A method, system and apparatus for ring seam welding of bicycle components

Abstract

The present application relates to the technical field of ring seam welding, and particularly relates to a bicycle part ring seam welding method, system and device. The method comprises: obtaining a thermal imaging image of a bicycle part in a welding process, and performing gray scale processing to obtain a gray scale image; determining a welding pool area, a heat affected zone and a spatter hot spot according to the gray scale distribution characteristics of the gray scale image; radiating outward from the center point of the welding pool area to determine the gradient feature sequence of the temperature variation of the heat affected zone in different edge pixel directions; determining the workpiece pipe diameter deviation degree according to the length difference and the gray scale distribution of the gradient feature sequence; analyzing the gray scale change characteristics of the welding pool area in the current frame and the previous frame gray scale image, and combining the area and distribution characteristics of the spatter hot spot in the current frame to evaluate the stability index; and adjusting the parameters of the current welding process in real time according to the stability index and the workpiece pipe diameter deviation degree. The present application can effectively realize real-time regulation and control of the welding process, and improve the welding stability.

Description

Technical Field

[0001] This invention relates to the field of circumferential welding technology, and specifically to a method, system, and apparatus for circumferential welding of bicycle parts. Background Technology

[0002] As bicycles demand higher levels of aesthetics and mechanical strength, the requirements for welding quality in bicycle component manufacturing are also increasing. To ensure welding quality, welding parameters need to be adjusted in real time based on the data changes in the circumferential seam area of ​​the components during the welding process, thereby improving the welding quality of bicycle components.

[0003] In related technologies, some automatic welding systems integrate process parameter libraries and automatically retrieve preset parameters for PID control based on the input material type and thickness. However, most current welding equipment does not have the ability to sense the real-time status of parts such as welding process temperature, molten pool state, and deformation trend. Once the process parameters are set, they are fixed values ​​and cannot be dynamically adjusted according to material changes or welding fluctuations. The lack of automatic adjustment capability in the welding process leads to uneven welding heat, affecting the stability of the overall welding control. Summary of the Invention

[0004] To address the technical problem in related technologies where fixed values ​​cannot achieve dynamic self-adjustment in response to material changes or welding fluctuations, leading to uneven welding heat and affecting the overall stability of welding control, this invention provides a method, system, and apparatus for circumferential seam welding of bicycle components. The specific technical solution adopted is as follows:

[0005] This invention proposes a method for circumferential welding of bicycle parts, the method comprising:

[0006] Thermal imaging images of bicycle parts during the welding process are acquired and then converted to grayscale to obtain grayscale images.

[0007] Based on the grayscale distribution characteristics of different regions in the grayscale image, the weld pool area, heat-affected zone, and spatter hotspots are determined;

[0008] Radience outward from the center point of the weld pool region determines the gradient feature sequence of temperature changes in the heat-affected zone at different edge pixel points; based on the length difference and grayscale distribution of the gradient feature sequence, the workpiece pipe diameter deviation of the current welding area is determined.

[0009] Analyze the grayscale change characteristics of the weld pool area in the grayscale images of the current frame and the previous frame, and combine the area and distribution characteristics of the spatter hotspots in the current frame to evaluate the stability index of the current welding process.

[0010] Based on the stability index and the deviation of the workpiece diameter, the parameters of the current welding process are adjusted in real time.

[0011] Furthermore, determining the weld pool area, heat-affected zone, and spatter hotspot based on the grayscale characteristics of different regions in the grayscale image includes:

[0012] The density clustering algorithm is used to perform gray-level clustering on the pixels in the welding influence area to determine different clustering regions. The mean gray value of the pixels in each clustering region is calculated as the gray-level index of the corresponding clustering region.

[0013] Clustered regions whose grayscale index is greater than a preset first index threshold are combined as welding pool regions.

[0014] In clustered regions other than the weld pool area, the mean difference of gray index between each clustered region and all other adjacent clustered regions is calculated. Clustered regions with a mean difference of gray index greater than a preset difference threshold and a larger gray index value compared with all other adjacent clustered regions are designated as spatter hotspots.

[0015] The area excluding the weld pool area and spatter hotspots is designated as the heat-affected zone.

[0016] Furthermore, the step of determining the gradient feature sequence of temperature changes in the heat-affected zone at different edge pixel points by radiating outwards from the center point of the weld pool region includes:

[0017] Using the morphological center point of the weld pool region as the radiation point, ray connections are made to different edge pixel points of the heat-affected region.

[0018] The difference between the gray value of each pixel on the ray and the gray value of the corresponding previous pixel is used as the gray gradient of the pixel. The pixels are sorted in order from near to far from the radiation point to obtain the gradient feature sequence of temperature change in the direction of the corresponding edge pixels.

