Intelligent verification and abnormal positioning method for three-dimensional topological relationship of power pipeline

CN122594767APending Publication Date: 2026-08-18SHANGHAI MINGCHUAN SURVEYING & MAPPING TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202610615569.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-05-07
Publication Date
2026-08-18

AI Technical Summary

Technical Problem

当前电力地下管线拓扑关系检查主要存在以下突出问题:其一,主流检查方式仍以人工目视检查为主,单条10公里管线的人工检查平均耗时达40工时,检查效率低下,且不同检查人员对规范的理解存在差异,检查标准不统一,隐蔽性拓扑问题发现率不足30%;其二,现有拓扑检查技术多聚焦于二维平面层面,仅能验证管线平面位置的连通性与交叉关系,无法反映地下管线三维空间的竖向高程、埋深、净距、交叉角度等关键信息,难以识别非开挖施工管线的三维走向失真、空间碰撞、竖向净距不足等隐蔽异常,而此类异常正是引发电网施工事故的核心诱因;其三,现有技术未结合电力行业专属规范与上海区域地质、城市管理特点构建针对性的拓扑规则体系,无法适配不同电压等级、不同敷设方式、不同区域的电力管线拓扑合规性校验需求,检查结果的实用性与合规性不足

Benefits of technology

[0008]本申请实施例的电力管线三维拓扑关系智能验证与异常定位方法,通过构建全流程的三维拓扑验证技术体系,从根本上解决了传统二维拓扑检查无法适配地下管线三维空间特性的技术痛点,实现了四大核心技术突破:其一,构建了适配上海城市特点的电力管线时空基准统一体系,实现了多源异构管线数据的标准化整合,解决了不同来源、不同坐标系、不同高程基准的管线数据融合难题,为三维拓扑验证奠定了精准的数据基础;其二,建立了电力管线领域本体库与语义关联规则体系,实现了管线空间要素与业务属性的语义级对齐,解决了传统检查中空间数据与属性数据脱节、业务规则适配性差的问题;其三,研发了三级体系的三维拓扑智能验证技术,融合图论分析、三维空间碰撞检测、规则引擎动态推理技术,实现了从网络连通性、三维几何拓扑到行业业务合规性的全维度校验,覆盖了地下电力管线全场景的拓扑异常类型;其四,构建了异常精准定位与分级体系,实现了拓扑异常的毫米级三维坐标定位与风险等级自动划分,结合三维可视化技术,为异常整改提供了精准、直观的技术支撑。

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122594767A_ABST
    Figure CN122594767A_ABST
Patent Text Reader

Abstract

This application discloses a method for intelligent verification and anomaly localization of three-dimensional topological relationships of power pipelines, including a three-dimensional data preprocessing module, a power pipeline ontology semantic modeling module, a three-dimensional topological relationship intelligent verification module, an anomaly accurate localization and classification module, and a visualization and report generation module. The three-dimensional data preprocessing module completes the standardized integration of multi-source power pipeline spatial data and the unification of spatiotemporal benchmarks. The power pipeline ontology semantic modeling module constructs an ontology library and semantic association rule system in the power pipeline domain based on the standardized three-dimensional pipeline dataset generated by preprocessing. This method breaks through the technical limitations of traditional two-dimensional topology inspection and realizes fully automatic, high-precision intelligent verification of three-dimensional spatial topological relationships of underground power pipelines. It has a high anomaly recognition rate, excellent computational efficiency, and strong environmental adaptability, and can be widely used in scenarios such as quality inspection of power pipeline tracking and measurement results, governance of existing pipeline data, and investigation of pipeline hidden dangers before construction.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of underground pipeline mapping and geographic information system technology, and in particular to a method for intelligent verification of three-dimensional topological relationships and anomaly location of power pipelines. Background Technology

[0002] With the deepening of Shanghai's "14th Five-Year Plan" for smart city construction, the scale of the power grid, the lifeline of the city's energy, continues to expand. By the end of 2024, the total length of power lines in Jinshan, Qingpu, Songjiang, Jiading, Minhang, and Xuhui districts of Shanghai had exceeded 5,800 kilometers, with underground cables accounting for more than 75%, forming a massive, complex, and spatially interconnected urban underground power network system. In the context of "digital twin city" construction, the accuracy and completeness of basic power line data, especially the compliance of three-dimensional spatial topology, directly affects the quality of the entire lifecycle management of power grid planning, construction, and operation.

