A 3D Visualization Modeling Method for Welds Based on Big Data Analysis

Abstract

This invention belongs to the field of welded pipe technology, specifically involving a 3D visualization modeling method for welds based on big data analysis. It includes the following steps: Step 1: Pipeline plane construction; Step 2: 3DMAX data modeling; Step 3: Unity3D production; Step 4: Determining and publishing to WEBGL; Step 5: 3D visualization analysis. The beneficial effects of this invention are: 1) Data display flexibility: Combining Linear Regression and PageRank algorithms for extraction, display, and analysis, all 3D models can serve as carriers for big data analysis, enabling data flow, display, and analysis through big data analysis within the 3D model. 2) 3D model scalability: This design facilitates the expansion of the platform model. New welds added during later service maintenance, including valve management added based on other requirements, can be plugged into the original model, reducing subsequent development costs.

Description

Technical Field

[0001] This invention belongs to the field of welded pipe technology, specifically relating to a three-dimensional visualization modeling method for welds based on big data analysis. Background Technology

[0002] Since the pipelines and welds of nuclear power plants are one of the important pressure boundaries of the power plant, the information on these pipelines and welds is currently scattered and distributed in different technical documents such as design documents, change documents, maintenance documents and in-service inspection documents. A unified visual management platform has not been formed, and there are problems such as low accuracy, timeliness, completeness, intuitiveness, convenience and efficiency in the collection, transmission, analysis and utilization of relevant information, which urgently need to be improved. Summary of the Invention

[0003] The purpose of this invention is to provide a 3D visualization modeling method for welds based on big data analysis. By querying data in the weld model, basic information can be obtained intuitively; by querying 720-degree panoramic images in different virtual rooms, the surrounding real environment of a weld can be viewed intuitively; based on big data services, querying any single weld in the 3D model can display basic weld data, work order data, status reports, etc., improving retrieval efficiency; through file queries, isometric diagrams, flowcharts, as-built drawings, and historical reports related to the weld data can be displayed.

[0004] The technical solution of this invention is as follows: A three-dimensional visualization modeling method for welds based on big data analysis, comprising the following steps:

[0005] Step 1: Pipeline plan construction;

[0006] Step 2: 3DMAX data modeling;

[0007] Step 3: Unity3D creation;

[0008] Step 4: Determine to publish WEBGL;

[0009] Step 5: 3D visualization analysis.

[0010] Step 1 involves using the collected weld-related data, including weld type attribute parameters, weld overhaul documents, and as-built materials, to convert the weld data into a graphical structure, where welds are represented as nodes and different relationships are represented as edges, thus constructing a network model of the weld data.

[0011] In step 1, for nodes connected by edges, corresponding weights are assigned according to their importance. The PageRank algorithm is used to calculate the importance score of each node. The process is iterated multiple times, and finally, based on the PageRank scores of the nodes, the important and critical welds in the weld data are determined.

[0012] In step 2, the pipeline obtained from the isometric drawing and the weld information obtained in step 1 are used to draw the model using 3DMAX. The pipe segments in 3DMAX are then spliced ​​together according to the drawings, and the node positions of each joint are reserved and named accordingly.

[0013] Step 3 involves creating relevant teaching experimental equipment and instruments through 3DMAX modeling, producing related animations, outputting files in FBX format, importing them into the Unity3D editor for integration and positioning, and finally entering interactive programming.

[0014] Step 5 includes:

[0015] Step 51: Design Direction;

[0016] Step 52: Design of the 3D display platform;

[0017] Step 53: Weld data modeling and design.

[0018] Step 51 includes weld big data interaction and query using GBASE8a storage, compression and intelligent indexing; the weld data, including related isometric diagrams, flowcharts, as-built drawings, historical reports, and on-site photos, are uniformly associated with the three-dimensional model points one-to-one, so that a single query condition can bring up all relevant information together.

