Weak current pipeline laying quality online inspection method and system

Abstract

The application relates to the technical field of quality inspection, and discloses a weak-current pipeline laying quality online inspection method and system.The method comprises the following steps: performing structured information analysis on the real-time construction state of target project laying to obtain quality parameter data; performing space-time coordinate mapping on the quality parameter data and the design specification of the target project laying to obtain a dynamic anchoring relationship; based on the dynamic anchoring relationship, continuously performing compliance evaluation on the quality parameter data to obtain a space-time deviation degree; based on the space-time deviation degree, performing key feature extraction on the quality parameter data to obtain a key abnormal mode; based on the design specification, constructing a logical topology graph of the target project laying, based on the logical topology graph, performing influence deduction on the key abnormal mode to obtain an abnormal risk grade; based on the abnormal risk grade, comprehensively integrating the key abnormal mode and the quality parameter data to obtain an inspection report; and the application can improve the efficiency of weak-current pipeline laying quality online inspection.

Description

Technical Field

[0001] This invention relates to the field of quality inspection technology, and in particular to an online inspection method and system for the quality of low-voltage pipeline laying. Background Technology

[0002] In the current quality inspection of low-voltage pipeline installation, data collection lacks a systematic multi-source integration mechanism, and the status data is fragmented. Furthermore, data preprocessing steps such as time alignment and redundancy cleaning rely on manual operation, resulting in low structure and insufficient accuracy of quality parameter data. This makes it impossible to provide standardized and calculable core data support for subsequent compliance assessments, leading to poor real-time linkage between construction status and quality data, and making it difficult to achieve early identification of quality problems.

[0003] Traditional inspection methods lack precise spatiotemporal mapping between design specifications and actual construction data. Compliance assessments often employ phased sampling, failing to cover the continuous state of the entire construction process, resulting in incomplete and untimely calculations of spatiotemporal deviations. Furthermore, anomaly pattern recognition lacks integration with engineering semantic constraints and continuous analysis, and risk level determination lacks quantitative topological analysis and path influence weight calculations, leading to ambiguous hierarchical classification of anomaly risks. Inspection report generation relies on manual integration, resulting in low efficiency and a lack of precise decision-making guidance. Therefore, improving the efficiency of online inspection of low-voltage cable laying quality has become an urgent problem to be solved. Summary of the Invention

[0004] This invention provides an online inspection method and system for the quality of low-voltage pipeline installation, in order to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides an online inspection method for the quality of low-voltage wiring installation, comprising: S01. Perform structured information parsing on the real-time construction status of the target project to obtain the quality parameter data of the real-time construction status. S02. Perform spatiotemporal coordinate mapping between the quality parameter data and the design specifications for the target project to obtain the dynamic anchoring relationship of the quality parameter data; S03. Based on the dynamic anchoring relationship, perform continuous frame compliance assessment on the quality parameter data to obtain the spatiotemporal deviation of the quality parameter data; S04. Based on the spatiotemporal deviation, key features are extracted from the quality parameter data to obtain the key anomaly patterns of the quality parameter data; S05. Based on the connection dependencies of the design specifications, construct the logical topology diagram of the target project, and based on the logical topology diagram, perform impact deduction on the key anomaly mode to obtain the anomaly risk level of the key anomaly mode. S06. Based on the aforementioned abnormal risk level, the key abnormal patterns and the quality parameter data are comprehensively integrated to obtain an online inspection report of the target project's installation.

[0006] In a preferred embodiment, the step of performing structured information parsing on the real-time construction status of the target project to obtain quality parameter data of the real-time construction status includes: Multi-source status acquisition is performed on the real-time construction status of the target project to obtain the status data of the real-time construction status. The state data is time-series aligned, and redundant information is cleaned from the aligned data to obtain a regular state sequence of the real-time construction state. The regular state sequence is deconstructed in a multidimensional way to obtain the quality feature sequence of the regular state sequence. The quality feature sequence is normalized to obtain the quality parameter data of the real-time construction status.

[0007] In a preferred embodiment, the step of performing spatiotemporal coordinate mapping between the quality parameter data and the design specifications for the target project to obtain the dynamic anchoring relationship of the quality parameter data includes: The design specifications for the target project are divided into spatiotemporal grids to obtain the reference spatiotemporal grid for the design specifications. Based on the reference spatiotemporal grid, the matching degree of the quality parameter data is verified to obtain the spatiotemporal deviation of the quality parameter data; Based on the spatiotemporal deviation, the mass parameter data is interpolated and fitted to obtain the continuous spatiotemporal trajectory of the mass parameter data; Based on the construction load data of the target project, the continuous spatiotemporal trajectory is subjected to load compensation correction to obtain the corrected spatiotemporal trajectory of the continuous spatiotemporal trajectory. The corrected spatiotemporal trajectory is associated and bound point by point with the reference spatiotemporal grid to obtain the dynamic anchoring relationship of the quality parameter data.

[0008] In a preferred embodiment, the step of performing continuous frame compliance assessment on the quality parameter data based on the dynamic anchoring relationship to obtain the spatiotemporal deviation of the quality parameter data includes: Based on the dynamic anchoring relationship, the quality parameter data is segmented for process consistency to obtain segmented quality data of the quality parameter data. A two-way compliance cross-validation is performed on the overlapping areas of the segmented quality data to obtain the judgment result of the segmented quality data. Based on the determination result, the segmented quality data is extrapolated within the segment to obtain the predicted compliance threshold of the segmented quality data. Based on the predicted compliance threshold, a forward-looking anomaly detection is performed on the quality parameter data to obtain parameter deviation markers for the quality parameter data; The spatiotemporal weighted integral of the parameter deviation marker is performed to obtain the spatiotemporal deviation of the quality parameter data.