[0019] Further, determining the workpiece diameter deviation in the current welding area based on the length difference and grayscale distribution of the gradient feature sequence includes:

[0020] The range of the number of elements contained in different gradient feature sequences is used as the first deviation analysis index.

[0021] Calculate the standard deviation of all elements in each gradient feature sequence as an indicator of sequence volatility.

[0022] The standard deviation of all sequence volatility indicators is used as the second deviation analysis indicator;

[0023] The product of the first deviation analysis index and the second deviation analysis index is normalized and used as the workpiece pipe diameter deviation.

[0024] Furthermore, the analysis of the grayscale change characteristics of the weld pool region in the grayscale images of the current frame and the previous frame, combined with the area and distribution characteristics of the spatter hotspots in the current frame, to evaluate the stability index of the current welding process includes:

[0025] By drawing a perpendicular line along the welding direction through the morphological center point of the weld pool region, the weld pool region is divided into two influence regions.

[0026] Based on the grayscale changes of pixels in different affected areas along the welding direction, the grayscale change characteristic index of the weld pool area is determined.

[0027] Calculate the absolute value of the difference between the grayscale change feature index of the current frame and the previous frame, and normalize the negative of the absolute value of the difference as the first welding stability coefficient of the current frame.

[0028] The second welding stability coefficient is determined based on the area of ​​the spatter hotspot and the density distribution of the spatter hotspot itself.

[0029] The sum of the first welding stability coefficient and the second welding stability coefficient is normalized and used as the stability index of the current welding process.

[0030] Furthermore, the step of determining the grayscale change characteristic index of the weld pool region based on the grayscale changes of pixels in different affected regions along the welding direction includes:

[0031] Calculate the absolute value of the grayscale value difference between each pixel and the next pixel in the welding direction to obtain the grayscale change value of the corresponding pixel.

[0032] The average grayscale change value of all pixels in each affected region is taken as the region change value;

[0033] The absolute value of the difference between the regional change values ​​of the two affected areas is used as the gray-scale change characteristic index of the weld pool area.

[0034] Furthermore, determining the second welding stability coefficient based on the area of ​​the spatter hotspot and the density distribution of the spatter hotspot itself includes:

[0035] Within a predetermined range centered on the morphological center point of the weld pool region, the number of spatter hotspots contained therein is determined as a density analysis index.

[0036] The product of the total area of ​​all the spatter hotspots and the density analysis index is normalized to obtain the second welding stability coefficient.

[0037] Furthermore, the real-time adjustment of parameters for the current welding process based on the stability index and workpiece diameter deviation includes:

[0038] Calculate the ratio of workpiece diameter deviation to stability index, and linearly map it to the range (0,2) as the adjustment index for the current moment;

[0039] The proportional gain of the PID controller is adjusted according to the aforementioned adjustment index to perform PID control on the welding process.

[0040] On the other hand, a circumferential weld system for bicycle components is also provided, the system comprising:

[0041] The acquisition module is used to acquire thermal imaging images of bicycle parts during the welding process and perform grayscale processing to obtain grayscale images;

[0042] The segmentation module is used to determine the weld pool area, heat-affected zone, and spatter hotspots based on the grayscale distribution characteristics of different regions in the grayscale image.

[0043] The deviation analysis module is used to determine the gradient feature sequence of temperature change in the heat-affected zone in different edge pixel directions, radiating outward from the center point of the weld pool area; and to determine the workpiece pipe diameter deviation in the current welding area based on the length difference and grayscale distribution of the gradient feature sequence.

[0044] The stability analysis module is used to analyze the grayscale change characteristics of the weld pool area in the grayscale image of the current frame and the previous frame, and to evaluate the stability index of the current welding process by combining the area and distribution characteristics of the spatter hotspots in the current frame.

[0045] The control module is used to adjust the parameters of the current welding process in real time based on the stability index and the deviation of the workpiece diameter.

[0046] On the other hand, a circumferential welding apparatus for bicycle parts is also provided. The apparatus includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of the method as described in any of the foregoing claims.