[0019] Step 52 includes enabling management between all systems in the plant, implementing basic operations of the 3D model, and simultaneously meeting the needs of hiding and displaying objects; the types of pipes and welds are distinguished in the 3D design by using different colors and markings.

[0020] Step 53 involves using multiple data platforms in series to combine pipeline weld information data, integrating the SAPHAHA big data platform, document management system platform, and its own 3D virtual simulation data to construct the pipeline weld data model, then performing HANA modeling based on actual business, storing MySQL data tables, and establishing a weld visualization management system model table.

[0021] The beneficial effects of this invention are as follows:

[0022] 1) Data display flexibility

[0023] It combines the Linear Regression algorithm and the PageRank algorithm for extraction, display and analysis. All 3D models can be used as carriers for big data analysis, enabling data flow, display and analysis through the implementation of big data analysis in 3D models.

[0024] 2) 3D model extensibility

[0025] This design approach facilitates the expansion of the platform model. New welds added during later service maintenance, as well as valve management and other new requirements, can be plugged into the original model, reducing subsequent development costs. Attached Figure Description

[0026] Figure 1 A flowchart illustrating a three-dimensional visualization modeling method for welds based on big data analysis provided by this invention;

[0027] Figure 2 A schematic diagram illustrating the principles of 3D modeling;

[0028] Figure 3 This is a diagram illustrating the interaction of data and information. Detailed Implementation

[0029] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0030] A 3D visualization modeling method for welds based on big data analytics combines big data analysis with 3D visualization modeling to comprehensively analyze and display weld data. First, data queries are performed within the weld model to intuitively obtain basic information. Second, 720-degree panoramic views are queried in different virtual rooms, enabling a direct view of the surrounding environment of a specific weld. Simultaneously, based on big data services, queries can be performed on any single weld, displaying basic weld data, work order data, status reports, etc., within the 3D model to improve retrieval efficiency. Finally, file queries display isometric diagrams, flowcharts, as-built drawings, and historical reports related to the weld data, thus providing a comprehensive view of weld information and optimizing welding process management and quality control.

[0031] The entire system piping is divided into segments, and initial modeling is performed in 3DS MAX to enable subsequent pipe splicing and pluggable applications. In 3DS MAX, based on the piping drawings, the main types appearing in the drawings are categorized by different colors, and modeling is performed using the XYZ axes to achieve point-to-point and segment-to-segment pluggable functionality.

[0032] Standardized design involves two steps: First, establishing comprehensive pipe classification systems, including shapes and names, organizing them by major categories, and creating unified standards. Second, classifying pipes according to design drawings, first creating individual pipe models, and then assembling them.

[0033] Once the models are assembled, they form a unified whole. During use, the connecting parts are automatically closed or hidden. During maintenance, any pipe can be clicked to perform a replacement operation. When adding content to the same pipe segment, additions can be made based on the original. When adding a new pipe segment, the model will be stretched or the incorrectly positioned pipe segment will be "pulled out". Once the new model is completed, it can be directly "inserted".

[0034] Based on the Linear Regression and PageRank algorithms, 3D models are connected to big data platforms to extract data. All 3D models can serve as carriers for big data analysis, enabling data circulation, display, and analysis through the implementation of big data analysis within the 3D models.

[0035] 1. Utilize the Linear Regression algorithm to extract various data objects related to welds from the big data platform, automatically alert to deviation data, and automatically supplement new weld data to achieve data extraction and circulation.

[0036] 2. The PageRank algorithm is used to design a correlation map of weld seam data information, and the weld seam data in the big data platform is displayed and analyzed in a three-dimensional model.

[0037] 3. By using a plug-and-play modeling method, it is possible to freely add, uninstall, and update its plugins, modules, extensions, etc. Pipelines that can be created by multiple people can be assembled by one person, which greatly improves efficiency and reduces maintenance costs.

