A welding model-based management method and system

Abstract

This invention discloses a welding model-based management method and system, including the acquisition of welding data, which includes welding structure data and weld seam data. The surface morphology of the weld seam is captured in images, and the acquired data is stored. Based on the acquired welding structure data, a 3D model of the welding structure is created using 3D modeling software. A lightweight engine supports interaction, and the weld seams are drawn in the 3D model. Each weld seam is assigned a unique ID and its status is indicated by color. This invention utilizes the acquisition of weld seam and welding structure data, creates 3D models of the weld seams, and establishes a unified weld seam platform. Clicking on a weld seam in the 3D model focuses on its physical location, and associated information such as weld seam attributes, process cards, and inspection reports can be retrieved with a single click, eliminating cross-system retrieval costs. This facilitates model-based management of weld seams, reducing weld seam search time and improving work efficiency.

Description

Technical Field

[0001] This invention relates to the field of welding model management, and in particular to a welding model management method and system. Background Technology

[0002] Welding, also known as fusion welding, is a manufacturing process and technology that uses heating, high temperature or high pressure to join metals or other thermoplastic materials such as plastics. Welding has a wide range of applications, and welding management is usually required when performing welding.

[0003] Current welding management typically relies on two-dimensional drawings, paper-based process documents, and scattered inspection records. Locating weld seams requires retrieving the corresponding weld seam document. Furthermore, if welding defects occur during the welding process, they are often not detected promptly. Instead, workers inspect the weld seam after welding, and if defects are found, they must then refer to the drawings to pinpoint the defect location. This method is time-consuming, labor-intensive, and inefficient. Therefore, it is necessary to implement model-based management of the welding structure during welding to monitor for welding defects in real time and directly locate and inspect the weld seams. Summary of the Invention

[0004] The purpose of this invention is to provide a welding modeling management method and system to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a welding modeling management method, comprising:

[0006] Step 1: Collect welding data, including welding structure data and weld seam data. Also, collect images of the surface morphology of the weld seam, identify the images, store the collected data, and establish a database.

[0007] Step 2: Based on the collected welding structure data, create a 3D model of the welding structure using 3D modeling software, use a lightweight engine to support interaction, and draw the weld seam in the 3D model;

[0008] Step 3: Based on the completed 3D model of the welded structure, input the information of each weld, assign a unique ID to each weld, use color to identify the weld status, and record the parameters of the weld welding process. Store the parameters and weld information in the database and associate them with the weld ID.

[0009] Step 4: Build a unified platform and develop weld positioning and attribute query functions based on the 3D model. The platform will input weld association information and images to realize information linkage function.

[0010] Step 5: Based on weld information, use welding process simulation software to plan and simulate the welding process, collect information during the simulation process, and use the simulation results to make predictions and generate set process parameters;

[0011] Step Six: Weld the structure. During the welding process, monitor it in real time, collect welding parameters and compare them with the set process parameters, and identify and locate defects. After welding is completed, the welding data is stored in the database.

[0012] Step 7: Conduct quality inspection on the completed weld, record the inspection results in the platform, and link them with the weld 3D model and related information. Based on the inspection results, determine whether the requirements are met.

[0013] Preferably, the welding structure data in step one includes shape, size and assembly relationship, and the weld data includes the shape of the weld, the location of the weld and the size of the weld.

[0014] Preferably, the acquisition and identification of the weld surface morphology image in step one specifically includes:

[0015] High-definition camera equipment and laser scanning equipment are used to acquire images of the initial surface appearance of the weld.

[0016] The defect identification algorithm identifies and determines whether there are defects in the initial appearance of the weld in real time. If defects are found, a defect mark is generated and the type and size of the defect mark are recorded. Then, when the 3D model is built, a corresponding warning mark is made on the 3D model, and a processing suggestion for the defect location is generated so that a matching welding method can be adopted during the welding process.

[0017] Preferably, the lightweight engine in step two supports interaction specifically including support for weld geometry feature annotation and process attribute binding. The weld geometry features include bevel type, length, and spatial coordinates, and the process attribute binding includes welding method, filler material, and preheating temperature.

[0018] Preferably, the weld information in step three includes weld type, weld grade, welding method, and welding material. The weld type includes butt weld, fillet weld, and fillet weld. The welding method includes manual arc welding, gas shielded welding, submerged arc welding, and welding robot. The welding material includes welding rod, welding wire, and flux. The weld process parameters in step three include welding current, voltage, and welding speed.