[0047] The present invention has the following beneficial effects:

[0048] This invention acquires thermal imaging images of bicycle components during the welding process, performs grayscale processing to obtain grayscale images, and divides the weld pool area, heat-affected zone, and spatter hotspots based on image features. Since welding analysis must consider the overall welding effect, this division allows for separate analysis of different areas, improving the reliability of the overall welding analysis. Based on the heat diffusion effect of the heat-affected zone in different directions, the workpiece diameter deviation of the welding area is analyzed. This deviation effectively characterizes the heat diffusion deviation effect of the heat-affected zone due to the pipe diameter characteristics during welding. Then, combining the grayscale variation characteristics of the weld pool area and the area and distribution characteristics of the spatter hotspots, a stability index for the current welding process is determined. This stability index combines the characteristics of both the weld pool area and the spatter hotspots to analyze the stability of the welding process. Finally, based on the stability index and the workpiece diameter deviation, the parameters of the current welding process are adjusted in real time. This invention enables independent analysis of different welding areas, combining the characteristics of all welding areas to achieve real-time control analysis, reducing uneven welding heat and improving the overall stability of welding control. Attached Figure Description

[0049] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 This is a flowchart of a circumferential weld method for bicycle parts provided in one embodiment of the present invention;

[0051] Figure 2 This is a schematic diagram of a grayscale image provided for an embodiment of the present invention. Detailed Implementation

[0052] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a circumferential weld method, system, and apparatus for bicycle parts according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0053] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0054] It should be noted that, for ease of calculation, all indicator data involved in the calculation in this embodiment of the invention have undergone data preprocessing to eliminate the influence of dimensions. The specific methods for eliminating the influence of dimensions are well known to those skilled in the art and are not limited here.

[0055] The following describes in detail, with reference to the accompanying drawings, a specific scheme for a circumferential weld method for bicycle parts provided by the present invention.

[0056] Please see Figure 1 The diagram illustrates a flowchart of a circumferential weld method for bicycle components according to an embodiment of the present invention. The method includes:

[0057] S101: Acquire thermal imaging images of bicycle parts during the welding process, and convert them to grayscale to obtain grayscale images.

[0058] In this embodiment of the invention, a far-infrared thermal imager can be used as the infrared thermal imaging sensor. The infrared thermal imaging sensor is installed above or to the side of the welding torch, looking down at the welding area. The infrared thermal imaging sensor is equipped with an air curtain (nitrogen / compressed air, pressure 0.5MPa) and a sapphire protective window (temperature resistance >1500℃) to prevent dust from adhering. The sampling frequency is set to 30Hz. During installation, it is necessary to ensure that the infrared thermal imaging sensor and the vision sensor maintain a safe distance from the welding trajectory to avoid collision.

[0059] In a specific welding scenario, bicycle parts (such as bottom bracket housings, rear fork connecting tubes, etc.) are fixed on a rotating fixture. After clamping, the control system records the welding path parameters, and the welding gun robotic arm enters the welding position. The material, thickness, and other parameters of the bicycle parts are input into the welding control system of the welding machine. The welding control system extracts the corresponding process recommendation values ​​from the built-in welding process database to obtain the initial welding parameters. The automatic rotation module is activated, and the workpiece rotates uniformly at a set speed. The temperature distribution of the bicycle parts during the circumferential seam welding process is collected by an infrared thermal imaging sensor, generating a thermal imaging image.

[0060] After acquiring the original thermal imaging image, this embodiment of the invention can also perform image preprocessing. The Otsu thresholding method is used to perform binary segmentation on the thermal image to obtain a binary segmentation image, and the regions with thermal characteristics are extracted to obtain the thermal imaging image of the bicycle parts in this embodiment of the invention. That is, the thermal imaging image only contains the regions affected by the welding temperature, and the background is eliminated. Then, a filtering algorithm is used to denoise the thermal image, wherein nonlocal mean filtering is selected to denoise the thermal image.

[0061] It should be noted that, in this embodiment of the invention, in order to facilitate image analysis, the thermal imaging image also needs to be grayscale processed to obtain a grayscale image, wherein the grayscale processing can specifically be mean grayscale processing.

[0062] The grayscale image contains information about the temperature distribution of the target surface; the higher the grayscale value of a pixel, the higher the temperature of the target object. (See also...) Figure 2 , Figure 2 This is a schematic diagram of a grayscale image provided for an embodiment of the present invention.

[0063] S102: Based on the grayscale distribution characteristics of different regions in the grayscale image, determine the weld pool area, heat-affected zone, and spatter hotspot.

[0064] During the welding process, the area can be divided into the weld pool region, the heat-affected zone (HAZ), and the spatter hotspot region based on differences in temperature and shape. The weld pool region is the core high-temperature area currently being targeted by the welding torch, representing the highest temperature. The HAZ exhibits a gradient distribution of temperature that gradually decreases around the weld pool region, and has a relatively large area. The spatter hotspot region originates from the weld pool region and has a relatively high temperature. The different regions can be defined by combining their characteristics.