[0003] Quality inspection of underground power line tracking and measurement results is a core step in ensuring the accuracy and reliability of pipeline data, and topological relationship verification is the core content of the quality inspection work. The current inspection of underground power pipeline topology faces the following prominent problems: First, the mainstream inspection method is still mainly manual visual inspection. The average time for manual inspection of a single 10-kilometer pipeline is 40 man-hours, resulting in low inspection efficiency. Furthermore, different inspectors have different understandings of the specifications, leading to inconsistent inspection standards and a detection rate of less than 30% for hidden topology problems. Second, existing topology inspection technologies mostly focus on the two-dimensional plane level, which can only verify the connectivity and intersection relationships of pipelines in the plane. It cannot reflect key information such as the vertical elevation, burial depth, clearance, and intersection angle of underground pipelines in three-dimensional space. It is difficult to identify hidden anomalies such as three-dimensional orientation distortion, spatial collision, and insufficient vertical clearance of trenchless construction pipelines. These anomalies are the core causes of power grid construction accidents. Third, existing technologies have not combined the power industry's specific specifications with the geological and urban management characteristics of Shanghai to build a targeted topology rule system. They cannot adapt to the topology compliance verification needs of power pipelines of different voltage levels, different laying methods, and different regions, resulting in insufficient practicality and compliance of the inspection results.

[0004] Industry statistics from 2024 show that power grid construction accidents caused by quality issues such as errors in topological relationships and spatial location deviations in surveying data accounted for 18.7% of all accidents, resulting in direct economic losses exceeding ten million yuan. The National Energy Administration's "Action Plan for Digital Transformation of the Power Industry (2023-2025)" clearly requires that "by 2025, the digital archiving rate of important power facilities should reach 100%, and the data quality qualification rate should not be less than 98%." Shanghai has also issued the "Implementation Opinions on the Tracking and Measurement of Underground Pipelines in Shanghai," which clearly requires that underground pipeline projects must adhere to the principle of "mandatory measurement for all construction projects, and shared resources," ensuring the accuracy of pipeline measurement results. Therefore, developing a method that can achieve fully automated, high-precision intelligent verification of the three-dimensional topological relationships of underground power pipelines and accurate anomaly location has become a pressing technical challenge for the industry. Summary of the Invention

[0005] This application aims to at least partially address one of the technical problems in the related art.

[0006] Therefore, the purpose of this application is to provide an intelligent verification and anomaly location method for the three-dimensional topology relationship of power pipelines, breaking through the technical limitations of traditional two-dimensional topology inspection, constructing a three-dimensional topology verification system adapted to power industry standards and Shanghai regional management requirements, realizing fully automatic, standardized, and high-precision intelligent verification and accurate anomaly location of the three-dimensional spatial topology relationship of underground power pipelines, improving inspection efficiency by 8-10 times compared to traditional manual methods, and achieving an anomaly identification rate of over 95%, providing reliable technical support for the safe operation and maintenance and refined management of urban underground power pipeline networks.

[0007] To achieve the above objectives, the first aspect of this application proposes an intelligent verification and anomaly localization method for three-dimensional topological relationships of power pipelines, including a three-dimensional data preprocessing module, a power pipeline ontology semantic modeling module, a three-dimensional topological relationship intelligent verification module, an anomaly accurate localization and classification module, and a visualization and report generation module. The three-dimensional data preprocessing module completes the standardized integration and spatiotemporal benchmark unification of multi-source power pipeline spatial data. The power pipeline ontology semantic modeling module constructs an ontology library and semantic association rule system in the power pipeline domain based on the standardized three-dimensional pipeline dataset generated by preprocessing. The three-dimensional topological relationship intelligent verification module realizes the compliance verification of the topological relationships of the pipeline network in all dimensions based on the ontology library and three-dimensional spatial analysis technology. The anomaly accurate localization and classification module performs three-dimensional spatial coordinate localization and risk level classification of the anomaly results identified by topology verification. The visualization and report generation module completes the three-dimensional visualization display of the anomaly results and the automatic output of standardized quality inspection reports.

[0008] The intelligent verification and anomaly location method for three-dimensional topology relationships of power pipelines in this application, by constructing a complete three-dimensional topology verification technology system, fundamentally solves the technical pain point that traditional two-dimensional topology inspection cannot adapt to the three-dimensional spatial characteristics of underground pipelines, achieving four core technological breakthroughs: First, it constructs a unified spatiotemporal reference system for power pipelines adapted to the characteristics of Shanghai, realizing the standardized integration of multi-source heterogeneous pipeline data, solving the problem of fusion of pipeline data from different sources, different coordinate systems, and different elevation references, and laying a precise data foundation for three-dimensional topology verification; Second, it establishes an ontology library and semantic association rule system for the power pipeline field, realizing the spatial relationship verification of pipelines. The semantic alignment of spatial elements and business attributes solves the problems of spatial data and attribute data being disconnected and business rule adaptability in traditional inspections. Third, a three-level system of three-dimensional topology intelligent verification technology has been developed, which integrates graph theory analysis, three-dimensional spatial collision detection, and rule engine dynamic reasoning technology to achieve full-dimensional verification from network connectivity and three-dimensional geometric topology to industry business compliance, covering topology anomaly types in all scenarios of underground power pipelines. Fourth, an anomaly precise location and classification system has been constructed, which realizes millimeter-level three-dimensional coordinate location of topology anomalies and automatic risk level classification. Combined with three-dimensional visualization technology, it provides accurate and intuitive technical support for anomaly rectification.