[0019] Preferably, the weld ID in step three includes project number, region, component number and weld serial number. The color-coded weld status in step three specifically includes the use of three colors: green, orange and red. Green represents qualified, orange represents one rework and red represents two reworks. The weld-related information entered into the platform in step four includes weld drawings, welding procedure qualification report, welding quality plan, inspection report and sampling information.

[0020] Preferably, step four, which involves platform construction and development of weld seam positioning and attribute query functions, specifically includes:

[0021] Front-end and back-end development. Front-end development includes designing the user interface, which provides a weld selection list, a 3D model display area, an attribute information display area, and a weld image display area. The weld selection list is categorized by weld type and location. The 3D model display area provides rotation, zoom, and pan buttons to view the model from different angles. Back-end development includes designing an interface for obtaining weld location information. It receives the weld ID from the front-end, queries the corresponding geometric information from the database and returns it. It supports queries based on weld ID or query conditions and returns the corresponding attribute data. It implements the query logic for weld location and attribute information, accurately retrieves data from the database according to the front-end request, and uses database connection pooling technology for access.

[0022] Using a 3D model processing library, the processed 3D model is loaded into the front-end page and rendered.

[0023] When a user selects a weld in the weld selection list, the front end requests the location information of the weld in the 3D model from the back end through the API interface. The back end queries the relevant information from the database based on the weld identifier and returns it to the front end. After receiving the location information, the front end highlights the weld in the 3D model. The highlighting is achieved by changing the color of the weld and adding a mark.

[0024] After the weld is located, the front end automatically obtains the attribute information of the weld from the back end and displays it in the attribute information display area. The front end enters the query conditions in the attribute information display area. The query conditions include weld type and welding material. The front end sends the query conditions to the back end, which queries the database and returns the results to the front end for display.

[0025] Preferably, in step four, the weld seam positioning is achieved by defining a coordinate range for each weld seam in the three-dimensional model. When a user selects a weld seam, the corresponding coordinate range is obtained based on its ID, and the geometric elements within that range are searched in the three-dimensional model.

[0026] Preferably, the welding process in step six is ​​monitored in real time, and the welding parameters are compared with the set process parameters, as well as defects are identified and located. Specifically, this includes:

[0027] During the welding process, sensors, monitoring equipment, and laser scanning equipment acquire real-time images of the welding process and collect parameters such as welding current, welding temperature, and welding speed, and then transmit the data in real time.

[0028] The parameters collected in real time are compared and analyzed with the preset process parameters. If the parameters exceed the allowable range, an alarm is issued and abnormal information is displayed on the platform.

[0029] During image acquisition, the defect recognition algorithm determines in real time whether there are defects, including cracks, pores and deformations. If a defect is found at the weld position, the platform maps the defect coordinates to the 3D model and generates an early warning mark, which flashes red on the platform. After clicking on the flashing position, the early warning mark is opened and the high-definition detection image of the defect and the quantitative data of the defect size are viewed.

[0030] The platform maps defect coordinates to a 3D model by attaching reflective target balls at the start and end points and corners of the weld, scanning the target balls with a laser tracker to obtain physical coordinates, marking the corresponding point coordinates in the digital model, establishing weld point coordinates in the 3D model, and outputting defect point cloud coordinates after the monitoring equipment detects the defect point. After receiving the converted defect coordinates and attributes, it generates spatial anchor points at the corresponding positions in the 3D model, and renders visualization elements and generates warning markers according to the defect type.

[0031] A welding modeling management system, comprising:

[0032] The data acquisition module is used to acquire welding data, including welding structure data and weld seam data. It also acquires images of the surface morphology of the weld seam, identifies the images, stores the acquired data, and establishes a database.

[0033] A 3D model creation module is used to create a 3D model of the welded structure based on the collected welded structure data using 3D modeling software, support interaction using a lightweight engine, and draw the weld seam in the 3D model.

[0034] The information entry module is used to enter the information of each weld seam based on the completed 3D model of the welded structure, assign a unique ID to each weld seam and use color to identify the weld seam status, and record the parameters of the weld seam welding process, so that the parameters and weld seam information are stored in the database and associated with the weld seam ID.

[0035] The platform construction module is used to build a unified platform and develop weld seam positioning and attribute query functions based on the 3D model. The platform inputs weld seam association information and images to realize information linkage function.

[0036] The welding process simulation module is used to plan and simulate welding processes based on weld information using welding process simulation software, collect information during the simulation process, and use the simulation results to make predictions and generate set process parameters.