[0065] Furthermore, in some embodiments of the present invention, determining the weld pool region, heat-affected zone, and spatter hotspot based on the grayscale characteristics of different regions in the grayscale image includes: performing grayscale clustering processing on the pixels within the weld-affected zone based on a density clustering algorithm to determine different cluster regions; calculating the average grayscale value of the pixels in each cluster region as the grayscale index of the corresponding cluster region; grouping cluster regions with grayscale indices greater than a preset first index threshold as the weld pool region; calculating the average difference in grayscale indices between each cluster region and all adjacent cluster regions in other cluster regions besides the weld pool region; and designating cluster regions with average grayscale indices greater than a preset difference threshold and larger grayscale indices compared to all adjacent cluster regions as spatter hotspots; and designating the region other than the weld pool region and spatter hotspot as the heat-affected zone.

[0066] In this embodiment of the invention, the density clustering algorithm can be specifically the k-means clustering algorithm, where the k value is obtained by the elbow method. Of course, it can also be based on other unsupervised density clustering algorithms to form a clustering region with pixels that have similar gray values ​​and are adjacent to each other.

[0067] First, based on the characteristics of the weld pool region having the highest temperature and the largest gray value, the weld pool region is analyzed. Specifically, clusters of regions with gray values ​​greater than a preset first threshold are defined as weld pool regions. This preset first threshold can be, for example, 200, meaning regions with gray values ​​above 200 are considered weld pool regions. Then, the heat-affected zone and spatter hotspots are further subdivided.

[0068] Spatter hotspots are caused by molten material splashing out from the weld pool area. Besides being a direct characteristic of the weld pool area, spatter hotspots are a typical sign of welding defects, especially in circumferential welding. The amount and distribution of spatter directly reflect welding stability, the rationality of heat input, and the stability of the weld pool. When welding parameters are too high, the more intense the metal vaporization, the worse the stability of the weld pool, and the easier it is to form a large number of high-temperature particles, i.e., spatter hotspots.

[0069] Therefore, its temperature is higher than that of the heat-affected zone. Since the heat-affected zone is affected by temperature and distance, its grayscale change is relatively linear. Therefore, the mean difference of grayscale index between each cluster and all other adjacent clusters is calculated. Clusters with a mean difference of grayscale index greater than a preset difference threshold and a larger grayscale index value compared to all other adjacent clusters are designated as spatter hotspots. The preset difference threshold represents the grayscale difference value between the spatter hotspot and the heat-affected zone. Optionally, the preset difference threshold can be, for example, 100. That is, when a sudden heat change occurs and there is a significant high grayscale characteristic compared to other surrounding clusters, it can be regarded as a spatter hotspot. The area other than the weld pool area and the spatter hotspot is designated as the heat-affected zone.

[0070] S103: Radige outward from the center point of the weld pool area to determine the gradient feature sequence of temperature change in the heat-affected zone at different edge pixel points; determine the workpiece diameter deviation of the current welding area based on the length difference and grayscale distribution of the gradient feature sequence.

[0071] To ensure the aesthetics of the weld and maintain the overall load-bearing strength of the bicycle, bicycle structures commonly contain numerous hollow components with non-uniform diameters. During the circumferential welding of these irregularly shaped or unequal-diameter hollow tubular components, differences in wall thickness, pipe diameter, and the structure of the adjacent area of ​​the weld lead to variations in heat conduction rates during the welding process. This can easily cause heat accumulation in certain locations, resulting in an abnormal increase in the temperature of the weld pool, thereby affecting the weld fusion quality and the consistency of the weld formation.

[0072] Therefore, during the welding process, it is necessary to assess the heat conduction at the current welding location in real time. In welding areas with uniform pipe diameter and wall thickness, the heat diffusion rate within the heat-affected zone after welding heat input is balanced. This is reflected in thermal imaging images as a relatively gentle temperature gradient along the same pixel distance within the heat-affected zone, with the temperature decreasing uniformly from the center of the molten pool outwards. However, when there are abrupt changes in pipe diameter, uneven wall thickness, or changes in joint structure in the welding area, the heat diffusion path changes, leading to abrupt changes in the temperature gradient within the local heat-affected zone. To analyze these abrupt temperature gradient changes, it is necessary to determine the gradient characteristic sequence.

[0073] Furthermore, in some embodiments of the present invention, the gradient feature sequence of temperature change in different edge pixel directions of the heat-affected zone is determined by radiating outward from the center point of the weld pool region. This includes: taking the morphological center point of the weld pool region as the radiation point and connecting rays to different edge pixel directions of the heat-affected zone; taking the difference between the gray value of each pixel on the ray and the gray value of the corresponding previous pixel as the gray gradient of the pixel, and sorting them in order from near to far from the radiation point to obtain the gradient feature sequence of temperature change in the corresponding edge pixel direction.

[0074] Starting from the morphological center point, rays are connected in the direction of different edge pixels to obtain corresponding rays. The gray value difference between the pixels through which the ray passes and the corresponding previous pixel is sorted according to the starting point of the ray to obtain the gradient feature sequence.