[0009] The core beneficial effects of this method are reflected in the following aspects: Inspection efficiency is greatly improved, reducing the time for topology inspection of a single 10-kilometer pipeline from 40 man-hours in the traditional manual method to less than 4 man-hours, improving inspection efficiency by 8-10 times, significantly reducing manual inspection costs, and adapting to the batch quality inspection needs of large-scale urban power pipeline data. The inspection accuracy and recognition rate are significantly improved. The recognition rate of hidden topological anomalies in the three-dimensional space of underground pipelines has increased from less than 30% by traditional manual methods to more than 95%. The plane positioning accuracy reaches ±0.05m and the elevation positioning accuracy reaches ±0.02m. It can effectively intercept erroneous data from flowing into the production system, reducing construction risks, downtime accidents and subsequent rework costs caused by data errors. The inspection standards are highly unified. The built-in three-level topology rule system strictly follows the "Underground Pipeline Surveying Standard" DG / TJ08-85, the "Power Engineering Cable Design Standard" GB50217-2018, and the Shanghai Municipal Underground Pipeline Tracking Measurement Management Requirements, which realizes the solidification and unification of inspection standards and eliminates the subjective differences of manual inspection. With strong scenario adaptability, it can be widely used in various scenarios such as quality inspection of tracking and measurement results of newly built power pipelines, data governance of existing pipelines, daily operation and maintenance data update review, and pipeline hidden danger investigation before construction. At the same time, through the flexible configuration of the rule base, it can adapt to the topology inspection needs of other urban underground pipelines such as gas, water supply, and communication, and has the potential for cross-industry promotion. With outstanding management support capabilities, it can automatically generate standardized quality inspection reports, realize closed-loop management of abnormal issues throughout the entire life cycle, provide high-quality three-dimensional pipeline data support for power grid planning, emergency repair, and asset life cycle management, and help the refined management of urban underground space and the construction of digital twin cities.

[0010] In one embodiment of this application, the multi-source data access and format conversion step of the three-dimensional data preprocessing module supports access to power pipeline tracking and measurement results data, design drawing data, laser point cloud data, and GIS vector data. A dedicated data parser performs lossless conversion between DWG, SHPP, LAS, and XLSX formats, extracting the three-dimensional spatial coordinates and attribute information of pipeline segments, pipeline feature points, manholes, and ancillary facilities. The spatiotemporal benchmark unification step constructs a three-level benchmark conversion system. The plane benchmark adopts a seven-parameter conversion model between the Shanghai local coordinate system and the CGCS2000 national geodetic coordinate system, with conversion accuracy controlled within ±0.05m. The elevation benchmark constructs a grid correction model based on the Wusong elevation and the 1985 national elevation benchmark, with elevation conversion residuals controlled within ±0.02m. The data cleaning and topology normalization step completes dirty data repair and topology normalization processing, and constructs an R* tree spatial index to improve spatial analysis efficiency.

[0011] In one embodiment of this application, the power pipeline ontology library construction step of the power pipeline ontology semantic modeling module is based on the Gruber ontology construction method to establish a power pipeline ontology library containing 5 major categories and 32 subcategories. The topological relationships between pipeline elements are described using the OWL language, and 87 attribute association rules are defined. The semantic parsing and entity mapping step uses the BiLSTM-CRF semantic parsing engine to complete entity recognition and element semantic mapping of unstructured text. The semantic conflict detection and resolution step establishes a conflict detection library with 78 core rules and completes hierarchical resolution through a four-level conflict handling strategy.

[0012] In one embodiment of this application, the three-dimensional topology rule base construction step of the intelligent verification module for three-dimensional topology relationships establishes a three-level topology rule system including basic topology rules, regional characteristic rules, and professional pipeline rules; the network connectivity topology verification step constructs a pipeline network adjacency matrix based on graph theory to complete the identification and verification of isolated pipelines, suspended nodes, and illegal closed loops; the three-dimensional spatial geometric topology verification step adopts an improved Douglas-Peucker algorithm, curvature analysis algorithm, and three-dimensional OBB bounding box collision detection algorithm to complete the identification of pipeline routing distortion, abrupt change points, spatial collisions, and insufficient clearance; the business rule compliance verification step is based on the Drools rule engine to complete the standard compliance verification of parameters such as pipeline burial depth, spacing, and intersection angle.

[0013] In one embodiment of this application, the abnormal three-dimensional coordinate positioning step of the abnormal precise positioning and classification module completes the three-dimensional precise positioning of the abnormal location through the "problem cube" marking system, with the positioning accuracy controlled within ±0.05m; the abnormal feature extraction and classification step divides the abnormal into four categories: connectivity abnormality, geometric topology abnormality, spatial collision abnormality, and business compliance abnormality; the risk level classification step constructs a multi-dimensional weighted risk assessment model to classify the abnormal into three levels: major risk, general risk, and minor risk; the automatic generation step of rectification suggestions automatically generates rectification measures that match the requirements of the specifications based on the rule knowledge graph.

