A method for real-time monitoring of welding speed

Abstract

The present application relates to a welding speed real-time monitoring method, comprising the following steps: pasting a reflective strip scale, collecting a welding image, identifying a welding arc position, obtaining a welding arc position coordinate, welding arc position coordinate correction processing, noise reduction filtering processing and welding speed calculation. The method can measure accurate welding speed in real time, and the device is simple and low in cost. The reflective strip scale with periodic pattern is used, the applicability of the system is enhanced, and the difficulty of image recognition algorithm is reduced. Through welding arc position coordinate correction processing, the position misjudgment of the welding point across the period can be avoided, and the welding direction can be obtained. The reflective strip scale is illuminated by the light generated by the welding arc, so that the reflective strip scale does not need to be illuminated by an external power supply or light source, and the problem of unclear image collection caused by insufficient light is avoided.

Description

Technical Field

[0001] This invention relates to the field of welding monitoring, and more particularly to a method for real-time welding speed monitoring. Background Technology

[0002] Welding technology is a crucial component of modern advanced manufacturing, encompassing almost all sectors of the machinery manufacturing industry. The level of welding technology significantly influences the overall level of a nation's manufacturing sector and is closely related to the development of modern advanced manufacturing. Currently, in specific fields such as nuclear power, large natural gas storage tanks, bridges, and pipelines, the complexity of welding operations, high strength requirements, and the need for fieldwork limit the application of general-purpose welding robots. In industrial production, a large amount of welding work is still done manually. However, the lack of mature measurement and monitoring technologies for manual welding process parameters often leads to difficulties in controlling the quality of manual welding. Many industries require 100% inspection of finished welded products, resulting in low work efficiency.

[0003] In the welding process, welding heat input is a key factor affecting weld quality. Excessive welding heat input can easily cause overheating of the joint and heat-affected zone, reducing the hardness and toughness of the weld and heat-affected zone. Insufficient welding heat input, on the other hand, can lead to defects such as incomplete penetration, incomplete fusion, and poor weld formation due to insufficient molten pool temperature. Welding speed is the primary factor influencing welding heat input. Currently, there is no mature solution for measuring welding speed in manual welding, making it difficult to effectively monitor the heat input and control the weld quality. To address this problem, a real-time welding speed monitoring method is proposed to solve the technical challenge of real-time monitoring of welding speed in existing manual welding processes. Summary of the Invention

[0004] This invention provides a method for real-time monitoring of welding speed to solve the technical problem that welding speed is difficult to monitor in real time during existing manual welding processes.

[0005] The technical solution adopted in this invention is as follows:

[0006] A method for real-time monitoring of welding speed, characterized by comprising the following steps:

[0007] S1. Apply reflective strip ruler: Apply reflective strip ruler next to the weld seam in a direction parallel to the weld seam; the application method includes, but is not limited to, adhesive and magnetic adsorption; the reflective strip ruler has a periodic colored pattern with a pattern period of T, and the colors of the reflective strip ruler pattern include, but are not limited to, white, red, yellow, green and blue;

[0008] S2. Acquiring welding images: Install a camera or industrial camera on the welding helmet to acquire welding images. Use the camera or industrial camera on the welding helmet to acquire images of the welding point and reflective strip ruler. Transmit the image acquisition data to the host computer via 5G or WIFI. Place welding goggles in front of the lens of the camera or industrial camera.

[0009] S3. Identify the welding arc position: Identify the area of ​​the welding arc and obtain the center point position a of the welding arc by using image recognition method; draw a perpendicular line from the center point position a of the welding arc to the reflective strip scale, and the corresponding foot position is the position b of the welding arc on the reflective strip scale.

[0010] S4. Obtain the coordinates of the welding arc position: Locate the periodic pattern c of the reflective strip scale where position b is located, and establish a coordinate system d with the starting position of the periodic pattern c as 0, the midpoint as 0.5T, and the ending position as T; determine the position coordinate x of position b on coordinate system d.