[0037] A real-time monitoring module is used to weld the structure. During the welding process, it performs real-time monitoring, collects welding parameters and compares them with set process parameters, and identifies and locates defects. After welding is completed, the welding data is stored in a database. The defect identification and location specifically includes:

[0038] During image acquisition, the defect recognition algorithm determines in real time whether there are defects, including cracks, pores and deformations. If a defect is found at the weld position, the platform maps the defect coordinates to the 3D model and generates an early warning mark, which flashes red on the platform. After clicking on the flashing position, the early warning mark is opened and the high-definition detection image of the defect and the quantitative data of the defect size are viewed.

[0039] The platform maps defect coordinates to a 3D model by attaching reflective target balls at the start and end points and corners of the weld, scanning the target balls with a laser tracker to obtain physical coordinates, marking the corresponding point coordinates in the digital model, establishing weld point coordinates in the 3D model, and outputting defect point cloud coordinates after the monitoring equipment detects the defect point. After receiving the converted defect coordinates and attributes, it generates spatial anchor points at the corresponding positions in the 3D model, and renders visualization elements and generates warning markers according to the defect type.

[0040] The quality inspection module is used to inspect the quality of the weld after welding is completed, and to record the inspection results into the platform. It is also linked with the 3D model of the weld and related information. Based on the inspection results, it is determined whether the requirements are met.

[0041] The technical effects and advantages of this invention are as follows:

[0042] (1) This invention utilizes data acquisition of weld seams and welded structures, creates a three-dimensional model of the weld seam, and establishes a unified platform for the weld seam. By clicking on the three-dimensional model weld seam, the physical location of the weld seam, the weld seam attributes, process cards, inspection reports and other related information can be retrieved with one click, eliminating the cost of cross-system retrieval. This facilitates the model-based management of weld seams, reduces the time spent searching for weld seams, and improves work efficiency.

[0043] (2) This invention utilizes the preliminary acquisition of weld morphology and real-time monitoring of weld during the welding process, as well as defect identification and location, to acquire the unwelded appearance of the weld, which is beneficial for adjusting the welding method accordingly. At the same time, the specific information of the preliminary weld is marked on the three-dimensional model. Meanwhile, during the welding process, the weld is monitored in real time, and the porosity / cracks identified by visual detection are mapped to the model in real time and given an early warning, which is beneficial for real-time monitoring of whether there are welding defects.

[0044] (3) By assigning an independent ID to each weld and using color to identify the weld status, this invention facilitates direct ID lookup of welds, improves weld lookup efficiency, enables rapid three-dimensional positioning, and allows for linkage with information for already modeled welds. Based on a unified platform, it enables querying related information such as weld three-dimensional models, drawings, welding process qualification, welding quality plan, inspection reports, and sampling information. Attached Figure Description

[0045] Figure 1 This is a schematic diagram of the welding modeling management method of the present invention.

[0046] Figure 2 This is a schematic diagram of the surface morphology image acquisition and recognition process of the weld seam in this invention.

[0047] Figure 3 This is a schematic diagram illustrating the process of building and developing the weld seam positioning and attribute query functions for the platform of this invention.

[0048] Figure 4 This is a block diagram of the welding modeling management system of the present invention. Detailed Implementation

[0049] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0050] This invention provides, for example Figure 1-4 The welding modeling management method shown includes the following specific steps:

[0051] Step 1: Collect welding data. Welding data includes weld structure data and weld seam data. Weld structure data includes shape, size, and assembly relationships, which facilitates a thorough understanding of the weld structure information for modeling. Weld seam data includes the weld shape, location, and size. Images of the weld seam surface morphology are captured and recognized, allowing for information on the welding surfaces to be welded before the structure is actually welded. This facilitates advance knowledge of weld seam information, enabling the adoption of appropriate welding methods and improving welding efficiency. The collected data is stored, allowing for the storage of the number of weld structures for later direct retrieval and historical understanding, and the establishment of a database. The specific steps of weld seam surface morphology image acquisition and recognition include:

[0052] The high-definition camera and laser scanning equipment work together as follows: The laser scanner projects blue structured light stripes (anti-arc interference wavelength) onto the weld surface, while the high-definition camera captures stripe deformation images under synchronous trigger control. The strong infrared light of the welding arc is filtered out by a narrow-band filter to ensure that the image signal-to-noise ratio is >35dB. The scanner moves at a speed of 120 lines / second, covering the full width of the weld. The camera acquires stripe images at 5 megapixels / 120fps, forming the original morphology data stream. 3D morphology generation is based on the principle of triangulation. By calculating the pixel displacement of the laser stripes in the image, combined with the baseline distance and focal length of the camera-laser, the surface height is calculated in real time, and the sub-pixel level stripe center line is extracted. Finally, a point 3D surface model is generated, which can detect undercut, porosity, and poor fusion defects ≥0.03mm. The laser scanning equipment is the laser scanner, which is beneficial for generating high-precision point cloud data, resisting arc light interference, and acquiring images of the initial surface morphology of the weld.