[0009] In a preferred embodiment, the step of extrapolating the segmented quality data based on the determination result to obtain the predicted compliance threshold of the segmented quality data includes: The segmented quality data is time-ordered to obtain a time-domain aligned data sequence of the segmented quality data. Based on the determination result, a nonlinear weight allocation is performed on the time-domain aligned data sequence to obtain the quality weight sequence of the time-domain aligned data sequence. Based on the quality weight sequence, the time-domain aligned data sequence is synthesized by weighted average to obtain the weighted average value of the time-domain aligned data sequence; The trend change rate of the time-domain aligned data sequence is analyzed to obtain the overall change rate of the time-domain aligned data sequence. Based on the weighted average, the overall rate of change, and the judgment result, the predicted compliance threshold for the segmented quality data is calculated, wherein the formula for calculating the predicted compliance threshold is: ; In the formula, The predicted compliance threshold, The index of the data point in the time-domain aligned data sequence. The number of data points in the time-domain aligned data sequence. The determination result is as follows. For index The corresponding quality parameter data, The quality parameter data for the last data point in the time-domain aligned data sequence. The quality parameter data is for the first data point in the time-domain aligned data sequence.

[0010] In a preferred embodiment, the step of extracting key features from the quality parameter data based on the spatiotemporal deviation to obtain the key anomaly patterns of the quality parameter data includes: Based on the spatiotemporal deviation, the anomaly persistence of the quality parameter data is evaluated to obtain the process persistence coefficient of the quality parameter data; Based on the process persistence coefficient, intermittent anomaly patterns are identified in the quality parameter data to obtain candidate anomaly patterns of the quality parameter data. Based on the preset construction tolerance deviation, the candidate anomaly patterns are filtered by engineering semantic constraints to obtain the engineering anomaly patterns of the candidate anomaly patterns. The abnormal patterns in the engineering are categorized and merged to obtain the key abnormal patterns in the quality parameter data.

[0011] In a preferred embodiment, based on the connection dependencies of the design specifications, a logical topology diagram of the target project is constructed, and based on the logical topology diagram, the impact of the key anomaly patterns is deduced to obtain the anomaly risk level of the key anomaly patterns, including: Semantic dependency destructuring is performed on the connection dependencies of the design specification to obtain the connection rules of the design specification; Based on the connection rules and the location information of the key anomaly pattern, neighborhood topology induction is performed on the key anomaly pattern to obtain a local logical topology subgraph of the key anomaly pattern. Based on the local logic topology subgraph, signal flow resistance analysis is performed on the key anomaly mode to obtain the local blocking characteristics of the key anomaly mode. Based on the local blocking characteristics and the connection rules, the key dependency path is inverted to obtain the logical topology diagram of the key anomaly pattern. Based on the logical topology graph, the structural vulnerability of the key anomaly pattern is quantified to obtain the path influence weight of the key anomaly pattern. Based on the path influence weight, a multi-level threshold adaptive determination is performed on the key anomaly pattern to obtain the anomaly risk level of the key anomaly pattern.

[0012] In a preferred embodiment, the step of quantifying the structural vulnerability of the key anomaly pattern based on the logical topology graph to obtain the path influence weight of the key anomaly pattern includes: The node betweenness centrality and edge clustering coefficient of the logical topology graph are extracted in parallel in two dimensions to obtain the topological structural feature parameters of the logical topology graph. Based on the logical topology diagram, spatial statistical characterization is performed on the key anomaly pattern to obtain the anomaly distribution characteristic parameters of the key anomaly pattern. Based on the topological structure feature parameters and the abnormal distribution feature parameters, the vulnerability score of each edge in the logical topology graph is calculated, wherein the formula for calculating the vulnerability score is: ; In the formula, For the edges in the logical topology graph. For the edge Vulnerability score, For node variables, For the edge endpoint nodes, For nodes betweenness centrality For nodes The edge aggregation coefficient, It is an exponential function. For nodes The topological shortest distance to the key anomaly pattern. The mean of the topological distribution among the abnormal distribution characteristic parameters is... The variance of the topological distribution in the abnormal distribution characteristic parameters; The vulnerability scores are mapped using a probability distribution to obtain the path influence weights of the key anomaly patterns.

[0013] In a preferred embodiment, the step of comprehensively integrating the key anomaly patterns and the quality parameter data based on the anomaly risk level to obtain an online inspection report for the target project's installation includes: Based on the aforementioned anomaly risk level, the key anomaly patterns are dynamically prioritized to obtain a priority sequence for the key anomaly patterns. Based on the priority sequence, the quality parameter data is correlated and labeled to obtain labeled data of the quality parameter data; Based on the priority sequence, the labeled data is subjected to temporal attention weighting to obtain the time-weighted sequence of the labeled data; The time-weighted sequence is organized into an itemized structure to obtain an online inspection report of the target project's installation.

[0014] To address the aforementioned problems, the present invention also provides an online inspection system for the quality of low-voltage wiring installation, the system comprising: The quality parameterization analysis module is used to perform structured information analysis on the real-time construction status of the target project to obtain the quality parameter data of the real-time construction status. The spatiotemporal reference alignment module is used to perform spatiotemporal coordinate mapping between the quality parameter data and the design specifications for the laying of the target project, so as to obtain the dynamic anchoring relationship of the quality parameter data. The compliance dynamic judgment module is used to perform continuous frame compliance evaluation on the quality parameter data based on the dynamic anchoring relationship, and obtain the spatiotemporal deviation of the quality parameter data. An anomaly feature extraction module is used to extract key features from the quality parameter data based on the spatiotemporal deviation to obtain the key anomaly patterns of the quality parameter data. The topology risk assessment module is used to construct a logical topology diagram of the target project based on the connection dependencies of the design specifications, and to perform impact deduction on the key anomaly modes based on the logical topology diagram to obtain the anomaly risk level of the key anomaly modes. The comprehensive report generation module is used to perform comprehensive quality integration of the key anomaly patterns and the quality parameter data based on the anomaly risk level, and obtain an online inspection report of the target project's installation.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention achieves temporal alignment, redundancy removal, and feature regularization of construction status data through multi-source status acquisition and structured information parsing. Combined with continuous frame compliance assessment and automated verification process, it significantly improves the overall efficiency of quality inspection of low-voltage pipeline laying, realizes real-time online monitoring of the construction process and automated advancement of the inspection process, quickly outputs standardized quality parameter data, and shortens the inspection cycle and report generation time.