[0075] It should be noted that the gray-level gradient at the radiation point is its own gray-level value, thereby determining the gradient feature sequence of gradient changes in the corresponding direction.

[0076] Furthermore, in some embodiments of the present invention, determining the workpiece diameter deviation of the current welding area based on the length difference and grayscale distribution of the gradient feature sequences includes: using the range of the number of elements contained in different gradient feature sequences as a first deviation analysis index; calculating the standard deviation of all elements in each gradient feature sequence as a sequence fluctuation index; using the standard deviation of all sequence fluctuation indices as a second deviation analysis index; and normalizing the product of the first deviation analysis index and the second deviation analysis index as the workpiece diameter deviation.

[0077] Among them, the larger the first deviation analysis index, the greater the range of element quantity, the greater the difference in length of different gradient feature sequences, which means the greater the difference in the diffusion effect produced during welding in different directions, and the more uneven the heating.

[0078] First, the standard deviation of all elements in each gradient feature sequence is calculated, and then the standard deviation of the standard deviation of all gradient feature sequences is calculated to perform fluctuation analysis on all gradient feature sequences. The larger the value of the second deviation analysis index, the greater the difference in heat diffusion effect in all directions and the more uneven the heating.

[0079] Based on the above feature analysis, the product of the first deviation analysis index and the second deviation analysis index is normalized and used as the workpiece pipe diameter deviation. The larger the value, the more likely the current welding area is to experience uneven heating during the circumferential weld due to the diameter of the bicycle parts, and the greater the degree of adjustment required for the welding parameters.

[0080] S104: Analyze the grayscale change characteristics of the weld pool area in the grayscale images of the current frame and the previous frame, and combine the area and distribution characteristics of the spatter hotspots in the current frame to evaluate the stability index of the current welding process.

[0081] After identifying welding anomalies caused by abnormal pipe diameter in bicycle components, further evaluation of the welding quality under the current welding parameters and the influence of the current pipe diameter is necessary, requiring analysis of the weld pool's state. The weld pool is a critical location during the welding process of bicycle components, representing the hottest part of the workpiece surface welding area. By analyzing the grayscale changes in the weld pool area across different frames, the overall thermal stability of the welding process can be determined.

[0082] Bicycle components, influenced by their application location and structural design, may exhibit variations not only in tube diameter but also in localized wall thickness. This geometric inhomogeneity directly impacts localized thermal conductivity during welding, causing variations in heat diffusion rates at different locations. Consequently, it affects the heat distribution of the weld pool, resulting in localized heat accumulation. In infrared thermography, this difference in thermal conductivity typically manifests as an uneven temperature gradient at the weld pool boundary. Specifically, this is reflected in narrower heat diffusion ranges or faster temperature decay rates in certain directions, leading to an asymmetric and irregular heat distribution characteristic in the weld pool.

[0083] Besides the direct characteristics of the molten pool area, spatter hotspots are one of the typical welding defects in the welding process, especially in circumferential welding. The amount and distribution of spatter directly reflect welding stability, the rationality of heat input, and the stability of the molten pool. When welding parameters are too high, the more intense the metal vaporization, the worse the molten pool stability characteristics, and the easier it is to form a large number of high-temperature particles. The initial molten pool stability index mainly considers the influence of molten pool stability on welding parameters from the perspective of heat loss, and it is also necessary to further adjust the molten pool stability characteristics from the perspective of excessive heat. In thermal imaging images, the number of spatter hotspots can directly reflect the intensity of heat input in the welding process. The more spatter hotspots there are and the larger their area, the more metal melts and spatters into the molten pool during the welding process, and the more unstable the welding process is.

[0084] Therefore, in this embodiment of the invention, the grayscale change characteristics of the weld pool region in the grayscale images of the current frame and the previous frame are analyzed, and the stability index of the current welding process is evaluated by combining the area and distribution characteristics of the spatter hotspots in the current frame. This includes: dividing the weld pool region into two influence regions by drawing a perpendicular line along the welding direction through the morphological center point of the weld pool region; determining the grayscale change characteristic index of the weld pool region based on the grayscale change of pixels in different influence regions along the welding direction; calculating the absolute value of the difference between the grayscale change characteristic index of the current frame and the previous frame, and normalizing the negative of the absolute value of the difference as the first welding stability coefficient of the current frame; determining the second welding stability coefficient based on the area of ​​the spatter hotspots and the density distribution of the spatter hotspots themselves; and normalizing the sum of the first welding stability coefficient and the second welding stability coefficient as the stability index of the current welding process.