[0053] The defect recognition algorithm performs real-time identification of the initial appearance of the weld to determine whether defects exist. The recognition steps of the algorithm are as follows: First, the acquired weld surface image is preprocessed (median filtering for noise reduction and contrast enhancement). Then, a convolutional neural network is used to extract the deep visual features of the defects (texture abrupt changes and geometric anomalies). Subsequently, the defect type (porosity / crack / undercut) and confidence level are output through the classification branch, while the defect location and bounding box are calculated through the regression branch. Finally, the defect depth is verified by combining 3D point cloud data (e.g., whether the crack extends into the interior), and false detections are filtered out with a confidence threshold (>0.7). This achieves a high efficiency detection with an accuracy of ≥95% and a single frame processing time of ≤100ms. If a defect exists, a defect marker is generated, and the type and size of the defect marker are recorded. Then, when the 3D model is built, a corresponding warning marker is made on the 3D model, and a processing suggestion for the defect location is generated so that a matching welding method can be adopted during the welding process.

[0054] Step 2: Based on the collected welding structure data, create a 3D model of the welding structure using 3D modeling software. Use the lightweight engine to support interaction and draw the weld in the 3D model. The lightweight engine supports interaction by supporting weld geometric feature annotation and process attribute binding. Weld geometric features include groove type, length and spatial coordinates. Process attribute binding includes welding method, filler material and preheating temperature.

[0055] Step 3: Based on the completed 3D model of the welded structure, input the information of each weld, assign a unique ID to each weld, and use color to indicate the weld status. This helps to give each weld its own unique identity for easy retrieval. Record the welding process parameters and store the parameters and weld information in the database, as well as link them to the weld ID. Weld information includes weld type, weld grade, welding method, and welding materials. Weld types include butt welds, fillet welds, and fillet welds. Welding methods include manual arc welding, gas shielded welding, submerged arc welding, and welding robots. Welding materials include welding rods, welding wires, and flux. Weld process parameters include welding current, voltage, and welding speed. Weld ID includes project number, area, component number, and weld serial number. The color-coded weld status uses green, orange, and red. Green represents qualified, orange represents one rework, and red represents two reworks. This allows for observation of the weld's status based on the color coding, providing a more intuitive understanding.

[0056] Step 4: Construct a unified platform and develop weld location and attribute query functions based on the 3D model. The platform inputs weld association information and images to achieve information linkage. The weld association information input by the platform includes weld drawings, welding procedure qualification reports, welding quality plans, inspection reports, and sampling information. The specific construction and development of the weld location and attribute query functions include:

[0057] Front-end and back-end development are involved. Front-end development includes designing the user interface, which provides a weld selection list, a 3D model display area, an attribute information display area, and a weld image display area. This facilitates user interaction with the platform using front-end technologies, such as weld selection, 3D model rotation, zooming, and attribute querying. For example, clicking on a weld in the weld selection list displays the weld in the 3D model display area and its detailed information in the attribute information display area. The weld selection list is categorized by weld type and location. The 3D model display area provides rotation, zoom, and translation buttons to view the model from different angles. Back-end development includes designing an interface for obtaining weld location information. This interface receives the weld ID from the front-end, queries the corresponding geometric information from the database, and returns it. It supports queries based on weld ID or query conditions and returns the corresponding attribute data. It implements the query logic for weld location and attribute information, accurately retrieves data from the database based on front-end requests, and uses database connection pooling technology for access. Back-end development facilitates the input, query, modification, and deletion of weld information, as well as the implementation of welding process planning and simulation algorithms, and the statistical analysis of welding quality inspection data.

[0058] Using a 3D model processing library, the processed 3D model is loaded into the front-end page and rendered.

[0059] When a user selects a weld in the weld selection list, the front end requests the location information of the weld in the 3D model from the back end through the API interface. The back end queries the relevant information from the database based on the weld identifier and returns it to the front end. After receiving the location information, the front end highlights the weld in the 3D model. The highlighting is achieved by changing the color of the weld and adding a mark.

[0060] After the weld is located, the front end automatically obtains the attribute information of the weld from the back end and displays it in the attribute information display area. The front end enters the query conditions in the attribute information display area. The query conditions include weld type and welding material. The front end sends the query conditions to the back end, the back end queries the database and returns the results to the front end for display.