[0016] 2. By leveraging spatiotemporal coordinate mapping to establish precise dynamic anchoring between quality parameters and design specifications, and through the construction of logical topology diagrams, quantification of structural vulnerability, and deduction of risk levels, the system achieves precise extraction of key anomaly patterns and scientific assessment of anomaly impacts. This enhances the accuracy of quality inspection and the comprehensiveness of risk identification, providing precise anomaly location and graded handling basis for engineering quality control, and ensuring the guidance and reliability of inspection results. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating an online inspection method for the quality of low-voltage pipeline installation according to an embodiment of the present invention. Figure 2 This is a functional module diagram of an online inspection system for the quality of low-voltage pipeline laying provided in an embodiment of the present invention; The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0018] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0019] This application provides an online inspection method for the quality of low-voltage cable laying. The executing entity of this online inspection method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the online inspection method for the quality of low-voltage cable laying can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0020] Reference Figure 1 The diagram shown is a flowchart illustrating an online inspection method for the quality of low-voltage cable laying according to an embodiment of the present invention. In this embodiment, the online inspection method for the quality of low-voltage cable laying includes: S01. Perform structured information parsing on the real-time construction status of the target project to obtain the quality parameter data of the real-time construction status. In this embodiment of the invention, the step of performing structured information parsing on the real-time construction status of the target project to obtain the quality parameter data of the real-time construction status includes: Multi-source status acquisition is performed on the real-time construction status of the target project to obtain the status data of the real-time construction status. The state data is time-series aligned, and redundant information is cleaned from the aligned data to obtain a regular state sequence of the real-time construction state. The regular state sequence is deconstructed in a multidimensional way to obtain the quality feature sequence of the regular state sequence. The quality feature sequence is normalized to obtain the quality parameter data of the real-time construction status.

[0021] Multi-source status acquisition uses various data acquisition devices deployed in the construction area of ​​low-voltage pipelines to comprehensively collect various raw information generated in real time during the laying process of the target project from multiple different monitoring perspectives, such as construction progress, laying location, connection status, and environmental conditions. All the raw information collected together constitutes the status data of the real-time construction status.

[0022] Using a preset fixed time interval as a unified benchmark, status data from different acquisition devices and monitoring angles are mapped to the same time node according to their respective time sequence to ensure that all data are consistent in the time dimension. Redundant information cleaning involves checking and aligning the status data one by one, removing duplicate records, interference information unrelated to the quality of low-voltage pipeline laying, and invalid information that cannot reflect the actual construction status. After these two operations, a regular status sequence of real-time construction status is formed.

[0023] For the regular state sequence, it is broken down according to the core concern dimensions of low-voltage pipeline laying quality. Specifically, from multiple fixed dimensions such as spatial layout, laying process, connection reliability, and material compatibility, quality-related information corresponding to each dimension is extracted from the regular state sequence. These quality-related information extracted from each dimension are integrated to form the quality feature sequence of the regular state sequence.

[0024] The feature information of each dimension in the quality feature sequence is standardized and uniformly processed. Specifically, the expression form and data format of different feature dimensions are adjusted to a unified standard form to eliminate the expression differences and format discrepancies caused by different collection methods. This ensures that all feature information is consistent and comparable in subsequent use. After standardization, the quality parameter data of the real-time construction status is obtained.

[0025] The beneficial effects include the ability to comprehensively collect various real-time raw information during the construction process of low-voltage pipeline laying, ensuring that the status data covers key dimensions of the entire construction scenario. By aligning the time sequence, data from different sources are kept consistent in the time dimension, while duplicate, interference, and invalid information are eliminated to ensure the validity and regularity of the data. Furthermore, by extracting feature information directly related to quality through multi-dimensional decomposition and combining it with standardized processing according to unified standards, differences in the expression and format of feature information are eliminated. This ensures that the final quality parameter data is comprehensive, accurate, consistent, and comparable, providing a reliable and standardized foundation for subsequent steps such as benchmarking of quality parameter data against design specifications and compliance assessment. This helps to improve the accuracy and efficiency of online inspection of low-voltage pipeline laying quality.

[0026] S02. Perform spatiotemporal coordinate mapping between the quality parameter data and the design specifications for the target project to obtain the dynamic anchoring relationship of the quality parameter data; In this embodiment of the invention, the step of performing spatiotemporal coordinate mapping between the quality parameter data and the design specifications for the target project to obtain the dynamic anchoring relationship of the quality parameter data includes: The design specifications for the target project are divided into spatiotemporal grids to obtain the reference spatiotemporal grid for the design specifications. Based on the reference spatiotemporal grid, the matching degree of the quality parameter data is verified to obtain the spatiotemporal deviation of the quality parameter data; Based on the spatiotemporal deviation, the mass parameter data is interpolated and fitted to obtain the continuous spatiotemporal trajectory of the mass parameter data; Based on the construction load data of the target project, the continuous spatiotemporal trajectory is subjected to load compensation correction to obtain the corrected spatiotemporal trajectory of the continuous spatiotemporal trajectory. The corrected spatiotemporal trajectory is associated and bound point by point with the reference spatiotemporal grid to obtain the dynamic anchoring relationship of the quality parameter data.

[0027] According to the spatiotemporal scope specified in the design specifications for the target project, the overall spatiotemporal area is evenly divided into several grid units of uniform size and clear boundaries. Each grid unit is marked with a unique spatiotemporal identifier. Through this systematic division, a reference spatiotemporal grid corresponding to the design specifications is formed.