[0085] It should be noted that the bicycle parts are continuously welded by rotation. This method causes morphological changes between consecutive frames due to rotation. These morphological changes mainly affect the heat diffusion effect, and during normal welding, a symmetrical effect is exhibited with the vertical line as the axis of symmetry. Therefore, in this embodiment of the invention, a vertical line is drawn along the welding direction through the morphological center point of the weld pool area, dividing the weld pool area into two influence areas. Based on the grayscale changes of pixels in different influence areas along the welding direction, the grayscale change characteristic index of the weld pool area is determined.

[0086] Furthermore, in some embodiments of the present invention, the grayscale change characteristic index of the weld pool region is determined based on the grayscale change of pixels in different influence regions in the welding direction, including: calculating the absolute value of the difference between the grayscale value of each pixel and the next pixel in the welding direction to obtain the grayscale change value of the corresponding pixel; taking the average grayscale change value of all pixels in each influence region as the region change value; and taking the absolute value of the difference between the region change values ​​of two influence regions as the grayscale change characteristic index of the weld pool region.

[0087] In this embodiment of the invention, by calculating the regional change value of the affected area, the absolute value of the difference between the regional change values ​​of the two affected areas is used as the gray-scale change characteristic index of the weld pool area. This index represents the symmetry effect of the weld pool area.

[0088] The absolute value of the difference between the grayscale change feature index of the current frame and the previous frame is used as the change value of the symmetry effect. The negative number of the absolute value of the difference is normalized and used as the first welding stability coefficient of the current frame. That is, the larger the first welding stability coefficient, the smaller the dynamic change of the welding pool area in the current frame, and the same symmetry effect is maintained, and the welding process is more stable.

[0089] Furthermore, in some embodiments of the present invention, determining a second welding stability coefficient based on the area of ​​the spatter hotspots and the density distribution of the spatter hotspots themselves includes: determining the number of spatter hotspots contained within a preset range centered on the morphological center point of the weld pool region as a density analysis index; and normalizing the negative of the product of the total area of ​​all spatter hotspots and the density analysis index as the second welding stability coefficient.

[0090] Since the number of hotspots directly reflects the intensity of heat input during welding, a greater number and larger area of ​​hotspots indicates that more metal melts and spatters into the weld pool during welding, making the welding process more unstable. Therefore, a preset range of 10cm in diameter can be defined to calculate density information, obtain density analysis index, and calculate the product of density analysis index and the total area of ​​all hotspots. The larger the product value, the higher the density and the larger the area of ​​the hotspots, indicating a more unstable welding process. The negative of the product value is normalized and used as the second welding stability coefficient.

[0091] Based on the above analysis, the sum of the first and second welding stability coefficients is normalized and used as the stability index of the current welding process. This stability index can analyze the stability of the welding process by combining the characteristics of the weld pool area and the spatter hotspots, and has high accuracy.

[0092] S105: Adjust the parameters of the current welding process in real time based on the stability index and the deviation of the workpiece diameter.

[0093] Based on the stability index and the workpiece diameter deviation, the parameters of the current welding process are adjusted in real time, including: calculating the ratio of the workpiece diameter deviation to the stability index, linearly mapping it to the range (0,2) as the adjustment index for the current moment; adjusting the proportional gain of the PID controller according to the adjustment index to perform PID control on the welding process.

[0094] In this embodiment of the invention, the workpiece diameter deviation represents the thermal diffusion deviation effect of the heat-affected zone due to the characteristics of the pipe diameter during the welding process, while the stability index represents the stability analysis of the welding process by combining the characteristics of the weld pool area and the spatter hotspots.

[0095] Therefore, the ratio of workpiece diameter deviation to stability index is directly mapped linearly to the range (0,2) as the adjustment index for the current moment. That is, the larger the value of the adjustment index, the greater the thermal diffusion deviation and unstable welding at the current moment, and the more adjustment effect needs to be increased. The linear mapping to the range (0,2) facilitates subsequent PID control.

[0096] In this embodiment of the invention, a PID controller can be used to adjust the welding torch current during the current welding process. The adjustment index is input into the PID controller as a proportional gain coefficient to adjust the welding torch current in real time. This reduces uneven welding heat caused by uneven pipe diameter during the circumferential welding of bicycle parts, and avoids common problems such as excessive / insufficient weld heat input, excessive spatter, and penetration deviation.

[0097] This invention acquires thermal imaging images of bicycle components during the welding process, performs grayscale processing to obtain grayscale images, and divides the weld pool area, heat-affected zone, and spatter hotspots based on image features. Since welding analysis must consider the overall welding effect, this division allows for separate analysis of different areas, improving the reliability of the overall welding analysis. Based on the heat diffusion effect of the heat-affected zone in different directions, the workpiece diameter deviation of the welding area is analyzed. This deviation effectively characterizes the heat diffusion deviation effect of the heat-affected zone due to the pipe diameter characteristics during welding. Then, combining the grayscale variation characteristics of the weld pool area and the area and distribution characteristics of the spatter hotspots, a stability index for the current welding process is determined. This stability index combines the characteristics of both the weld pool area and the spatter hotspots to analyze the stability of the welding process. Finally, based on the stability index and the workpiece diameter deviation, the parameters of the current welding process are adjusted in real time. This invention enables independent analysis of different welding areas, combining the characteristics of all welding areas to achieve real-time control analysis, reducing uneven welding heat and improving the overall stability of welding control.