[0061] Step 5: Based on weld information, use welding process simulation software to plan and simulate the welding process, collect information during the simulation process, and use the simulation results to make predictions and generate set process parameters. This will help to optimize and adjust the welding process based on the simulation results, such as adjusting the welding sequence and welding parameters, to ensure welding quality.

[0062] Step Six: Weld the structure. During the welding process, monitor it in real time, collect welding parameters and compare them with the set process parameters, and identify and locate defects. After welding is completed, the welding data is stored in the database.

[0063] Step 7: Conduct quality inspection on the completed welds. Inspection methods include visual inspection, non-destructive testing (such as ultrasonic testing, radiographic testing, and magnetic particle testing), and mechanical property testing. The inspection results are entered into the management system and linked with the weld's 3D model and related information. Based on the inspection results, the quality of the weld is evaluated, and the results are entered into the platform and linked with the weld's 3D model and related information. Based on the inspection results, it is determined whether the requirements are met. This also helps to record the reasons for non-compliance and the handling measures for non-compliant welds, and mark them in the 3D model for subsequent tracking and handling.

[0064] Specifically, in step four, weld seam positioning involves defining a coordinate range for each weld seam in the 3D model. When a user selects a weld seam, the corresponding coordinate range is obtained based on its ID, and the geometric elements within that range are searched in the 3D model. The weld seam positioning includes the following algorithm formula:

[0065] ,

[0066] in, Indicates the left and right bright line segments. This represents the rightmost end of the line drawing. This represents the leftmost part of the line drawing. The x-coordinate of a point The ordinate of a point. The x-coordinate represents the target weld. The vertical coordinate of the target weld is used to locate the weld and directly obtain its coordinate position.

[0067] Specifically, step six involves real-time monitoring of the welding process, comparing welding parameters with the set process parameters, and identifying and locating defects. This includes:

[0068] During welding, sensors, monitoring equipment, and laser scanning equipment acquire images of the welding process in real time. These sensors include arc sensors, vision sensors, laser sensors, temperature sensors, pressure sensors, displacement sensors, and ultrasonic sensors. Arc sensors detect changes in the physical characteristics of the welding arc (such as current, voltage, and arc light) to determine the relative position of the welding torch and the weld seam in real time, enabling automatic weld seam tracking. Combined with changes in welding parameters, the height, angle, and speed of the welding torch are adjusted to ensure the stability of the welding process. Vision sensors capture images of the weld seam using a camera, extract weld seam features using image processing technology, and achieve accurate weld seam identification and positioning. They also monitor the shape, size, and temperature distribution of the weld pool in real time to determine welding quality, identify defects such as porosity and cracks during the welding process, and adjust welding parameters promptly. Laser sensors scan the weld seam using a laser beam, detecting changes in the laser reflection signal to... The system features high-precision tracking of weld seams to acquire their 3D contour information, providing precise path planning for the welding robot. It also monitors weld seam width in real time to ensure consistent welding quality. Temperature sensors monitor the temperature of the welding area in real time to prevent welding defects caused by overheating or undercooling, and control heat input by adjusting welding parameters to ensure the performance of the weld joint. Pressure sensors monitor the contact force between the welding torch and the workpiece to ensure stable pressure during welding. Changes in contact force are used to determine the quality of the weld joint. Displacement sensors monitor the position and displacement of the welding torch in real time to ensure the accuracy of the welding path. They also monitor workpiece deformation during welding and adjust welding parameters accordingly. Ultrasonic sensors detect internal weld defects such as cracks and porosity by utilizing the reflection and propagation characteristics of ultrasonic waves. They measure workpiece thickness to ensure the strength of the weld joint and collect welding current, welding temperature, and welding speed parameters, then transmit the data in real time.

[0069] The parameters collected in real time are compared and analyzed with the preset process parameters. If the parameters exceed the allowable range, an alarm is issued and abnormal information is displayed on the platform.

[0070] During image acquisition, the defect recognition algorithm determines the presence of defects in real time. Defects include cracks, porosity, and deformation. If a defect is found at the weld location, the platform maps the defect coordinates to the 3D model and generates a warning marker, which flashes red on the platform. Clicking on the flashing location opens the warning marker and allows viewing of a high-resolution image of the defect and quantitative data on its size. The formulas involved in defect recognition are as follows:

[0071]

[0072] in, Indicates the confidence level that a defect exists. This represents the activation function. The bias term of the detection box. Indicates the convolution kernel weights. Represents the pixel features of the input image. This indicates the kernel size. This formula is used to perform binary classification of porosity and cracks in weld images and outputs bounding box coordinates.