[0028] The spatiotemporal information contained in the quality parameter data is compared one by one with the grid cells in the reference spatiotemporal grid to confirm whether the spatiotemporal position corresponding to each quality parameter data falls within the spatiotemporal boundary of the corresponding grid cell. This direct comparison clarifies the degree of fit between the data and the grid cell, thereby determining the spatiotemporal deviation of each quality parameter data relative to the reference spatiotemporal grid.

[0029] To address the spatiotemporal biases in the quality parameter data, based on the spatiotemporal information and bias values ​​of adjacent valid data points, complete intermediate data points are added within the spatiotemporal interval corresponding to the bias to fill data gaps. This allows the originally potentially discrete quality parameter data to form a coherent, unbroken trajectory, ultimately resulting in a continuous spatiotemporal trajectory of the quality parameter data.

[0030] Collect relevant information on construction loads during the laying process of the target project, analyze the influence of load magnitude and distribution range on the laying status of low-voltage pipelines, and make targeted adjustments to the trajectory segments affected by loads in the continuous spatiotemporal trajectory based on this law to eliminate trajectory deviations caused by load factors, thereby obtaining the corrected spatiotemporal trajectory of the continuous spatiotemporal trajectory.

[0031] By traversing every data point on the correction spatiotemporal trajectory, the corresponding grid cell in the reference spatiotemporal grid is found based on the spatiotemporal information of the data point. A fixed correspondence is established between the data point and the corresponding grid cell to ensure that each data point can be accurately associated with a specific position in the reference spatiotemporal grid. Through this comprehensive point-by-point association, the dynamic anchoring relationship of the quality parameter data is obtained.

[0032] The beneficial effects are that the reference spatiotemporal grid formed by the systematic division provides a unified and clear spatiotemporal benchmark for quality parameter data. By comparing the clear spatiotemporal deviations one by one, the spatiotemporal fit between the data and the design specifications is accurately reflected. The continuous spatiotemporal trajectory formed by interpolation fills the data gaps, ensuring the continuity and integrity of the data. Load compensation correction eliminates the interference of construction loads on the trajectory, ensuring the authenticity and accuracy of the trajectory. Finally, through the dynamic anchoring relationship established by point-by-point association, the precise correspondence between quality parameter data and design specifications in the spatiotemporal dimension is achieved. This provides a reliable and accurate foundation for subsequent compliance assessments of quality parameter data and other links, ensuring the scientificity and effectiveness of the entire online inspection process.

[0033] S03. Based on the dynamic anchoring relationship, perform continuous frame compliance assessment on the quality parameter data to obtain the spatiotemporal deviation of the quality parameter data; In this embodiment of the invention, the step of performing continuous frame compliance assessment on the quality parameter data based on the dynamic anchoring relationship to obtain the spatiotemporal deviation of the quality parameter data includes: Based on the dynamic anchoring relationship, the quality parameter data is segmented for process consistency to obtain segmented quality data of the quality parameter data. A two-way compliance cross-validation is performed on the overlapping areas of the segmented quality data to obtain the judgment result of the segmented quality data. Based on the determination result, the segmented quality data is extrapolated within the segment to obtain the predicted compliance threshold of the segmented quality data. Based on the predicted compliance threshold, a forward-looking anomaly detection is performed on the quality parameter data to obtain parameter deviation markers for the quality parameter data; The spatiotemporal weighted integral of the parameter deviation marker is performed to obtain the spatiotemporal deviation of the quality parameter data.

[0034] Based on the determination result, the segmented quality data is extrapolated intra-segmentally to obtain the predicted compliance threshold of the segmented quality data, including: The segmented quality data is time-ordered to obtain a time-domain aligned data sequence of the segmented quality data. Based on the determination result, a nonlinear weight allocation is performed on the time-domain aligned data sequence to obtain the quality weight sequence of the time-domain aligned data sequence. Based on the quality weight sequence, the time-domain aligned data sequence is synthesized by weighted average to obtain the weighted average value of the time-domain aligned data sequence; The trend change rate of the time-domain aligned data sequence is analyzed to obtain the overall change rate of the time-domain aligned data sequence. Based on the weighted average, the overall rate of change, and the judgment result, the predicted compliance threshold for the segmented quality data is calculated, wherein the formula for calculating the predicted compliance threshold is: ; In the formula, The predicted compliance threshold, The index of the data point in the time-domain aligned data sequence. The number of data points in the time-domain aligned data sequence. The determination result is as follows. For index The corresponding quality parameter data, The quality parameter data for the last data point in the time-domain aligned data sequence. The quality parameter data is for the first data point in the time-domain aligned data sequence.

[0035] Based on the dynamic anchoring relationship between quality parameter data and design specifications, the changing characteristics of quality parameters during construction are analyzed, and continuous segments with stable trends and consistent characteristics are identified in the data. The quality parameter data is then divided into multiple independent and complete segments, each corresponding to a coherent stage in the construction process, thus obtaining segmented quality data of the quality parameter data.

[0036] For overlapping areas between two adjacent segments in the segmented quality data, bidirectional verification is performed based on the compliance standards of the two segments. First, the compliance requirements of the first segment are used to verify the compliance of the overlapping area data, and then the compliance requirements of the second segment are used to verify the compliance of the same overlapping area data. The combined results of the two verifications lead to a clear conclusion as to whether the overlapping area data complies with the specifications, and thus the judgment result of the segmented quality data is obtained.

[0037] The segmented quality data is reorganized in chronological order, and the distribution of data points on the time axis is adjusted so that all data points correspond to the same time interval, eliminating time misalignment between different data points and forming a time-domain aligned data sequence of segmented quality data.