[0014] In one embodiment of this application, the 3D lightweight rendering step of the visualization and report generation module, based on the Cesium engine and 3DTiles layered loading technology, enables smooth browsing of the pipeline's 3D scene; the anomaly visualization interaction step enables anomaly point highlighting, spatial analysis, and quick querying; the automatic quality inspection report generation step supports custom templates and automatically generates quality inspection reports that comply with national standards and Shanghai municipal local management requirements, achieving closed-loop management of anomaly rectification. Attached Figure Description

[0015] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is an overall flowchart of a method for intelligent verification and anomaly localization of three-dimensional topology of power pipelines according to an embodiment of this application; Figure 2 This is a detailed flowchart illustrating the intelligent verification and anomaly localization of three-dimensional topological relationships according to an embodiment of this application. Detailed Implementation

[0016] Embodiments of this application are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application. Rather, embodiments of this application include all variations, modifications, and equivalents falling within the spirit and scope of the appended claims.

[0017] The following describes, with reference to the accompanying drawings, the intelligent verification and anomaly location method for three-dimensional topology relationships of power pipelines according to an embodiment of this application.

[0018] The intelligent verification and anomaly location method for three-dimensional topological relationships of power pipelines provided in this application embodiment can be applied to scenarios such as quality inspection of underground power pipeline tracking and measurement results in Shanghai, management of existing pipeline data, and investigation of pipeline hidden dangers before construction. It strictly follows local and industry standards such as the "Implementation Opinions on the Tracking and Measurement of Underground Pipelines in Shanghai" and the "Underground Pipeline Surveying Standard" DG / TJ08-85, providing accurate technical support for the safe operation and maintenance of urban underground power networks.

[0019] like Figure 1 and Figure 2 As shown in the figure, the intelligent verification and anomaly localization method for three-dimensional topology relationships of power pipelines in this application includes a three-dimensional data preprocessing module, a power pipeline ontology semantic modeling module, a three-dimensional topology relationship intelligent verification module, an anomaly accurate localization and classification module, and a visualization and report generation module.

[0020] Specifically, the core objective of the 3D data preprocessing module is to standardize and integrate multi-source power pipeline spatial data and unify spatiotemporal references, laying a precise and standardized data foundation for subsequent 3D topology verification. This module sequentially executes the steps of multi-source data access and format conversion, spatiotemporal reference unification, and data cleaning and topology normalization. In actual execution, multi-source data access and format conversion requires accessing power pipeline tracking and measurement results data, CAD design drawings, 3D laser point cloud data, and GIS vector pipeline data from the target area. A dedicated data parser performs lossless conversion between DWG, SHPP, LAS, and XLSX formats, accurately extracting the 3D coordinates of the start and end points of pipeline segments, pipe diameter, material, voltage level, 3D coordinates of pipeline feature points (tees, elbows, joints), and 3D boundary and attribute information of ancillary facilities such as cable wells and switch stations; spatiotemporal reference unification... A three-level datum transformation system needs to be constructed. The seven-parameter transformation between the Shanghai local coordinate system and the CGCS2000 national geodetic coordinate system will be completed through 12 high-level control points. The transformation between the Wusong elevation and the 1985 national elevation datum will be completed through a grid correction model constructed from 156 leveling points. At the same time, a UTC time-scale system will be established to unify the data time base. Data cleaning and topology normalization need to complete the automatic repair of dirty data, elimination of pseudo nodes, deletion of redundant vertices, and construction of an R* tree spatial index to improve spatial query efficiency by more than 40%.

[0021] Furthermore, the power pipeline ontology semantic modeling module constructs a power pipeline domain ontology library and semantic association rule system based on the preprocessed standardized 3D pipeline dataset, realizing semantic-level fusion of pipeline spatial elements and business attributes. This module sequentially executes the steps of power pipeline domain ontology library construction, semantic parsing and entity mapping, and semantic conflict detection and resolution. In practical implementation, the construction of the power pipeline ontology library needs to be based on the Gruber ontology construction method. This involves establishing a power pipeline ontology library comprising 5 major categories and 32 subcategories, including voltage level, laying method, pipeline material, functional type, and ownership unit. The OWL language is used to describe the topological relationships such as connections, inclusions, and intersections between pipeline segments, feature points, manholes, and ancillary facilities, defining 87 attribute association rules. Semantic parsing and entity mapping require the use of the BiLSTM-CRF semantic parsing engine to complete the recognition of power-related entities in drawing annotations and unstructured text, achieving an F1 score of 92.4%. This establishes a field-level semantic mapping between pipeline spatial elements and ontology library entities. Semantic conflict detection and resolution require a conflict detection library based on 78 core rules, employing a four-level conflict handling strategy to automatically resolve semantic conflicts at different levels, ensuring the consistency of semantic information for pipeline elements.