[0011] S5. Welding arc position coordinate correction processing: Set the acquisition time of the image data acquisition system to t1, t2, ..., t n-1 , t n , ...; where t n-1 and t n For adjacent data acquisition times, t n-1 and t n The coordinates of the welding arc position determined in step S4 are denoted as x. n-1 and x n , will t n The correction value of the welding arc position coordinates at time x is denoted as x. n The welding arc position coordinates are corrected as follows:

[0012]

[0013] S6. Noise Reduction and Filtering Process: The welding arc position coordinate correction value sequence {x1′, x2′, ..., x} obtained in step S5 is processed by applying the noise reduction and filtering process to the noise reduction and filtering process. n After performing noise reduction filtering on the coordinates {x1″, x2″, ..., x}, a new coordinate sequence is obtained. n The methods for noise reduction filtering include, but are not limited to: low-pass filtering, Gaussian filtering, and mean filtering.

[0014] S7. Welding speed calculation: The coordinate sequence {x1″, x2″, ..., x...} obtained after noise reduction and filtering in step S5. n "} Calculate the welding speed and set t n The velocity at time v n Welding speed v n The calculation method is as follows:

[0015] v n =(x n "-x n-1 ”) / (t n -t n-1 )

[0016] Furthermore, the reflective strip ruler mentioned in step S1 is 1 to 5 cm away from the weld.

[0017] Furthermore, the reflective strip scale pattern mentioned in step S1 is an alternating square pattern or stripe pattern, with a pattern period T of 1 to 10 cm.

[0018] Furthermore, the light transmittance of the welded protective lens described in step S2 is 2% to 15%.

[0019] Furthermore, in the process of acquiring welding images described in step S2, the light illuminating the reflective scale comes from the welding arc.

[0020] The present invention has the following beneficial effects:

[0021] The present invention provides a real-time welding speed monitoring method, comprising the following steps: attaching reflective strip rulers, acquiring welding images, identifying the welding arc position, obtaining welding arc position coordinates, welding arc position coordinate correction processing, noise reduction filtering processing, and welding speed calculation. The method proposed in this invention can measure the welding speed accurately in real time. Compared with the prior art, it has the following advantages: (1) It can conveniently measure the welding speed. The reflective strip scale is arranged by pasting or magnetic adsorption. The camera or industrial camera is mounted on the welding helmet and will not obstruct the view. The equipment arrangement is simple and the cost is low. (2) The reflective strip scale is a periodic pattern. By using image recognition method and the related algorithm proposed in this invention, the position change of adjacent time can be accurately obtained. Then the speed and position can be calculated. There is no need to arrange the scale with absolute coordinate scale. Even if the reflective strip scale is partially damaged, it can still be used, which enhances the applicability of the system and reduces the difficulty of the image recognition algorithm. (3) By correcting the welding arc position coordinates, not only can the position misjudgment of the welding point across the cycle be avoided, but the welding direction can also be obtained. (4) The light generated by the welding arc is used to illuminate the reflective strip scale. There is no need to add an external power supply or light source to illuminate the reflective strip scale. It also avoids the problem of unclear image acquisition due to insufficient light.

[0022] In addition to the objectives, features, and advantages described above, the present invention has other objectives, features, and advantages. The invention will now be described in further detail with reference to the accompanying drawings. Attached Figure Description

[0023] The accompanying drawings, which form part of this application, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0024] Figure 1 This is a flowchart of the steps of a real-time welding speed monitoring method proposed in this invention;

[0025] Figure 2 It is a single-layer periodic grid pattern according to an embodiment of the present invention;

[0026] Figure 3 It is a double-layer periodic grid pattern according to an embodiment of the present invention;

[0027] Figure 4 It is a periodic stripe pattern according to an embodiment of the present invention;

[0028] Figure 5 This is a schematic diagram illustrating the principle of identifying the welding arc position and obtaining the welding arc position coordinates in the method described in this embodiment of the invention. In the diagram, 1 represents the welding arc area, 2 represents the welding arc center point position a, 3 represents the welding arc position b on the reflective strip scale, 4 represents the reflective strip scale, 5 represents the coordinate system d, 6 represents the periodic pattern c of position b on the reflective strip scale, 7 represents the color area before position b, 8 represents the color area where position b is located, and 9 represents the color area after position b. Detailed Implementation