[0038] Based on the judgment results of the segmented quality data, different weights are assigned to each data point in the time-domain aligned data sequence. Data points that are compliant are given higher weights, and data points that are non-compliant are given lower weights. The weight allocation does not adopt a uniform distribution method, but presents a non-linear difference according to the compliance status, and finally obtains the quality weight sequence of the time-domain aligned data sequence.

[0039] The weighted average of the time-domain aligned data sequence is obtained by multiplying the value of each data point in the time-domain aligned data sequence by the corresponding weight in the quality weight sequence, calculating the sum of all products, and then dividing by the sum of all weights in the quality weight sequence.

[0040] The numerical difference between two adjacent data points in the time-domain aligned data sequence is calculated one by one. Combined with the time intervals corresponding to the data points, the unit time change of each adjacent data pair is determined. The changes of all adjacent data pairs are integrated to extract the change pattern of the entire data sequence in the time dimension, and the overall change rate of the time-domain aligned data sequence is obtained.

[0041] Based on the weighted average of the time-domain aligned data series, the data is adjusted according to the trend and magnitude of the overall rate of change. Then, it is further corrected in combination with the compliance status reflected by the judgment results, so that the results not only fit the average level and trend of the data itself, but also meet the requirements of compliance judgment, and finally the predicted compliance threshold of segmented quality data is obtained.

[0042] The predicted compliance threshold is used as the compliance judgment standard for quality parameter data. The relationship between each data point in the quality parameter data and the predicted compliance threshold is compared one by one. If the data point exceeds the range of the predicted compliance threshold, the data point is marked and its deviation from the compliance standard is clearly recorded, thus obtaining the parameter deviation mark of the quality parameter data.

[0043] Based on the construction stage and spatial location of the parameter deviation marker, corresponding spatiotemporal weights are set. The weights are higher for critical construction stages and important pipeline spatial areas, and lower for non-critical stages and minor areas. The deviation degree of each parameter deviation marker is multiplied by the corresponding spatiotemporal weight, and then all the product results are summed. The spatiotemporal deviation of the quality parameter data is obtained through this integration method.

[0044] The indexes are derived from a time-aligned data sequence, allocated sequentially according to the chronological order of data points within the sequence, with each data point corresponding to a unique sequence identifier. The number of data points is directly obtained by counting the total number of data points in the time-aligned data sequence. The judgment result comes from bidirectional compliance cross-validation of overlapping areas of segmented quality data. First, the compliance requirements of the first segment are used to verify the compliance of the overlapping area data, and then the compliance requirements of the second segment are used to verify the compliance of the same overlapping area data. The combined results of the two verifications yield a clear conclusion. The quality parameter data corresponding to the indexes are directly extracted from the time-aligned data sequence, with each index position corresponding to the data content at the same position in the sequence. The quality parameter data of the last data point in the time-aligned data sequence is directly selected from the data at the end of the sequence. The quality parameter data of the first data point in the time-aligned data sequence is directly selected from the data at the beginning of the sequence.

[0045] This formula assigns time-related weights to each data point in a time-domain aligned data sequence, adjusts the weights based on the judgment results, incorporates the variation amplitude at the beginning and end of the data sequence, and corrects it using the judgment results to comprehensively calculate the predicted compliance threshold for segmented quality data. Its core principle is to balance the weight differences of data points in the time dimension, the impact of compliance judgment results, and the changing trends of the data itself. This ensures that the obtained predicted compliance threshold not only conforms to the overall distribution characteristics of the data but also reflects compliance requirements and data change patterns, providing a clear and practical judgment standard for the forward-looking anomaly detection of subsequent quality parameter data.

[0046] As data points progress sequentially through the time-aligned data sequence according to their indices, the time-related weights gradually change. Data points further down the index have a larger weight, and their impact on the first part of the calculation result is more significant. The judgment result directly adjusts the values ​​of both parts of the calculation result. Simultaneously, by correcting the magnitude of changes in the first and last data points, the overall result is adjusted accordingly based on the different judgment results. The differences between the first and last data points in the sequence are transformed into an impact on the result through a fixed calculation method. Combined with the impact of time weights, this ensures that changes in the predicted compliance threshold both follow the temporal distribution characteristics of the data and respond to the magnitude of changes in the compliance judgment result and the data itself, forming a change trend that highly aligns with the actual data situation.

[0047] The beneficial effects include: dividing quality parameter data according to the continuous construction stages to form independent and complete segmented quality data, making the data presentation more logical and targeted; combining bidirectional compliance cross-checking of overlapping area data to ensure the rigor and accuracy of the judgment results; time-domain alignment processing to eliminate data time misalignment and keep the data consistent in the time dimension; quality weight sequence highlighting the importance of compliance data through differentiated allocation; weighted average value closely matching the core level of the data; overall change rate accurately capturing the change pattern of the data in the time dimension; predicting compliance thresholds and adjusting and correcting them based on multiple factors to ensure that they are realistic and meet compliance requirements; parameter deviation markers accurately identifying data points that deviate from compliance standards; and differentiated setting of spatiotemporal weights fully considering the importance of construction stages and spatial locations. Finally, the spatiotemporal deviation obtained through integration comprehensively and accurately reflects the deviation of quality parameter data from the specifications, providing a reliable basis for subsequent key anomaly pattern analysis and other links, and helping to improve the accuracy and scientific nature of online inspection of low-voltage pipeline laying quality.