[0073] The formulas for mapping coordinates are as follows:

[0074]

[0075] in, This indicates the depth of the defect point from the camera. Indicates the camera's focal length. Indicates the baseline distance of the binocular cameras. This formula represents the disparity between corresponding points in the left and right images. It is used to map the defect location in the 2D image to the coordinate system of the 3D model of the weld.

[0076] The mapped coordinates also include the following formula:

[0077]

[0078] in, Indicates the position correction amount. Indicates the speed of the welding torch movement. Indicates the acceleration of the welding torch. This indicates the delay time from image acquisition to processing. This method is used to ensure that the error between the marked position of the defect in the 3D model and the actual position is ≤0.5mm.

[0079] The platform maps defect coordinates to a 3D model by attaching reflective target balls at the start and end points and corners of the weld, scanning the target balls with a laser tracker to obtain physical coordinates, marking the corresponding point coordinates in the digital model, establishing weld point coordinates in the 3D model, and outputting defect point cloud coordinates after the monitoring equipment detects the defect point. After receiving the converted defect coordinates and attributes, it generates spatial anchor points at the corresponding positions in the 3D model, and renders visualization elements and generates warning markers according to the defect type.

[0080] A welding modeling management system, comprising:

[0081] The data acquisition module is used to collect welding data, including welding structure data and weld seam data. It also performs image acquisition and recognition of the surface morphology of the weld seam, stores the collected data, and establishes a database.

[0082] The 3D model creation module is used to create a 3D model of the welded structure based on the collected welded structure data using 3D modeling software. It uses a lightweight engine to support interaction and draws the weld seam in the 3D model.

[0083] The information entry module is used to enter the information of each weld seam based on the completed 3D model of the welded structure, assign a unique ID to each weld seam and use color to identify the weld seam status, and record the parameters of the weld seam welding process, so that the parameters and weld seam information are stored in the database and associated with the weld seam ID.

[0084] The platform construction module is used to build a unified platform and develop weld positioning and attribute query functions based on the 3D model. The platform inputs weld association information and images to realize information linkage function.

[0085] The welding process simulation module is used to plan and simulate welding processes based on weld information using welding process simulation software. It collects information during the simulation process, uses the simulation results to make predictions, and generates set process parameters.

[0086] The real-time monitoring module is used to weld the structure. During the welding process, it performs real-time monitoring, collects welding parameters and compares them with the set process parameters, and identifies and locates defects. After welding is completed, the welding data is stored in the database. The defect identification and location specifically includes:

[0087] During image acquisition, the defect recognition algorithm determines in real time whether there are defects, including cracks, pores and deformations. If a defect is found at the weld position, the platform maps the defect coordinates to the 3D model and generates an early warning mark, which flashes red on the platform. After clicking on the flashing position, the early warning mark is opened and the high-definition detection image of the defect and the quantitative data of the defect size are viewed.

[0088] The platform maps defect coordinates to a 3D model by attaching reflective target balls at the start and end points and corners of the weld, scanning the target balls with a laser tracker to obtain physical coordinates, marking the corresponding point coordinates in the digital model, establishing weld point coordinates in the 3D model, and outputting defect point cloud coordinates after the monitoring equipment detects the defect point. After receiving the converted defect coordinates and attributes, it generates spatial anchor points at the corresponding positions in the 3D model, and renders visualization elements and generates warning markers according to the defect type.

[0089] The quality inspection module is used to inspect the quality of the completed welds and record the inspection results into the platform. It also links the results to the 3D weld model and related information to determine whether the welds meet the requirements. The data acquisition module connects to the 3D model creation module via data transmission. The information entry module connects to the 3D model creation module via data interaction. The platform construction module connects to the 3D model creation module, information entry module, and data acquisition module via data interaction. The welding process simulation module connects to the platform construction module via data interaction. The real-time monitoring module connects to the platform construction module and the 3D model creation module via data interaction. Finally, the quality inspection module connects to the platform construction module via data interaction.