[0038] A method for 3D visualization modeling of welds based on big data analysis includes the following steps:

[0039] Step 1: Pipeline Plan Construction

[0040] Pipeline isometric drawings, also known as pipe segment drawings or open-plan drawings, are a type of three-dimensional drawing used for industrial pipelines and equipment. This drawing fully depicts the specific locations of pipe fittings, welds, valves, etc., in a planar design. In the initial design phase, before transferring to 3D Max for modeling, isometric drawings of pipelines are prepared, and the pipeline routing, upper and lower structure of the plant building are analyzed, converting the planar data (including floor height data, spacing, and dimensions at a 1:1 scale). Specifically, this includes the following:

[0041] Weld data transformation utilizes collected weld-related data, including various weld attribute parameters, overhaul documents, and as-built material information. The weld data is converted into a graphical structure, where welds are represented as nodes, and different relationships (such as adjacent welds, similarity, etc.) are represented as edges, constructing a network model of the weld data. Furthermore, nodes connected by edges are assigned corresponding weights based on their importance. The PageRank algorithm is used to calculate the importance score of each node, iterating multiple times. Finally, based on the PageRank scores of the nodes, important and critical welds in the weld data are identified, thereby aiding in the adjustment and optimization of welding processes. The results are validated using the LinearRegression algorithm.

[0042] Step 2: 3DMAX Data Modeling

[0043] Based on the isometric drawing of the pipeline and the weld information obtained in step 1, a model is drawn using 3DMAX. The pipe segments from 3DMAX are then assembled according to the drawings, with each joint having pre-defined node positions and distinct names. This ensures the replaceability, controllability, and flexibility of the pipe containing each weld.

[0044] Create pipe sections and paths

[0045] Use the spline tool to draw the cross-section (such as a circle) and path of the pipe.

[0046] Create pipes using lofting.

[0047] Edit Polygon Modifier

[0048] Creating Pipe Interface Details

[0049] Use the "Insert" command to add details to the pipe joint, such as the pipe thickness. Use the "Extrude" command again, this time with a negative value, to extrude inwards to create the inner edge of the pipe.

[0050] Making hard edges

[0051] Remove end face and smooth

[0052] Special pipe modeling

[0053] For curved pipes, you can use the "Bend" modifier to adjust the curvature of the pipe. In the "Modify" panel, select the "Bend" modifier and set the bending angle and axis to achieve the desired bending effect.

[0054] These steps allow you to create a detailed and realistic pipe model in 3ds Max and correctly connect all the individual pipe segments to form a continuous pipe network.

[0055] Step 3: Unity3D Production

[0056] Using 3DMAX modeling, relevant teaching experimental equipment and instruments were created, and related animations were produced. These were then output as FBX files, imported into the Unity3D editor for integration and positioning, and finally entered into interactive programming. Through in-depth analysis and research of the Unity3D engine's architecture and working principles, virtual simulation of the welded seam plant structure was achieved, along with spatial mimicry design, enabling 3D display and animation preview of the plant. Secondly, the "communication" mechanism between Unity3D and page information data was analyzed and studied, allowing for data and spatial integration.

[0057] Step 4: Determine to publish WEBGL

[0058] The introduction of WebGL allows models to be embedded into web pages for manipulation, greatly increasing user interactivity. By publishing WebGL, Vue utilizes components for page rendering.

[0059] In the BuildSettings folder, select the scenario you want to publish.

[0060] Click Player's Settings to modify the relevant configurations of the package, such as name, resolution, and other parameters.

[0061] Click Build, select the location to publish to, and click Select Folder to start packaging and publishing.

[0062] After publishing, a WEBGL file is obtained, which is then loaded and rendered using WEB components.

[0063] Step 5: 3D Visualization Analysis

[0064] In accordance with the management requirements for weld seams during the Qinshan Nuclear Power Plant's maintenance operations, we improved the refined management of weld seam data. Through data analysis and on-site inspections, we identified existing problems and developed a multi-functional digital management platform for weld seams. This single platform enables the creation of a virtual 3D visualization model and establishes relationships between attribute objects, reducing on-site troubleshooting work, thereby lowering radiation dose, improving overall work efficiency, and truly achieving a new management model with less contact and fewer queries.