[0029] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0030] Figure 1 This is a flowchart of the steps of a real-time welding speed monitoring method proposed in this invention, including attaching reflective strip rulers, acquiring welding images, identifying the welding arc position, obtaining the welding arc position coordinates, welding arc position coordinate correction processing, noise reduction filtering processing, and welding speed calculation. Figure 2 , Figure 3 and Figure 4 It is a typical periodic pattern of reflective strip scales. Figure 2 It is a single-layer periodic checkered pattern composed of two different colored squares. Figure 3 It is a double-layered periodic checkered pattern composed of two different colored squares. Figure 4 It is a periodic striped pattern composed of two different colored striped areas. Figure 5This is a schematic diagram of the principle of identifying and obtaining the position of the welding arc. The periodic pattern c(6) corresponding to position b on the reflective strip scale can be composed of the color area (7) before position b and the color area (8) where position b is located, or it can be composed of the color area (8) where position b is located and the color area (9) after position b. The selection of the periodic pattern needs to be consistent throughout the process.

[0031] like Figure 1 As shown in this embodiment, a real-time welding speed monitoring method includes the following steps:

[0032] S1. Applying reflective strip rulers: Apply reflective strip rulers next to the weld seam in a direction parallel to the weld seam; the application method includes, but is not limited to, adhesive bonding and magnetic adsorption; the reflective strip rulers have periodic colored patterns with a pattern period of T, and the colors of the reflective strip ruler patterns include, but are not limited to, white, red, yellow, green, and blue; the reflective strip rulers are 1-5cm away from the weld seam; the reflective strip ruler patterns are alternating square or stripe patterns with a pattern period T of 1-10cm;

[0033] S2. Acquiring Welding Images: A camera or industrial camera for acquiring welding images is installed on the welding helmet. The camera or industrial camera on the welding helmet is used to acquire images of the welding point and the reflective strip ruler. The image acquisition data is transmitted to the host computer via 5G or WIFI. A welding safety lens is placed in front of the lens of the camera or industrial camera. The light transmittance of the welding safety lens is 2% to 15%. During the acquisition of welding images, the light illuminating the reflective strip ruler comes from the welding arc.

[0034] S3. Identify the welding arc position: Identify the area of ​​the welding arc and obtain the center point position a of the welding arc by using image recognition method; draw a perpendicular line from the center point position a of the welding arc to the reflective strip scale, and the corresponding foot position is the position b of the welding arc on the reflective strip scale.

[0035] S4. Obtain the coordinates of the welding arc position: Locate position b in the periodic pattern c of the reflective strip scale, and establish a coordinate system d with the starting position of the periodic pattern c as 0, the midpoint as 0.5T, and the ending position as T; determine the position coordinate x of position b in coordinate system d.

[0036] S5. Welding arc position coordinate correction processing: Set the acquisition time of the image data acquisition system to t1, t2, ..., t n-1 , t n , ...; where t n-1 and t n For adjacent data acquisition times, t n-1 and t n The coordinates of the welding arc position determined in step S4 are denoted as x.n-1 and x n , will t n The correction value of the welding arc position coordinates at time x is denoted as x. n The welding arc position coordinates are corrected as follows:

[0037]

[0038] S6. Noise Reduction and Filtering: The welding arc position coordinate correction value sequence {x1′, x2′, ..., x} obtained in step S5 is processed by noise reduction and filtering. n After performing noise reduction filtering on the coordinates {x1″, x2″, ..., x}, a new coordinate sequence is obtained. n The methods for noise reduction filtering include, but are not limited to, low-pass filtering, Gaussian filtering, and mean filtering.