[0090] The working principle of this invention is as follows: Welding data is collected, images of the weld surface morphology are acquired and recognized, the collected data is stored, and a database is established. A 3D model of the welded structure is created using 3D modeling software, with a lightweight engine supporting interaction. The welds are drawn in the 3D model. Based on the established 3D model, information for each weld is entered, each weld is assigned a unique ID, and its status is indicated by color. Welding process parameters are recorded, and these parameters and weld information are stored in the database and associated with the weld ID. A unified platform is constructed, and weld positioning and attributes are developed based on the 3D model. The platform features a query function, allowing users to input weld-related information and images to achieve information linkage. Based on weld information, welding process simulation software is used for welding process planning and simulation. Information from the simulation process is collected, and the simulation results are used for prediction to generate set process parameters. Welding is then performed on the structure. During the welding process, real-time monitoring is conducted, and welding parameters are compared with the set process parameters, as well as defect identification and location. After welding is completed, the welding data is stored in the database. The completed weld is then subjected to quality inspection, and the inspection results are entered into the platform and linked with the weld 3D model and related information. Based on the inspection results, it is determined whether the requirements are met.

[0091] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A welding model-based management method, characterized in that, The specific steps include the following: Step 1: Collect welding data, including welding structure data and weld seam data. Also, collect images of the surface morphology of the weld seam, identify the images, store the collected data, and establish a database. Step 2: Based on the collected welding structure data, create a 3D model of the welding structure using 3D modeling software, use a lightweight engine to support interaction, and draw the weld seam in the 3D model; Step 3: Based on the completed 3D model of the welded structure, input the information of each weld, assign a unique ID to each weld, use color to identify the weld status, and record the parameters of the weld welding process. Store the parameters and weld information in the database and associate them with the weld ID. Step 4: Build a unified platform and develop weld positioning and attribute query functions based on the 3D model. The platform will input weld association information and images to realize information linkage function. Step 5: Based on weld information, use welding process simulation software to plan and simulate the welding process, collect information during the simulation process, and use the simulation results to make predictions and generate set process parameters; Step Six: Weld the structure. During the welding process, real-time monitoring is performed, and welding parameters are compared with the set process parameters, as well as defect identification and location. After welding is completed, the welding data is stored in the database. The defect identification and location specifically includes: During image acquisition, the defect recognition algorithm determines in real time whether there are defects, including cracks, pores and deformations. If a defect is found at the weld position, the platform maps the defect coordinates to the 3D model and generates an early warning mark, which flashes red on the platform. After clicking on the flashing position, the early warning mark is opened and the high-definition detection image of the defect and the quantitative data of the defect size are viewed. The platform maps defect coordinates to a 3D model by attaching reflective target balls at the start and end points and corners of the weld, scanning the target balls with a laser tracker to obtain physical coordinates, marking the corresponding point coordinates in the digital model, establishing weld point coordinates in the 3D model, and outputting defect point cloud coordinates after the monitoring equipment detects the defect point. After receiving the converted defect coordinates and attributes, it generates spatial anchor points at the corresponding positions in the 3D model, and renders visualization elements and generates warning markers according to the defect type. Step 7: Conduct quality inspection on the completed weld, record the inspection results in the platform, and link them with the weld 3D model and related information. Based on the inspection results, determine whether the requirements are met.

2. The welding model-based management method according to claim 1, characterized in that, The welding structure data in step one includes shape, size and assembly relationship, and the weld data includes weld shape, weld location and weld size.

3. The welding model-based management method according to claim 1, characterized in that, The acquisition and identification of the surface morphology image of the weld in step one specifically includes: High-definition camera equipment and laser scanning equipment are used to acquire images of the initial surface appearance of the weld. The defect identification algorithm identifies and determines whether there are defects in the initial appearance of the weld in real time. If defects are found, a defect mark is generated and the type and size of the defect mark are recorded. Then, when the 3D model is built, a corresponding warning mark is made on the 3D model, and a processing suggestion for the defect location is generated so that a matching welding method can be adopted during the welding process.

4. The welding model-based management method according to claim 1, characterized in that, The lightweight engine in step two supports interaction in the following specific ways: it supports weld geometry feature annotation and process attribute binding. The weld geometry features include bevel type, length and spatial coordinates. The process attribute binding includes welding method, filler material and preheating temperature.

5. The welding model-based management method according to claim 1, characterized in that, The weld information in step three includes weld type, weld grade, welding method, and welding materials. The weld type includes butt weld, fillet weld, and fillet weld. The welding method includes manual arc welding, gas shielded welding, submerged arc welding, and welding robot. The welding materials include welding rods, welding wire, and flux. The weld process parameters in step three include welding current, voltage, and welding speed.

6. The welding model-based management method according to claim 1, characterized in that, The weld ID in step three includes the project number, region, component number, and weld serial number. The color-coded weld status in step three specifically uses three colors: green, orange, and red. Green represents qualified, orange represents one rework, and red represents two reworks. The weld-related information entered into the platform in step four includes weld drawings, welding procedure qualification reports, welding quality plans, inspection reports, and sampling information.

