre-processed temperature heat map, including: correcting the temperature range in the temperature heat map, using linear transformation to map the temperature value to The pixel value range is set, and isotherms are extracted at preset pixel value intervals. Pixels with temperature values ​​within a specific range in the temperature heat map are set to 255, and other pixels are set to 0. The K3M algorithm is used to extract the skeleton of the pixel area with a value of 255, and mapped back to the temperature heat map to obtain a set of isotherms. If the isotherms are evenly spaced and continuous, it means that the temperature distribution in the welding area is relatively uniform. If the isotherms are broken, aggregated, or discontinuous, it indicates that there may be welding defects. According to the curvature change of the pixel points on each isothermal line in the isothermal line set, the temperature anomaly value of the isothermal line is determined, and the distance between the solder point and the isothermal line is used as the weight of the temperature anomaly value, including: Determine the position coordinates of the solder joint, calculate the shortest distance from all pixels on the marking line in the isothermal line set to the solder joint, and use the shortest distance as the weight of the temperature anomaly value of the marking line for weighted summation to obtain the comprehensive anomaly value; Obtain the intersection distance sequence between the solder joint and the isothermal line in different directions. Based on the intersection distance sequence, analyze the uniformity of the isothermal line intervals and obtain the coefficient of variation of the intersection distance sequence. Calculate the mean similarity between the distance sequences in all adjacent directions and construct the temperature gradient anomaly value based on the coefficient of variation and the mean similarity. The product of the curvature anomaly value and the comprehensive anomaly value is used with an activation function to obtain the abnormality degree value of the welding area. According to the abnormality degree value and the preset abnormality threshold, it is judged whether there is an abnormality in the welding area and the welding operation is adjusted.

2. The automated production line control method according to claim 1, characterized in that: The temperature anomaly values ​​of the isotherms include: Taking any isotherm in the isotherm set as the marking line, calculate the sum of the squares of the differences between the curvature of each pixel on the marking line and the mean curvature of all pixels, and take the square root of the mean sum of the squares of the differences as the temperature anomaly value of the marking line.

3. The automated production line control method according to claim 1, characterized in that: The temperature anomaly value of the isotherm also includes: Taking any isotherm in the isotherm set as the marking line, calculate the ratio of the curvature of each pixel on the marking line to the mean curvature of all pixels, sum the absolute values ​​of the differences between the ratio and 1, and divide the sum by the number of pixels to obtain the temperature anomaly value of the marking line.

4. The automated production line control method according to claim 1, characterized in that: The intersection distance sequence is obtained by taking the welding point as the center, selecting the intersection points of the ray in the preset direction and each isotherm, calculating the distance from each intersection point to the welding point, and obtaining the distance sequence in the preset direction.

5. The automated production line control method according to claim 1, characterized in that: The coefficient of variation of the intersection distance sequence includes: The distance sequence in each direction is differentiated to obtain a differential sequence, the standard deviation and mean of the differential sequence are calculated, and the ratio between the standard deviation and the mean is taken as the coefficient of variation of the differential sequence.

6. The automated production line control method according to claim 1, characterized in that: The temperature gradient abnormal value includes: Taking the isotherm corresponding to any pixel value as the marker line and any direction as the target direction, the average value of the sum of the ratios of the curvature of all marker lines in the isotherm set in the target direction to the average curvature is calculated to obtain the average rate of change of the marker line. The mean function is used to calculate the absolute value of the difference between 1 and the average rate of change multiplied by the coefficient of variation of the isotherm interval sequence in the target direction to obtain the degree of temperature anomaly in the target direction of the welding area. A negative exponential function is used to map the mean similarity between the isothermal line distance sequences in the target direction centered on the solder point to obtain the mean similarity. The product of the temperature anomaly degree and the mean similarity is taken as the total temperature gradient anomaly value of the soldering area.

7. The automated production line control method according to claim 1, characterized in that: The method of determining whether there is an abnormality in the welding area according to the abnormality degree value and the preset abnormality threshold value and adjusting the welding operation includes: In response to the abnormality level value being less than the preset abnormality threshold, the welding area is normal; otherwise, if it is greater than or equal to the preset abnormality threshold, the welding area is abnormal, and the welding parameters or welding path are adjusted, where the welding parameters include: welding current, welding voltage and welding speed.

8. An automated production line control system, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the automated production line control method according to any one of claims 1 to 7 is implemented.

Images