[0039] S7. Welding speed calculation: Based on the coordinate sequence {x1″, x2″, ..., x...} obtained after noise reduction and filtering in step S5. n ″}, calculate the welding speed, and set t n The velocity at time v n Welding speed v n The calculation method is as follows:

[0040] v n =(x n "-x n-1 ”) / (t n -t n-1 )

[0041] In the above-mentioned real-time welding speed monitoring method, the reflective strip scale can be... Figure 2 , Figure 3 or Figure 4 The periodic pattern shown can also be other periodic patterns, such as diagonal stripes or multi-layered periodic checkered patterns. The color of the reflective strip scale can be any combination of two of white, red, yellow, green, and blue, or it can be a periodic pattern composed of multiple colors. The welding goggles in step S2 can be welding goggles with a light transmittance of 2% to 15%, or commercially available welding goggles with color numbers 7 to 12. The principle for identifying and obtaining the welding arc position is as follows: Figure 5As shown, the pattern period 'c' on the reflective strip scale can be selected arbitrarily. For example, a periodic pattern composed of yellow and green squares can have the pattern period from left to right alternating between yellow and green squares, or vice versa. This method allows for arbitrary selection of the pattern period while ensuring that the final welding speed value and accuracy remain unaffected, greatly improving its applicability in practical applications. Furthermore, if the reflective strip scale is partially damaged, it can be removed, and the remaining portion can continue to be used. Figure 5 As shown, in step S4, the coordinate system d is a one-dimensional coordinate system Ox, with the origin being the starting position of the pattern period c, and the direction parallel to the reflective strip scale pointing to the end point of the pattern period c is the positive direction of the coordinate system Ox.

[0042] Example 1:

[0043] See Figure 1 , Figure 2 and Figure 5 The present invention provides a method for real-time monitoring of welding speed, comprising the following steps:

[0044] S1. Applying reflective strip markers: Apply reflective strip markers parallel to the weld seam, 1cm away from the weld seam, using adhesive. The reflective strip markers have periodic colored patterns with a period of 1cm. The colors of the reflective strip marker patterns are white and red. The reflective strip marker patterns are as follows: Figure 2 The single-layer periodic checkered pattern shown;

[0045] S2. Acquiring Welding Images: A camera for acquiring welding images is installed on the welding helmet. The camera on the welding helmet is used to acquire images of the welding point and the reflective strip ruler. The image acquisition data is transmitted to the host computer via 5G. A welding safety lens is placed in front of the lens of the camera. The light transmittance of the welding safety lens is 2%. During the process of acquiring welding images, the light illuminating the reflective strip ruler comes from the welding arc.

[0046] S3. Identify the welding arc position: Identify the area of ​​the welding arc and obtain the center point position a of the welding arc by using image recognition method; draw a perpendicular line from the center point position a of the welding arc to the reflective strip scale, and the corresponding foot position is the position b of the welding arc on the reflective strip scale.

[0047] S4. Obtain the coordinates of the welding arc position: Locate position b in the periodic pattern c of the reflective strip scale, and establish a coordinate system d with the starting position of the periodic pattern c as 0, the midpoint as 0.5T, and the ending position as T; determine the position coordinate x of position b in coordinate system d.

[0048] S5. Welding arc position coordinate correction processing: Set the acquisition time of the image data acquisition system to t1, t2, ..., t n-1 , t n , ...; where t n-1 and t n For adjacent data acquisition times, t n-1 and t n The coordinates of the welding arc position determined in step S4 are denoted as x. n-1 and x n , will t n The correction value of the welding arc position coordinates at time x is denoted as x. n The welding arc position coordinates are corrected as follows:

[0049]

[0050] S6. Noise Reduction and Filtering: The welding arc position coordinate correction value sequence {x1′, x2′, ..., x} obtained in step S5 is processed by noise reduction and filtering. n After performing noise reduction filtering on the coordinates {x1″, x2″, ..., x}, a new coordinate sequence is obtained. n The methods for noise reduction filtering include, but are not limited to, low-pass filtering, Gaussian filtering, and mean filtering.