7. The welding model-based management method according to claim 1, characterized in that, Step four, which involves platform construction and development of weld seam location and attribute query functions, specifically includes: Front-end and back-end development. Front-end development includes designing the user interface, which provides a weld selection list, a 3D model display area, an attribute information display area, and a weld image display area. The weld selection list is categorized by weld type and location. The 3D model display area provides rotation, zoom, and pan buttons to view the model from different angles. Back-end development includes designing an interface for obtaining weld location information. It receives the weld ID from the front-end, queries the corresponding geometric information from the database and returns it. It supports queries based on weld ID or query conditions and returns the corresponding attribute data. It implements the query logic for weld location and attribute information, accurately retrieves data from the database according to the front-end request, and uses database connection pooling technology for access. Using a 3D model processing library, the processed 3D model is loaded into the front-end page and rendered. When a user selects a weld in the weld selection list, the front end requests the location information of the weld in the 3D model from the back end through the API interface. The back end queries the relevant information from the database based on the weld identifier and returns it to the front end. After receiving the location information, the front end highlights the weld in the 3D model. The highlighting is achieved by changing the color of the weld and adding a mark. After the weld is located, the front end automatically obtains the attribute information of the weld from the back end and displays it in the attribute information display area. The front end enters the query conditions in the attribute information display area. The query conditions include weld type and welding material. The front end sends the query conditions to the back end, which queries the database and returns the results to the front end for display.

8. The welding model-based management method according to claim 1, characterized in that, In step four, the weld seam positioning involves defining a coordinate range for each weld seam in the 3D model. When a user selects a weld seam, the corresponding coordinate range is obtained based on its ID, and the geometric elements within that range are searched in the 3D model.

9. The welding model-based management method according to claim 1, characterized in that, The welding process in step six is ​​monitored in real time, and the welding parameters are compared with the set process parameters. Specifically, this includes: During the welding process, sensors, monitoring equipment, and laser scanning equipment acquire real-time images of the welding process and collect parameters such as welding current, welding temperature, and welding speed, and then transmit the data in real time. The parameters collected in real time are compared and analyzed with the preset process parameters. If the parameters exceed the allowable range, an alarm is issued and abnormal information is displayed on the platform.

10. A welding model-based management system, characterized in that, A welding modeling management method as described in any one of claims 1-9, comprising: The data acquisition module is used to acquire welding data, including welding structure data and weld seam data. It also acquires images of the surface morphology of the weld seam, identifies the images, stores the acquired data, and establishes a database. A 3D model creation module is used to create a 3D model of the welded structure based on the collected welded structure data using 3D modeling software, support interaction using a lightweight engine, and draw the weld seam in the 3D model. The information entry module is used to enter the information of each weld seam based on the completed 3D model of the welded structure, assign a unique ID to each weld seam and use color to identify the weld seam status, and record the parameters of the weld seam welding process, so that the parameters and weld seam information are stored in the database and associated with the weld seam ID. The platform construction module is used to build a unified platform and develop weld seam positioning and attribute query functions based on the 3D model. The platform inputs weld seam association information and images to realize information linkage function. The welding process simulation module is used to plan and simulate welding processes based on weld information using welding process simulation software, collect information during the simulation process, and use the simulation results to make predictions and generate set process parameters. A real-time monitoring module is used to weld the structure. During the welding process, it performs real-time monitoring, collects welding parameters and compares them with set process parameters, and identifies and locates defects. After welding is completed, the welding data is stored in a database. The defect identification and location specifically includes: During image acquisition, the defect recognition algorithm determines in real time whether there are defects, including cracks, pores and deformations. If a defect is found at the weld position, the platform maps the defect coordinates to the 3D model and generates an early warning mark, which flashes red on the platform. After clicking on the flashing position, the early warning mark is opened and the high-definition detection image of the defect and the quantitative data of the defect size are viewed. The platform maps defect coordinates to a 3D model by attaching reflective target balls at the start and end points and corners of the weld, scanning the target balls with a laser tracker to obtain physical coordinates, marking the corresponding point coordinates in the digital model, establishing weld point coordinates in the 3D model, and outputting defect point cloud coordinates after the monitoring equipment detects the defect point. After receiving the converted defect coordinates and attributes, it generates spatial anchor points at the corresponding positions in the 3D model, and renders visualization elements and generates warning markers according to the defect type. The quality inspection module is used to inspect the quality of the weld after welding is completed, and to record the inspection results into the platform. It is also linked with the 3D model of the weld and related information. Based on the inspection results, it is determined whether the requirements are met.

Images