[0098] On the other hand, the present invention also provides a circumferential weld system for bicycle parts, the system comprising:

[0099] The acquisition module is used to acquire thermal imaging images of bicycle parts during the welding process and perform grayscale processing to obtain grayscale images;

[0100] The segmentation module is used to determine the weld pool area, heat-affected zone, and spatter hotspots based on the grayscale distribution characteristics of different regions in the grayscale image.

[0101] The deviation analysis module is used to determine the gradient feature sequence of temperature change in the heat-affected zone in different edge pixel directions, radiating outward from the center point of the weld pool area; and to determine the workpiece pipe diameter deviation in the current welding area based on the length difference and grayscale distribution of the gradient feature sequence.

[0102] The stability analysis module is used to analyze the grayscale change characteristics of the weld pool area in the grayscale image of the current frame and the previous frame, and to evaluate the stability index of the current welding process by combining the area and distribution characteristics of the spatter hotspots in the current frame.

[0103] The control module is used to adjust the parameters of the current welding process in real time based on the stability index and the deviation of the workpiece diameter.

[0104] In an embodiment of the present invention, a bicycle component circumferential seam welding system implements the steps of any of the methods described above during the execution of various tasks.

[0105] On the other hand, the present invention also provides a circumferential welding device for bicycle parts, the device including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of any of the methods described above.

[0106] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0107] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for circumferential welding of bicycle parts, characterized in that, The method for circumferential welding of bicycle parts includes: Thermal imaging images of bicycle parts during the welding process are acquired and then converted to grayscale to obtain grayscale images. Based on the grayscale distribution characteristics of different regions in the grayscale image, the weld pool area, heat-affected zone, and spatter hotspots are determined; Radience outward from the center point of the weld pool region determines the gradient feature sequence of temperature changes in the heat-affected zone at different edge pixel points; based on the length difference and grayscale distribution of the gradient feature sequence, the workpiece pipe diameter deviation of the current welding area is determined. Analyze the grayscale change characteristics of the weld pool area in the grayscale images of the current frame and the previous frame, and combine the area and distribution characteristics of the spatter hotspots in the current frame to evaluate the stability index of the current welding process. Based on the stability index and the deviation of the workpiece diameter, the parameters of the current welding process are adjusted in real time. Based on the grayscale characteristics of different regions in the grayscale image, the weld pool area, heat-affected zone, and spatter hotspots are determined as follows: The density clustering algorithm is used to perform gray-level clustering on the pixels in the welding influence area to determine different clustering regions. The mean gray value of the pixels in each clustering region is calculated as the gray-level index of the corresponding clustering region. Clustered regions whose grayscale index is greater than a preset first index threshold are combined as welding pool regions. In clustered regions other than the weld pool area, the mean value of the gray index difference between each clustered region and all other clustered regions connected to it is calculated. Clustered regions with a mean gray index difference greater than a preset difference threshold and a larger gray index value compared to all other clustered regions connected to it are designated as spatter hotspots. The area excluding the weld pool area and spatter hotspots is designated as the heat-affected zone. The gradient feature sequence for determining the temperature change of the heat-affected zone in different edge pixel directions, radiating outward from the center point of the weld pool region, includes: Using the morphological center point of the weld pool region as the radiation point, rays are connected to different edge pixels of the heat-affected zone. The difference between the gray value of each pixel on the ray and the gray value of the corresponding previous pixel is used as the gray gradient of the pixel. The pixels are sorted in order from near to far from the radiation point to obtain the gradient feature sequence of temperature change in the direction of the corresponding edge pixels.

2. The circumferential weld method for bicycle parts as described in claim 1, characterized in that, The step of determining the workpiece pipe diameter deviation in the current welding area based on the length difference and grayscale distribution of the gradient feature sequence includes: The range of the number of elements contained in different gradient feature sequences is used as the first deviation analysis index. Calculate the standard deviation of all elements in each gradient feature sequence as an indicator of sequence volatility. The standard deviation of all sequence volatility indicators is used as the second deviation analysis indicator; The product of the first deviation analysis index and the second deviation analysis index is normalized and used as the workpiece pipe diameter deviation.