[0051] S7. Welding speed calculation: Based on the coordinate sequence {x1″, x2″, ..., x...} obtained after noise reduction and filtering in step S5. n ″}, calculate the welding speed, and set t n The velocity at time v n Welding speed v n The calculation method is as follows:

[0052] v n =(x n "-x n-1 ”) / (t n -t n-1 )

[0053] Example 2:

[0054] See Figure 1 , Figure 3 and Figure 5 The present invention provides a method for real-time monitoring of welding speed, comprising the following steps:

[0055] S1. Applying reflective strip markers: Apply reflective strip markers parallel to the weld seam, 5cm away from the weld seam, using magnetic adsorption. The reflective strip markers have periodic colored patterns with a period of 10cm. The colors of the reflective strip marker patterns are yellow and green. The reflective strip marker patterns are as follows: Figure 3The double-layered periodic checkered pattern shown;

[0056] S2. Acquiring Welding Images: An industrial camera for acquiring welding images is installed on the welding helmet. The industrial camera on the welding helmet is used to acquire images of the welding point and the reflective strip ruler. The image acquisition data is transmitted to the host computer via WIFI. A welding safety lens is placed in front of the lens of the industrial camera. The light transmittance of the welding safety lens is 15%. During the acquisition of welding images, the light illuminating the reflective strip ruler comes from the welding arc.

[0057] S3. Identify the welding arc position: Identify the area of ​​the welding arc and obtain the center point position a of the welding arc by using image recognition method; draw a perpendicular line from the center point position a of the welding arc to the reflective strip scale, and the corresponding foot position is the position b of the welding arc on the reflective strip scale.

[0058] S4. Obtain the coordinates of the welding arc position: Locate position b in the periodic pattern c of the reflective strip scale, and establish a coordinate system d with the starting position of the periodic pattern c as 0, the midpoint as 0.5T, and the ending position as T; determine the position coordinate x of position b in coordinate system d.

[0059] S5. Welding arc position coordinate correction processing: Set the acquisition time of the image data acquisition system to t1, t2, ..., t n-1 , t n , ...; where t n-1 and t n For adjacent data acquisition times, t n-1 and t n The coordinates of the welding arc position determined in step S4 are denoted as x. n-1 and x n , will t n The correction value of the welding arc position coordinates at time x is denoted as x. n The welding arc position coordinates are corrected as follows:

[0060]

[0061] S6. Noise Reduction and Filtering: The welding arc position coordinate correction value sequence {x1′, x2′, ..., x} obtained in step S5 is processed by noise reduction and filtering. n After performing noise reduction filtering on the coordinates {x1″, x2″, ..., x}, a new coordinate sequence is obtained. n The methods for noise reduction filtering include, but are not limited to, low-pass filtering, Gaussian filtering, and mean filtering.

[0062] S7. Welding speed calculation: Based on the coordinate sequence {x1″, x2″, ..., x...} obtained after noise reduction and filtering in step S5. n ″}, calculate the welding speed, and set tn The velocity at time v n Welding speed v n The calculation method is as follows:

[0063] v n =(x n "-x n-1 ”) / (t n -t n-1 )

[0064] Example 3:

[0065] See Figure 1 , Figure 4 and Figure 5 The present invention provides a method for real-time monitoring of welding speed, comprising the following steps:

[0066] S1. Applying reflective strip markers: Apply reflective strip markers parallel to the weld seam, 3cm away from the weld seam, using adhesive. The reflective strip markers have periodic colored patterns with a period of 5cm. The colors of the reflective strip marker patterns are blue and red. The reflective strip marker patterns are as follows: Figure 4 The periodic stripe pattern shown;

[0067] S2. Acquiring Welding Images: A camera for acquiring welding images is installed on the welding helmet. The camera on the welding helmet is used to acquire images of the welding point and the reflective strip ruler. The image acquisition data is transmitted to the host computer via WIFI. A welding safety lens is placed in front of the lens of the camera. The light transmittance of the welding safety lens is 10%. During the acquisition of welding images, the light illuminating the reflective strip ruler comes from the welding arc.

[0068] S3. Identify the welding arc position: Identify the area of ​​the welding arc and obtain the center point position a of the welding arc by using image recognition method; draw a perpendicular line from the center point position a of the welding arc to the reflective strip scale, and the corresponding foot position is the position b of the welding arc on the reflective strip scale.

[0069] S4. Obtain the coordinates of the welding arc position: Locate position b in the periodic pattern c of the reflective strip scale, and establish a coordinate system d with the starting position of the periodic pattern c as 0, the midpoint as 0.5T, and the ending position as T; determine the position coordinate x of position b in coordinate system d.