[0022] Furthermore, the three-dimensional topology intelligent verification module is the core module of this method. Based on the ontology library and three-dimensional spatial analysis technology, it realizes the compliance verification of the pipeline network topology relationship in all dimensions. This module sequentially performs the steps of three-dimensional topology rule base construction, network connectivity topology verification, three-dimensional spatial geometric topology verification, and business rule compliance verification. In practical implementation, the construction of the 3D topology rule base requires the establishment of a three-level topology rule system. The basic topology rules include 32 general rules such as pipeline connectivity, pipeline-manhole connectivity, and minimum pipeline clearance. Regionally specific rules include 15 specialized rules for soft soil areas in Shanghai, such as "pipeline settlement curvature radius ≥ 50D," "pipeline burial depth ≥ 5m below the riverbed" for river-crossing sections, and "safe distance between pipelines and structures" for central urban areas. Specialized pipeline rules are stratified according to voltage levels (10kV, 35kV, 110kV, 220kV), and include 40 power industry-specific rules such as "parallel spacing between 220kV cables and communication pipelines ≥ 10m" and "crossing angle of pipelines of different voltage levels ≥ 30°." Network connectivity topology verification requires constructing an adjacency matrix for the power pipeline network based on graph theory. A baseline network with pipeline feature points as nodes is constructed using Delaunay triangles. Isolated pipelines, isolated points, and suspended nodes are identified through connectivity component analysis. Loop detection is achieved through adjacency matrix eigenvalue analysis, verifying variable... The hierarchical connectivity and current flow logic of the power plant, switch station, and distribution station must be verified for compliance. Three-dimensional spatial geometric topology verification requires the use of an improved Douglas-Peucker algorithm, setting an accuracy threshold of ε=0.05m to detect pipeline routing distortion, and using a curvature analysis algorithm to set a curvature threshold of κ>0.15 to identify pipeline abrupt change points. A three-dimensional spatial relationship matrix of pipeline-building-geological bodies must be constructed. A three-dimensional OBB bounding box collision detection algorithm and the separating axis theorem must be used to calculate the minimum three-dimensional clearance between pipelines, identifying spatial intersections, overlaps, collisions, and insufficient clearance anomalies between pipelines and between pipelines and other structures. Business rule compliance verification must be based on the dynamic reasoning mechanism of the Drools rule engine, with a weight allocation of 0.5 for national standards, 0.3 for industry standards, and 0.2 for enterprise standards. This must be combined with the "Underground Pipeline Surveying Standard" DG / TJ08-85 and the "Power Engineering Cable Design Standard" GB50217-2018 to complete the compliance verification of pipeline burial depth, pipeline spacing, intersection angle, and laying curvature radius.

[0023] Furthermore, the anomaly precise location and classification module performs three-dimensional spatial coordinate location and risk level classification on the anomaly results identified by topology verification. This module sequentially executes the steps of anomaly three-dimensional coordinate location, anomaly feature extraction and classification, risk level classification, and automatic generation of rectification suggestions. In actual execution, anomaly three-dimensional coordinate location requires extracting the three-dimensional spatial coordinates of anomaly elements for each type of anomaly identified. The "problem cube" marking system is used to accurately locate the anomaly position, recording the plane coordinates, elevation, burial depth, and associated pipeline element information of the anomaly point, with positioning accuracy controlled within ±0.05m. Anomaly feature extraction and classification requires feature extraction of anomalies, classifying them into four categories: connectivity anomalies, geometric topology anomalies, spatial collision anomalies, and business compliance anomalies. Feature parameters such as anomaly amplitude, impact range, associated pipeline voltage level, and laying method are extracted. Risk level classification requires constructing a multi-dimensional weighted risk assessment model, in which pipeline voltage... The risk value is calculated by weighting the level (0.4), anomaly type (0.3), impact scope (0.2), and mandatory specification attribute (0.1). Anomalies are classified into three levels: major risk, general risk, and minor risk. Major risk refers to anomalies that violate mandatory national specifications and directly affect the safe operation of the power grid. General risk refers to anomalies that violate non-mandatory specifications and affect the daily operation and maintenance of pipelines. Minor risk refers to anomalies that are not properly marked and do not affect structural safety. The automatic generation of rectification suggestions is based on the power pipeline ontology database and rule knowledge graph. For different categories and levels of anomalies, the automatic generation of rectification measures, parameter thresholds, and construction suggestions that match the requirements of the specifications is required.

[0024] Furthermore, the visualization and report generation module completes the 3D visualization display of abnormal results and the automatic output of standardized quality inspection reports. This module sequentially executes the steps of 3D lightweight rendering, abnormal visualization interaction, and automatic quality inspection report generation. In actual execution, the 3D lightweight rendering requires the development of a web-based visualization module based on the Cesium engine, using 3DTiles layered loading technology to achieve smooth browsing of the 3D scene of power pipelines, abnormal locations, and related elements; the abnormal visualization interaction needs to realize interactive operations such as highlighting abnormal points, displaying attribute information pop-ups, spatial buffer analysis, and continuity analysis, supporting quick jumps to abnormal locations and one-click query of related elements; the automatic quality inspection report generation needs to support custom report templates, automatically fill in inspection results, abnormal details, spatial location, risk level, and rectification suggestions, and generate PDF and Word format quality inspection reports that comply with GB / T24356 "Quality Inspection and Acceptance of Surveying and Mapping Results" and the requirements of Shanghai Municipal Underground Pipeline Tracking Measurement Management, synchronously record the problem rectification status, and perform a second automatic check on the rectified data to achieve closed-loop management of the entire life cycle of abnormal issues.