[0048] By obtaining the necessary basic information for formula calculation in a clear and direct manner, the index is allocated according to the chronological order of data points in the time-domain aligned data sequence, with each data point corresponding to a unique identifier. The number of data points is obtained by counting the total number of data points contained in the sequence. The judgment result is obtained by performing bidirectional compliance cross-validation on the overlapping areas of segmented quality data. Various quality parameter data are directly extracted from the corresponding positions in the time-domain aligned data sequence, ensuring the accuracy and reliability of the basic information used in the formula. When calculating, the formula takes into account the weight differences of the time dimension of data points, the impact of compliance judgment results, and the changing trends of the data itself. This ensures that the predicted compliance threshold of segmented quality data not only fits the overall distribution characteristics of the data but also reflects compliance requirements and data change patterns. At the same time, as the data point indexing progresses, the time-related weights change according to a pattern, and the judgment results are adjusted accordingly to the calculation results. The differences between the beginning and end of the data sequence are also incorporated into the calculation in a fixed way, making the changes in the predicted compliance threshold highly consistent with the actual data situation. Ultimately, this provides a clear and practical judgment standard for the prospective anomaly detection of subsequent quality parameter data, ensuring the accuracy and effectiveness of the compliance assessment process.

[0049] S04. Based on the spatiotemporal deviation, key features are extracted from the quality parameter data to obtain the key anomaly patterns of the quality parameter data; In this embodiment of the invention, the step of extracting key features from the quality parameter data based on the spatiotemporal deviation to obtain the key anomaly patterns of the quality parameter data includes: Based on the spatiotemporal deviation, the anomaly persistence of the quality parameter data is evaluated to obtain the process persistence coefficient of the quality parameter data; Based on the process persistence coefficient, intermittent anomaly patterns are identified in the quality parameter data to obtain candidate anomaly patterns of the quality parameter data. Based on the preset construction tolerance deviation, the candidate anomaly patterns are filtered by engineering semantic constraints to obtain the engineering anomaly patterns of the candidate anomaly patterns. The abnormal patterns in the engineering are categorized and merged to obtain the key abnormal patterns in the quality parameter data.

[0050] When assessing the persistence of anomalies in quality parameter data based on spatiotemporal deviation, a fixed assessment time interval is first determined. The quality parameter data is then evenly divided into several consecutive time periods according to this time interval. The spatiotemporal deviation of the quality parameter data is then analyzed for each time period to determine whether the spatiotemporal deviation remains within the abnormal range in each time period. The duration of each abnormal state from start to finish is recorded, and the frequency of occurrence of the abnormal state throughout the entire assessment period is also statistically analyzed. Based on a comprehensive consideration of the duration and frequency of occurrence, a specific value that reflects the persistence characteristic is assigned to each abnormal state. This value is the process persistence coefficient of the quality parameter data.

[0051] When identifying intermittent anomaly patterns in quality parameter data based on process persistence coefficients, a fixed coefficient judgment standard is first set. The process persistence coefficient of each anomaly state is compared with this standard. When the process persistence coefficient is lower than the standard, the corresponding anomaly state is determined to have intermittent characteristics. Then, these anomalies determined to be intermittent are tracked one by one, and the specific time point of each occurrence, the duration of each occurrence, and the corresponding changes in quality parameter data are recorded in detail. These anomalies with the same intermittent characteristics (such as consistent occurrence intervals and duration ranges) are organized into independent and distinguishable patterns. These patterns are the candidate anomaly patterns of quality parameter data.

[0052] When screening candidate anomaly patterns based on preset construction tolerances, the specific content of the preset construction tolerances is first clarified. This content includes the allowable range of positional deviations, angle deviations, and connection stability during the laying of low-voltage pipelines. Then, the quality parameter data corresponding to each candidate anomaly pattern is compared with each item of the construction tolerance to check whether the deviation in the candidate anomaly pattern exceeds the range specified by the construction tolerance. If the deviation of the candidate anomaly pattern is within the construction tolerance range, the candidate anomaly pattern is eliminated. If the deviation exceeds the construction tolerance range, the candidate anomaly pattern is retained. These retained candidate anomaly patterns are the engineering anomaly patterns.

[0053] When categorizing and consolidating engineering anomaly patterns, first determine the classification criteria. The classification criteria can be based on the core links of low-voltage pipeline laying (such as conduit installation, cable laying, and interface connection). Then, analyze the core characteristics of each engineering anomaly pattern, determine the laying link to which it belongs, and group engineering anomaly patterns belonging to the same laying link into one category. Next, integrate the engineering anomaly patterns in each category in detail, remove duplicate descriptions, and extract the core characteristics common to this category of anomaly patterns. Finally, each category of anomaly patterns is the key anomaly pattern of the quality parameter data.

[0054] The beneficial effects are as follows: By splitting quality parameter data into segments within a fixed evaluation time interval and analyzing the spatiotemporal deviation of each segment, and combining the duration and frequency of abnormal states to determine the process persistence coefficient, the persistence characteristics of abnormal states can be accurately captured, providing a reliable quantitative basis for subsequent anomaly pattern identification; by comparing the process persistence coefficient with the fixed coefficient judgment standard, the occurrence time, duration, and corresponding quality parameter changes of intermittent anomalies can be tracked and recorded, forming independent and distinguishable candidate anomaly patterns, which can accurately screen out anomalies with intermittent characteristics and avoid omissions or misjudgments; and the allowable ranges of positional deviation, angular deviation, and connection stability in construction tolerances are clarified, and compared with candidate anomaly patterns. By comparing quality parameter data one by one, candidate patterns with deviations within the allowable range are eliminated, while abnormal engineering patterns exceeding the standard are retained. This ensures that the retained abnormal patterns meet the actual requirements for judging violations in the project and filters out invalid interference. The classification criteria are determined according to the core links of low-voltage pipeline laying. The core characteristics of engineering abnormal patterns are analyzed and classified. Details are integrated, duplicate descriptions are removed, and common core features are extracted to form key abnormal patterns. This makes the abnormal patterns more systematic and clearly presents the core abnormal issues of different laying links. It provides accurate and clear abnormal analysis results for subsequent abnormal risk projection and quality integration, and helps to improve the accuracy and pertinence of abnormal identification in online inspection of low-voltage pipeline laying quality.