[0065] Step 51: Design Direction

[0066] The big data interaction and query of weld seams uses GBASE8a storage, compression and intelligent indexing, plus full-text indexing technology, to work together for retrieval.

[0067] Basic concept: The method of segmenting a sequence of values ​​into individual words is called word segmentation. Natural word segmentation uses certain identifiers as natural delimiters between words. The "3" in ternary word segmentation means that when there are more than three consecutive English letters or numbers, they are segmented into groups of three. If the original data cannot be segmented normally into three words, it is directly stored as a key in the index. Chinese text is usually segmented using single characters. For English letters and numbers, both natural word segmentation and ternary word segmentation are used throughout the text.

[0068] Detailed design and analysis of weld data

[0069] Refined data management links weld data, including related isometric diagrams, flowcharts, as-built drawings, historical reports, and on-site photos, to specific points (welds) in the 3D model on a one-to-one basis. This allows a single query to retrieve all relevant data, minimizing search costs. Simultaneously, it maximizes integration with the virtual 3D model and enables data interaction, facilitating bidirectional "transmission" between the data and the 3D model.

[0070] Step 52: Design of 3D Display Platform

[0071] The system enables management across all systems in the plant, allows basic operations on the 3D model, and supports the hiding and display of objects to meet different needs. Pipe types and weld types are distinguished by different colors and labels in the 3D design. When designing the 3D weld model, basic operations such as zooming in and out and stretching are retained, and the model can be viewed based on different object attribute fields during querying, achieving the hiding and display function.

[0072] Step 53: Weld Data Modeling Design

[0073] The pipeline weld information data is currently constructed by combining multiple data platforms in series, integrating the SAPHAHA big data platform, the document management system platform, and its own 3D virtual simulation data to build the pipeline weld data model.

[0074] The 3D simulation data modeling relies on the MAX technology field. At the same time, HANA modeling is carried out according to actual business, and MySQL data tables are stored to establish a weld seam visualization management system model table.

[0075] Due to the large volume of weld information data from nuclear power plants, the diverse sources of information, and the complex formats—including EAM production, ECM documents, management support, and ERP systems—incomplete information processing and summarization resulted in instances where specific attribute data for weld objects was entered incorrectly, as well as other data entry errors.

[0076] Data related to weld seams in nuclear power plants is scattered. Data from historical changes and design modifications in welding records is easily lost or incomplete, with old and new data mixed together, making verification and retrieval extremely difficult. The disorganized nature of nuclear power plant weld seam data, due to the lack of standardized centralized management, leads to significant inaccuracies or discrepancies with actual site conditions. Historical data verification of nuclear power plant weld seams is difficult; data verification and confirmation are challenging; and there is a lack of technical means for analysis and judgment.

[0077] This invention utilizes a big data platform to achieve comprehensive, full-process, and all-scenario control over weld seam data sourcing, modeling, and display. By employing a data extraction tool, data from various systems is extracted into the big data platform, and unified modeling is achieved using the HANA tool. Using the big data platform as the data source for the weld seam 3D model solves the problem of data fragmentation across systems, enabling the querying of various data based on system layout within 3D space. Furthermore, the data and 3D model are developed into a universal version, facilitating use by other business stakeholders. An approval process is integrated to check and correct erroneous attribute information in real time. The data model standardizes schema field attributes, and data changes can be monitored in real time through model queries. The platform sets data permissions, categorized into different levels: 1) query only, 2) review / approval allowed, and 3) modification allowed. Finally, the model, through modification, submission, display, and download functions, achieves data collection, verification, and presentation. Example

[0078] This design was used in the Qinshan Nuclear Power Plant's weld seam visualization management system, employing a completely realistic 3D power plant model. Based on the 3D design model (AECL) of the Fangjiashan section of the Qinshan Nuclear Power Plant, a 3D digital model of the power plant was created using Unity3D modeling. This model was created in a 1:1 scale, ensuring consistency with the actual physical structure and accurately reproducing the actual conditions of the power plant.