[0025] To clearly illustrate the above embodiments, in one embodiment of this application, a new 10kV underground power pipeline construction project in Minhang District, Shanghai is used as an implementation case. The intelligent verification and anomaly location method for three-dimensional topology relationships of power pipelines of this application is executed. The specific steps are as follows: 1. Data Preparation: Access to underground pipeline tracking and measurement data, CAD design drawings, 1:500 topographic map GIS data, and pipeline point cloud data for this project. The total pipeline length is 8.7 kilometers, including both direct burial and trenchless pipe jacking methods, involving 32 cable wells and 2 switch stations along the route; 2. 3D Data Preprocessing Implementation: Multi-format data conversion was performed using a dedicated data parser to extract the 3D coordinates and attribute information of 126 pipeline segments and 342 pipeline feature points; The seven-parameter transformation model completed the transformation from the Shanghai local coordinate system to the CGCS2000 national geodetic coordinate system with a transformation accuracy of ±0.03m. The grid correction model completed the transformation from the Wusong elevation to the 1985 national elevation datum with an elevation residual of ±0.015m. Dirty data repair and topology normalization were completed, and an R* tree spatial index was constructed. 3. Implementation of semantic modeling for power pipeline ontology: Based on preprocessed data, semantic mapping between pipeline elements and the 10kV power pipeline ontology library was completed using BiLST. The M-CRF engine completes entity recognition of drawing annotations, detects and resolves 3 semantic conflicts, and ensures consistency between attribute information and spatial elements; 4. Implementation of intelligent verification of 3D topology relationships: Load a three-level topology rule base, and sequentially execute network connectivity verification, 3D spatial geometric topology verification, and business rule compliance verification to complete the topology relationship verification of the entire pipeline network; 5. Implementation of precise anomaly location and classification: A total of 12 anomalies were identified through topology verification, including 1 connectivity anomaly, 2 spatial collision anomalies, and 9 business compliance anomalies. The "problem cube" system completes the 3D coordinate location of all anomalies. After risk assessment, 1 major risk, 3 general risks, and 8 minor risks were classified, and rectification suggestions were automatically generated for each anomaly; 6. Implementation of visualization and report generation: The visualization and interactive query of all anomalies are completed through the web-based 3D scene, and a quality inspection report that meets the requirements of Shanghai's underground pipeline tracking and measurement management is automatically generated. At the same time, an anomaly rectification tracking ledger is established to achieve closed-loop management of rectification.

[0026] Understandably, the method in this application addresses the industry pain points of three-dimensional topology inspection of underground power lines. It constructs a fully optimized technical solution from five core aspects: "unified spatiotemporal benchmark, semantic modeling, three-dimensional topology verification, anomaly location and classification, and visualization output". This solution breaks through the technical limitations of traditional two-dimensional topology inspection and realizes fully automatic and high-precision intelligent verification of the three-dimensional topology relationship of underground power lines.

[0027] As a possible application, this method can be used for the quality inspection of the as-built measurement results of newly constructed power pipelines, providing a basis for compliance verification for pipeline engineering planning and final acceptance.

[0028] As another possible scenario, this method can be applied to urban existing power pipeline data governance projects to complete batch verification of topological relationships and anomaly investigation of large-scale pipeline data, thereby improving the quality of pipeline basic data.

[0029] As another possible scenario, this method can be applied to the investigation of potential underground pipeline hazards before construction projects, accurately identifying power line topological anomalies and spatial collision hazards within the construction impact area, and ensuring construction safety.

[0030] It should be noted that the three-level topology rule library built into this method supports user-defined configuration. The rule parameters and thresholds can be flexibly adjusted according to local regulations in different regions and industry requirements for different voltage levels to adapt to the inspection needs of different regions and scenarios.

[0031] Optionally, this method can be seamlessly integrated with the Shanghai Municipal Underground Pipeline Tracking and Measurement Management Platform and the Power Grid GIS Platform to achieve real-time sharing and data linkage of inspection results, and be integrated into the entire life cycle management process of power pipelines.

[0032] In summary, the intelligent verification and anomaly location method for three-dimensional topology relationships of power pipelines in this application overcomes the technical limitations of traditional two-dimensional topology inspection by constructing a full-process three-dimensional topology verification technology system. It achieves fully automatic, standardized, and high-precision intelligent verification and accurate anomaly location of three-dimensional spatial topology relationships of underground power pipelines. The inspection efficiency is 8-10 times higher than that of traditional manual methods, and the anomaly identification rate can reach over 95%. It effectively solves the industry pain points of low efficiency, inconsistent standards, and insufficient identification rate of hidden anomalies in traditional inspection methods. It can be widely applied to scenarios such as power pipeline quality inspection, data governance, and hidden danger investigation, providing reliable technical support for the safe operation and maintenance and refined management of urban underground power networks. It has significant engineering application value and industry promotion significance.