[0070] S5. Welding arc position coordinate correction processing: Set the acquisition time of the image data acquisition system to t1, t2, ..., t n-1 , t n , ...; where t n-1 and t n For adjacent data acquisition times, t n-1 and tn The coordinates of the welding arc position determined in step S4 are denoted as x. n-1 and x n , will t n The correction value of the welding arc position coordinates at time x is denoted as x. n The welding arc position coordinates are corrected as follows:

[0071]

[0072] S6. Noise Reduction and Filtering: The welding arc position coordinate correction value sequence {x1′, x2′, ..., x} obtained in step S5 is processed by noise reduction and filtering. n After performing noise reduction filtering on the coordinates {x1″, x2″, ..., x}, a new coordinate sequence is obtained. n The methods for noise reduction filtering include, but are not limited to, low-pass filtering, Gaussian filtering, and mean filtering.

[0073] S7. Welding speed calculation: Based on the coordinate sequence {x1″, x2″, ..., x...} obtained after noise reduction and filtering in step S5. n ″}, calculate the welding speed, and set t n The velocity at time v n Welding speed v n The calculation method is as follows:

[0074] v n =(x n "-x n-1 ”) / (t n -t n-1 )

[0075] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for real-time monitoring of welding speed, characterized in that, Includes the following steps: S1. Apply reflective strip ruler: Apply reflective strip ruler next to the weld seam in a direction parallel to the weld seam; the application method includes adhesive or magnetic adsorption; the reflective strip ruler has a periodic colored pattern with a pattern period of T, and the color of the reflective strip ruler pattern includes white, red, yellow, green or blue; S2. Acquiring welding images: Install a camera or industrial camera on the welding helmet to acquire welding images. Use the camera or industrial camera on the welding helmet to acquire images of the welding point and reflective strip ruler. Transmit the image acquisition data to the host computer via 5G or WIFI. Place welding goggles in front of the lens of the camera or industrial camera. S3. Identify the welding arc position: Identify the area of ​​the welding arc and obtain the position a of the center point of the welding arc through image recognition. Draw a perpendicular line from position a, the center point of the welding arc, to the reflective strip scale. The corresponding foot position of the perpendicular is the position b of the welding arc on the reflective strip scale. S4. Obtain the coordinates of the welding arc position: Locate position b in the periodic pattern c of the reflective strip scale, and establish a coordinate system d with the starting position of the periodic pattern c as 0, the midpoint as 0.5T, and the ending position as T; determine the position coordinate x of position b in coordinate system d. S5. Welding arc position coordinate correction processing: Set the acquisition time of the image data acquisition system to t1, t2, ..., t n-1 , t n , ...; where t n-1 and t n For adjacent data acquisition times, t n-1 and t n The coordinates of the welding arc position determined in step S4 are denoted as x. n-1 and x n , will t n The correction value of the welding arc position coordinates at time is denoted as The welding arc position coordinates are corrected as follows: S6. Noise Reduction and Filtering: The welding arc position coordinate correction value sequence obtained in step S5 is processed by... , , ..., After performing noise reduction filtering, a new coordinate sequence is obtained. ,… Noise reduction filtering methods include low-pass filtering, Gaussian filtering, and mean filtering. S7. Welding speed calculation: The coordinate sequence obtained after noise reduction and filtering in step S6 { , , ..., Calculate the welding speed and set t. n The velocity at time v n Welding speed v n The calculation method is shown in the following formula: 。 2. The method for real-time monitoring of welding speed as described in claim 1, characterized in that: The distance between the reflective strip ruler and the weld seam in step S1 is 1~5cm.

3. The method for real-time monitoring of welding speed as described in claim 1, characterized in that: The reflective strip scale pattern mentioned in step S1 is an alternating square pattern or stripe pattern, with a pattern period T of 1~10cm.

4. The method for real-time monitoring of welding speed as described in claim 1, characterized in that: The light transmittance of the welding goggles described in step S2 is 2% to 15%.

5. The method for real-time monitoring of welding speed as described in claim 1, characterized in that: In the process of acquiring welding images described in step S2, the light illuminating the reflective scale comes from the welding arc.

Images