[0079] (1) Obtain all the drawings required for the 25 systems of Qinshan Nuclear Power Plant, including RIS, RCP, RCV, VVP, TEG, etc., including complete flowcharts and isometric drawings (including main pipelines and branch pipes).

[0080] (2) Perform data cleaning, and convert the drawing data into tabular data according to the data specifications and weld attribute data requirements of the model. (This includes recording all basic information contained in the drawings, such as room number, pipe section number, flowchart and isometric drawing file number, etc., and dividing the data into pipe-data object-room number diagrams according to every two weld nodes.)

[0081] (3) 3DMAX uses the weld drawing data information to perform detailed modeling, dividing the entire system pipeline into segments. In 3DMAX, according to the pipeline drawing, the major types appearing in the drawing are classified by different colors. At the same time, modeling is performed using the XYZ axis to achieve point-to-point and pipe segment-to-pipe segment insertion and removal.

[0082] (4) The packaged model files are then used for animation rendering in Unity3D, and finally a WEBGL package is generated. (WEBGL is memory-optimized and its size is compressed according to hardware requirements.)

[0083] (5) Data association design: The weld seam data and WEBGL model data are associated using the HAHA big data database and integrated tabular data. Image files are used to perform path queries within a small range and data attachment using data node names.

[0084] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications to the technical solutions described in the foregoing embodiments, or equivalent substitutions for some or all of the technical features, do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for weld 3D visualization modeling based on big data analysis, characterized in that, It comprises the following steps: Step 1: pipeline plane erection; The step 1 includes converting the weld data into a graph structure by using the collected weld-related data, including various types of attribute parameters of the weld, the weld overhaul file, and the completion material, wherein the weld is represented as a node, and different relationships are represented as edges, and a network model of the weld data is constructed; in the step 1, for the nodes connected by the edges, corresponding weights are given according to the importance thereof, the importance score of each node is calculated by using the PageRank algorithm, and multiple iterations are performed, and finally the important and key welds in the weld data are determined according to the PageRank score of the nodes; Step 2: 3DMAX data modeling; The step 2 draws a model by 3DMAX according to the pipeline obtained from the pipeline isometric graph and the weld information obtained in the step 1, splices the max pipe sections according to the drawing, and reserves node positions and distinguishes the naming for each opening; Step 3: UNITY3D production; The step 3 creates relevant teaching experimental equipment and instruments by 3DMAX modeling, produces relevant animations, then outputs a file in the FBX format, imports the file into a unity3d editor for integration and positioning, and finally enters interactive programming; Step 4: determining to publish WEBGL; Step 5: three-dimensional visualization analysis.

2. The weld 3D visualization modeling method based on big data analysis of claim 1, wherein, The step 5 comprises: Step 51: design direction; Step 52: three-dimensional display platform design; Step 53: weld data modeling design.

3. The weld 3D visualization modeling method based on big data analysis of claim 2, wherein: The step 51 includes that the weld big data interaction and query adopt GBASE8a storage, compression and intelligent indexing; the weld data including relevant isometric graphs, flowcharts, completion drawings, historical reports and field photos are associated with three-dimensional model points one by one, so that all related materials are brought out together under one query condition.

4. The weld 3D visualization modeling method based on big data analysis of claim 2, wherein: The step 52 includes realizing management between all systems in a plant, realizing basic operations of the three-dimensional model, and simultaneously satisfying object hiding and display of different requirements; the type of the pipeline and the type of the weld are distinguished according to different colors and marks in the three-dimensional design.

5. The weld 3D visualization modeling method based on big data analysis of claim 2, wherein: The step 53 includes that the pipeline weld information data is composed of a plurality of data platforms in series, integrates a SAPHAHA big data platform, a document management system platform and three-dimensional virtual simulation data of itself, constructs a pipeline weld data model, then performs HANA modeling according to actual business, stores a MYSQL data table, and establishes a weld visualization management system model table.

Images