[0033] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0034] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for intelligent verification and anomaly localization of three-dimensional topological relationships in power pipelines, characterized in that, It includes a 3D data preprocessing module, a power pipeline ontology semantic modeling module, a 3D topology intelligent verification module, an anomaly accurate location and classification module, and a visualization and report generation module; The 3D data preprocessing module completes the standardized integration and spatiotemporal benchmark unification of multi-source power pipeline spatial data. The power pipeline ontology semantic modeling module constructs an ontology library and semantic association rule system for the power pipeline domain based on the standardized 3D pipeline dataset generated by preprocessing. The 3D topology intelligent verification module realizes the compliance verification of pipeline network topology relationships in all dimensions based on the ontology library and 3D spatial analysis technology. The anomaly accurate location and classification module performs 3D spatial coordinate location and risk level classification of the anomaly results identified by topology verification. The visualization and report generation module completes the 3D visualization display of anomaly results and the automatic output of standardized quality inspection reports.

2. The intelligent verification and anomaly location method for three-dimensional topology relationships of power pipelines according to claim 1, characterized in that, The three-dimensional data preprocessing module includes the following steps: S101. Multi-source data access and format conversion: Access power pipeline tracking and measurement results data, design drawing data, laser point cloud data, and GIS vector data; develop a dedicated data parser to complete lossless conversion of dwg, shp, las, and xlsx formats; and extract the three-dimensional spatial coordinates and attribute information of pipeline segments, pipeline feature points, manholes, and ancillary facilities. S102. Unification of Spatiotemporal References: A three-level reference transformation system is constructed to achieve standardization and unification of plane coordinates and elevation references. The plane reference adopts a seven-parameter transformation model between the Shanghai local coordinate system and the CGCS2000 national geodetic coordinate system, with the transformation accuracy controlled within ±0.05m. The elevation reference is constructed based on a grid correction model of the Wusong elevation and the 1985 national elevation reference, with the elevation transformation residual controlled within ±0.02m. The time reference establishes a time-scaled system based on UTC time to achieve millisecond-level time synchronization and historical data metadata traceability. S103. Data Cleaning and Topology Normalization: Based on the rule engine, the system automatically repairs dirty data with missing fields, out-of-limit coordinates, and outlier pipe diameters. It repairs missing data and discrete points through interpolation and smoothing algorithms, performs topology normalization processing such as pseudo-node elimination, redundant vertex deletion, and duplicate feature removal, and constructs an R* tree spatial index to improve the efficiency of subsequent spatial analysis.

3. The intelligent verification and anomaly location method for three-dimensional topology relationships of power pipelines according to claim 1, characterized in that, The semantic modeling module for the power pipeline ontology includes the following steps: S201. Construction of the ontology library in the field of power pipelines: Based on the Gruber ontology construction method, a power pipeline ontology library containing 5 major categories and 32 subcategories is established, covering the core dimensions of voltage level, laying method, pipeline material, functional type, and ownership unit. The connection, containment, and intersection topology relationships between pipeline segments, feature points, manholes, and ancillary facilities are described using the OWL language, and 87 attribute association rules are defined. S202, Semantic parsing and entity mapping: The BiLSTM-CRF semantic parsing engine is used to complete entity recognition of unstructured text and drawing annotations, establish semantic mapping relationship between pipeline space elements and entities in the ontology library, and realize automatic field-level mapping of "IfcCable→Cable model" and "IfcJunctionBox→Junction box number". S203. Semantic Conflict Detection and Resolution: Establish a semantic conflict detection rule base containing 78 core rules, construct a four-level conflict handling strategy based on DS evidence theory, and complete hierarchical resolution of automatic processing, rule processing, manual processing, and expert consultation to ensure the consistency and compliance of semantic information of pipeline elements.

4. The intelligent verification and anomaly location method for three-dimensional topology relationships of power pipelines according to claim 1, characterized in that, The intelligent verification module for three-dimensional topological relationships includes the following steps: S301. Construction of a 3D Topology Rule Base: Establish a three-level topology rule system, including basic topology rules, regional characteristic rules, and professional pipeline rules. The basic topology rules include 32 general rules on pipeline connectivity, pipeline-manhole connection, and minimum pipeline clearance. The regional characteristic rules set 15 special rules for soft soil areas, river crossings, and central urban areas in Shanghai. The professional pipeline rules are layered according to voltage level, with 40 power industry-specific rules. S302. Network connectivity topology verification: Construct an adjacency matrix for the power pipeline network based on graph theory, perform connectivity component analysis to identify isolated pipelines, isolated nodes, and suspended nodes, identify illegal closed loops in the pipeline network through loop detection algorithms, and verify the hierarchical connectivity relationship and current flow logic compliance of substation-switching station-distribution station. S303, 3D Spatial Geometric Topology Verification: An improved Douglas-Peucker algorithm is used to detect pipeline routing distortion, with an accuracy threshold of ε=0.05m; a curvature analysis algorithm is used to identify pipeline abrupt change points, with a curvature threshold of κ>0.15; a 3D spatial relationship matrix of pipeline-building-geological body is constructed, and a 3D OBB bounding box collision detection algorithm is used to identify spatial intersections, overlaps, and collision anomalies between pipelines and between pipelines and other structures; S304. Business Rule Compliance Verification: Based on the dynamic reasoning mechanism of the Drools rule engine, combined with the "Underground Pipeline Surveying Standard" DG / TJ08-85 and the "Power Engineering Cable Design Standard" GB50217-2018, the compliance verification of pipeline burial depth, pipeline spacing, crossing angle, and laying curvature radius is completed, realizing the adaptive allocation and dynamic reasoning of the weights of national standards, industry standards, and enterprise standards.