[0055] S05. Based on the connection dependencies of the design specifications, construct the logical topology diagram of the target project, and based on the logical topology diagram, perform impact deduction on the key anomaly mode to obtain the anomaly risk level of the key anomaly mode. In this embodiment of the invention, the step of constructing a logical topology diagram of the target project deployment based on the connection dependencies of the design specifications, and performing impact deduction on the key anomaly patterns based on the logical topology diagram to obtain the anomaly risk level of the key anomaly patterns, includes: Semantic dependency destructuring is performed on the connection dependencies of the design specification to obtain the connection rules of the design specification; Based on the connection rules and the location information of the key anomaly pattern, neighborhood topology induction is performed on the key anomaly pattern to obtain a local logical topology subgraph of the key anomaly pattern. Based on the local logic topology subgraph, signal flow resistance analysis is performed on the key anomaly mode to obtain the local blocking characteristics of the key anomaly mode. Based on the local blocking characteristics and the connection rules, the key dependency path is inverted to obtain the logical topology diagram of the key anomaly pattern. Based on the logical topology graph, the structural vulnerability of the key anomaly pattern is quantified to obtain the path influence weight of the key anomaly pattern. Based on the path influence weight, a multi-level threshold adaptive determination is performed on the key anomaly pattern to obtain the anomaly risk level of the key anomaly pattern.

[0056] The step of quantifying the structural vulnerability of the key anomaly patterns based on the logical topology graph to obtain the path influence weights of the key anomaly patterns includes: The node betweenness centrality and edge clustering coefficient of the logical topology graph are extracted in parallel in two dimensions to obtain the topological structural feature parameters of the logical topology graph. Based on the logical topology diagram, spatial statistical characterization is performed on the key anomaly pattern to obtain the anomaly distribution characteristic parameters of the key anomaly pattern. Based on the topological structure feature parameters and the abnormal distribution feature parameters, the vulnerability score of each edge in the logical topology graph is calculated, wherein the formula for calculating the vulnerability score is: ; In the formula, For the edges in the logical topology graph. For the edge Vulnerability score, For node variables, For the edge endpoint nodes, For nodes betweenness centrality For nodes The edge aggregation coefficient, It is an exponential function. For nodes The topological shortest distance to the key anomaly pattern. The mean of the topological distribution among the abnormal distribution characteristic parameters is... The variance of the topological distribution in the abnormal distribution characteristic parameters; The vulnerability scores are mapped using a probability distribution to obtain the path influence weights of the key anomaly patterns.

[0057] When deconstructing the semantic dependencies of the connection relationships in the design specifications, the connection requirements between the components of the low-voltage pipeline in the design specifications are sorted out one by one. The core information such as the connection sequence of each component, the necessary supporting components, the prohibited connection combinations, and the stability requirements of the connection are clarified. These sorted-out specific requirements are organized into a clear and directly referable set of rules, which is the connection rules of the design specifications.

[0058] When performing neighborhood topology guidance based on connection rules and location information of key anomaly patterns, the specific area of ​​the low-voltage pipeline where the key anomaly pattern is located is first determined according to the location information of the key anomaly pattern. Then, according to the connection rules, all pipeline components in the area that have direct connection relationships with the key anomaly pattern, as well as the connection methods between these components, are identified. Each relevant component is treated as an independent node, and the connection relationship between components is treated as the association line between nodes. A local graph containing only the relevant nodes and association lines of the neighborhood is constructed to obtain the local logical topology subgraph of the key anomaly pattern.

[0059] When performing signal flow resistance analysis based on local logical topology subgraphs, the normal transmission path of weak electrical signals within the subgraph is simulated. Each transmission path is checked one by one to see if transmission is hindered due to the presence of key abnormal modes. The specific nodes or connection locations where the obstruction occurs, the total number of affected transmission paths, and the specific manifestations of signal transmission interruption or attenuation are recorded. These recorded information are then integrated into a system to form local blocking characteristics that can reflect the impact of abnormalities on signal transmission.

[0060] When performing critical dependency path inversion based on local blockage characteristics and connection rules, starting from the obstruction location determined by the local blockage characteristics, the normal connection logic specified by the connection rules is traced backward to find all upstream dependent pipelines and downstream related pipelines that cannot transmit signals normally due to the blockage. The connection relationship and dependency order between these pipelines are clarified, and all affected pipeline components and connection relationships are presented as a complete structured graph to obtain the logical topology diagram of the critical anomaly mode.

[0061] When performing a two-dimensional parallel analysis of the logical topology graph, the betweenness centrality of nodes and the edge clustering coefficient are analyzed simultaneously. The betweenness centrality of a node is determined by counting the number of times the node appears in the shortest path between any two nodes in the graph. The edge clustering coefficient is determined by calculating the ratio of the actual number of connections between other nodes directly connected to the two endpoints of the edge to the maximum possible number of connections between these nodes. By integrating the betweenness centrality of all nodes and the clustering coefficient of all edges, the topological characteristic parameters of the logical topology graph are obtained.

[0062] When performing spatial statistical representation based on logical topology graphs, spatial information such as the number of specific locations of key anomaly patterns in the graph, the topological distance between each anomaly location, and the proportion of the number of nodes and edges covered by the anomaly pattern to the total number of nodes and edges in the entire logical topology graph are collected. These statistical results are then summarized and organized to form anomaly distribution characteristic parameters that can comprehensively reflect the spatial distribution of key anomaly patterns.

[0063] When calculating the vulnerability score of each edge based on topological structure feature parameters and anomaly distribution feature parameters, for each edge in the logical topology graph, the possibility of the edge failing or malfunctioning under the influence of the key anomaly mode is comprehensively evaluated by combining its corresponding node betweenness centrality, edge clustering coefficient, topological distance to the key anomaly mode, and overall density of the anomaly distribution. Each edge is assigned a specific value that can accurately reflect this possibility, and this value is the vulnerability score of the edge.