3. The circumferential weld method for bicycle parts as described in claim 1, characterized in that, The analysis of the grayscale change characteristics of the weld pool region in the grayscale images of the current frame and the previous frame, combined with the area and distribution characteristics of the spatter hotspots in the current frame, evaluates the stability index of the current welding process, including: By drawing a perpendicular line along the welding direction through the morphological center point of the weld pool region, the weld pool region is divided into two influence regions. Based on the grayscale changes of pixels in different affected areas along the welding direction, the grayscale change characteristic index of the weld pool area is determined. Calculate the absolute value of the difference between the grayscale change feature index of the current frame and the previous frame, and normalize the negative of the absolute value of the difference as the first welding stability coefficient of the current frame. The second welding stability coefficient is determined based on the area of ​​the spatter hotspot and the density distribution of the spatter hotspot itself. The sum of the first welding stability coefficient and the second welding stability coefficient is normalized and used as the stability index of the current welding process.

4. The circumferential weld method for bicycle parts as described in claim 3, characterized in that, The method of determining the grayscale change characteristic index of the weld pool region based on the grayscale change of pixels in different affected regions along the welding direction includes: Calculate the absolute value of the grayscale value difference between each pixel and the next pixel in the welding direction to obtain the grayscale change value of the corresponding pixel. The average grayscale change value of all pixels in each affected region is taken as the region change value; The absolute value of the difference between the regional change values ​​of the two affected areas is used as the gray-scale change characteristic index of the weld pool area.

5. A method for circumferential welding of bicycle parts as described in claim 4, characterized in that, The determination of the second welding stability coefficient based on the area of ​​the spatter hotspot and the density distribution of the spatter hotspot itself includes: Within a predetermined range centered on the morphological center point of the weld pool region, the number of spatter hotspots contained therein is determined as a density analysis index. The product of the total area of ​​all the spatter hotspots and the density analysis index is normalized, and the negative of the product value is used as the second welding stability coefficient.

6. The circumferential weld method for bicycle parts as described in claim 1, characterized in that, The real-time adjustment of parameters for the current welding process based on the stability index and workpiece diameter deviation includes: Calculate the ratio of workpiece diameter deviation to stability index, and linearly map this ratio to the range of 0-2 as the adjustment index at the current moment; The proportional gain of the PID controller is adjusted according to the aforementioned adjustment index to perform PID control on the welding process.

7. A circumferential weld system for bicycle parts, characterized in that, The system includes: The acquisition module is used to acquire thermal imaging images of bicycle parts during the welding process and perform grayscale processing to obtain grayscale images; The segmentation module is used to determine the weld pool area, heat-affected zone, and spatter hotspots based on the grayscale distribution characteristics of different regions in the grayscale image. The deviation analysis module is used to determine the gradient feature sequence of temperature change in the heat-affected zone in different edge pixel directions, radiating outward from the center point of the weld pool area; and to determine the workpiece pipe diameter deviation in the current welding area based on the length difference and grayscale distribution of the gradient feature sequence. The stability analysis module is used to analyze the grayscale change characteristics of the weld pool area in the grayscale image of the current frame and the previous frame, and to evaluate the stability index of the current welding process by combining the area and distribution characteristics of the spatter hotspots in the current frame. The control module is used to adjust the parameters of the current welding process in real time based on the stability index and the deviation of the workpiece pipe diameter. Based on the grayscale characteristics of different regions in the grayscale image, the weld pool area, heat-affected zone, and spatter hotspots are determined as follows: The density clustering algorithm is used to perform gray-level clustering on the pixels in the welding influence area to determine different clustering regions. The mean gray value of the pixels in each clustering region is calculated as the gray-level index of the corresponding clustering region. Clustered regions whose grayscale index is greater than a preset first index threshold are combined as welding pool regions. In clustered regions other than the weld pool area, the mean value of the gray index difference between each clustered region and all other clustered regions connected to it is calculated. Clustered regions with a mean gray index difference greater than a preset difference threshold and a larger gray index value compared to all other clustered regions connected to it are designated as spatter hotspots. The area excluding the weld pool area and spatter hotspots is designated as the heat-affected zone. The gradient feature sequence for determining the temperature change of the heat-affected zone in different edge pixel directions, radiating outward from the center point of the weld pool region, includes: Using the morphological center point of the weld pool region as the radiation point, rays are connected to different edge pixels of the heat-affected zone. The difference between the gray value of each pixel on the ray and the gray value of the corresponding previous pixel is used as the gray gradient of the pixel. The pixels are sorted in order from near to far from the radiation point to obtain the gradient feature sequence of temperature change in the direction of the corresponding edge pixels.

8. A circumferential weldment device for bicycle parts, the device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the circumferential welding method for bicycle parts as described in any one of claims 1 to 6.

Images