5. The intelligent verification and anomaly location method for three-dimensional topology relationships of power pipelines according to claim 4, characterized in that, The anomaly precise location and classification module includes the following steps: S401, Anomaly 3D Coordinate Location: For various anomalies identified by topology verification, the 3D spatial coordinates of the anomaly elements are extracted. The "problem cube" marking system is used to accurately locate the anomaly position and record the plane coordinates, elevation, burial depth, and associated pipeline element information of the anomaly point. The positioning accuracy is controlled within ±0.05m. S402. Anomaly Feature Extraction and Classification: Anomalies are feature extracted and classified into four categories: connectivity anomalies, geometric topology anomalies, spatial collision anomalies, and business compliance anomalies. For each type of anomaly, feature parameters such as anomaly amplitude, impact range, associated pipeline voltage level, and laying method are extracted. S403. Risk Level Classification: Based on the abnormal characteristic parameters, a risk assessment model is constructed, which is divided into three levels according to the degree of impact: major risk, general risk, and minor risk. Among them, major risk refers to anomalies that affect the safe operation of the power grid and violate mandatory regulations; general risk refers to non-mandatory violation anomalies that affect pipeline operation and maintenance; and minor risk refers to anomalies related to mapping and annotation that do not affect structural safety. S404 Automatic generation of rectification suggestions: Based on the power pipeline ontology library and rule knowledge graph, rectification measures and parameter suggestions matching the specifications are automatically generated for different categories and levels of anomalies.

6. The intelligent verification and anomaly location method for three-dimensional topology relationships of power pipelines according to claim 1, characterized in that, The visualization and report generation module includes the following steps: S501, 3D Lightweight Rendering: Based on the Cesium engine, a web-based visualization module is developed, using 3DTiles layered loading technology to achieve smooth browsing and visualization of 3D power pipeline scenes, abnormal locations, and related elements; S502, Anomaly Visualization Interaction: Enables interactive operations such as highlighting anomaly points, displaying attribute information pop-ups, analyzing spatial buffers, and analyzing connectivity, and supports quick navigation to anomaly locations and one-click querying of related elements; S503 Automatic Quality Inspection Report Generation: Supports custom report templates, automatically fills in inspection results, anomaly details, spatial location, risk level, and rectification suggestions, and generates PDF and Word format quality inspection reports that comply with GB / T24356 "Quality Inspection and Acceptance of Surveying and Mapping Results" and the requirements of Shanghai Municipality's underground pipeline tracking measurement management. It also records the status of problem rectification in real time, realizing closed-loop management of rectification.

7. The intelligent verification and anomaly location method for three-dimensional topology relationships of power pipelines according to claim 2, characterized in that, In S102, the plane datum conversion was achieved by setting up 12 high-level control points and using a Leica TS60 total station for field measurement, with a plane coordinate measurement error of ≤±3mm. The elevation datum conversion involved setting up 156 leveling points in a 2km×2km grid throughout the city, using a DNA03 electronic level for second-order leveling, and constructing a grid correction model.

8. The intelligent verification and anomaly location method for three-dimensional topology relationships of power pipelines according to claim 4, characterized in that, In S302, a Delaunay triangle is used to construct a baseline pipeline network with pipeline feature points as nodes. The connectivity of the pipeline network is verified by robust least squares method, and loop detection is achieved based on adjacency matrix eigenvalue analysis to identify illegal closed-loop pipelines.

9. The intelligent verification and anomaly location method for three-dimensional topology relationships of power pipelines according to claim 4, characterized in that, In S303, 3D OBB bounding box collision detection constructs a minimum directed bounding box for pipeline segments, determines the intersection relationship between bounding boxes using the separating axis theorem, calculates the minimum 3D clearance between pipelines, compares it with the clearance threshold in the rule base, and identifies anomalies with insufficient clearance.

10. The intelligent verification and anomaly location method for three-dimensional topology relationships of power pipelines according to claim 5, characterized in that, The risk assessment model in S403 adopts a multi-dimensional weight allocation, with pipeline voltage level weighted at 0.4, anomaly type weighted at 0.3, impact range weighted at 0.2, and mandatory specification attribute weighted at 0.

1. The risk value is calculated by weighted summation, and the corresponding risk level is classified.