[0064] When mapping the vulnerability scores to a probability distribution, the vulnerability scores of all edges are collected first, the overall distribution range of these scores is determined, and the distribution range is divided into several continuous intervals. Each interval corresponds to a fixed probability value. The probability value corresponding to each vulnerability score is determined according to the interval in which it is located. Then, the probability value is converted into a weight that can reflect the degree of influence of the edge on the overall pipeline function. The weights of all edges are integrated to obtain the path influence weight of the critical anomaly pattern.

[0065] When performing multi-level threshold adaptive determination based on path influence weight, multiple different weight threshold intervals are pre-set. Each interval corresponds to a specific abnormal risk level, and there is no overlap between intervals. The path influence weight is compared with these threshold intervals one by one to determine the threshold interval to which the path influence weight belongs. The risk level corresponding to this interval is the abnormal risk level of the key abnormal pattern.

[0066] Edges are selected directly from the logical topology graph of the critical anomaly patterns, with the endpoints of each edge being the nodes connected to its two endpoints. The betweenness centrality of a node is determined by counting the number of times it appears in the shortest path between any two nodes in the logical topology graph. The edge clustering coefficient of a node is determined by the ratio of the actual number of connections between other nodes directly connected to the two endpoints of the edge to the maximum possible number of connections between these nodes. The topological shortest distance from a node to the critical anomaly pattern is obtained by finding the length of the shortest path between the node and the critical anomaly pattern in the logical topology graph. The topological distribution mean and variance are directly extracted from the anomaly distribution characteristic parameters of the critical anomaly patterns.

[0067] This formula integrates the topological structural characteristic parameters of the logical topology graph with the anomaly distribution characteristic parameters of the key anomaly patterns. By combining information such as node betweenness centrality, edge clustering coefficient, topological shortest distance from a node to the key anomaly pattern, topological distribution mean, and topological distribution variance, it comprehensively assesses the probability of each edge in the logical topology graph failing or malfunctioning under the influence of the key anomaly patterns. Finally, it obtains the vulnerability score of the edge that can accurately reflect this probability, providing a reliable basis for obtaining the subsequent path influence weights.

[0068] The higher the betweenness centrality of a node, the more prominent the corresponding computational results, and the edge vulnerability score changes accordingly. Changes in the node edge clustering coefficient adjust the amplitude of the computational results. The closer the topological shortest distance from a node to the key anomaly pattern is to the topological distribution mean, the more stable the corresponding exponential results. Changes in the topological distribution variance adjust the degree of influence of this stability. Considering the interaction of these factors, the edge vulnerability score exhibits a variation pattern that highly matches the logical topological structure characteristics and anomaly distribution characteristics, truly reflecting the vulnerability of the edges.

[0069] The beneficial effects are as follows: By systematically reviewing the connection requirements of each component of low-voltage pipelines in the design specifications, core information such as connection sequence, supporting components, prohibited combinations, and stability requirements is clarified, forming a clear and well-organized connection rule, providing a clear and referable basis for all subsequent stages; Based on the connection rule and the location information of key anomaly patterns, specific areas are identified, directly related pipeline components and connection methods are found, and a local logical topology subgraph is constructed to achieve precise focusing on key anomaly-related areas; The normal transmission path of low-voltage signals is simulated, and transmission obstructions are checked one by one, recording the obstruction location, affected path, and signal performance, and integrating them into local congestion characteristics to accurately reflect the specific impact of anomalies on signal transmission; Tracing upstream and downstream dependent and related pipelines from the obstruction location, the connection relationship and dependency sequence are clarified, and a complete logical topology diagram of affected components and connections is presented to fully understand the scope of anomaly impact; At the same time, the betweenness centrality of nodes is analyzed. By integrating edge aggregation coefficients into topological structure characteristic parameters, the system grasps the structural characteristics of the logical topology; it statistically analyzes the spatial distribution information of key anomaly patterns and summarizes it into anomaly distribution characteristic parameters, comprehensively reflecting the spatial distribution of anomalies; combining topological structure and anomaly distribution characteristic parameters, it comprehensively assesses the possibility of failure or malfunction of each edge and assigns a vulnerability score, accurately quantifying the vulnerability of edges; by dividing score intervals, matching probability values, and converting them into influence weights, it obtains path influence weights, clearly reflecting the degree of influence of each edge on the overall pipeline function; it presets non-overlapping multi-level weight threshold intervals and corresponding anomaly risk levels, and determines the interval and corresponding risk level by comparing path influence weights, clarifying the risk level of key anomaly patterns. The entire process is progressive and precisely connected at each stage, providing a comprehensive and reliable risk assessment basis for subsequent integrated quality integration, ensuring the accuracy and scientific nature of online inspection of low-voltage pipeline laying quality.

[0070] Edges and their endpoints are directly selected from the logical topology graph of the key anomaly patterns. The acquisition of node betweenness centrality, edge clustering coefficient, topological shortest distance from nodes to key anomaly patterns, and the extraction of the topological distribution mean and variance all rely closely on the anomaly distribution characteristic parameters of the logical topology graph and the key anomaly patterns, ensuring the relevance and accuracy of the underlying data used in the calculations. The formula integrates the topological structure characteristic parameters of the logical topology graph with the anomaly distribution characteristic parameters of the key anomaly patterns, comprehensively considering information such as node betweenness centrality, edge clustering coefficient, topological shortest distance, topological distribution mean, and topological distribution variance, enabling precise evaluation of each edge. The possibility of failure or malfunction under the influence of key anomaly patterns is analyzed to obtain the vulnerability score of the edge, which can reliably reflect the possibility, providing strong support for obtaining the subsequent path influence weight. At the same time, the interaction of node betweenness centrality, edge clustering coefficient, topological shortest distance and topological distribution mean and variance makes the edge vulnerability score show a change pattern that is highly consistent with the logical topological structure characteristics and anomaly distribution characteristics. It can truly reflect the vulnerability of the edge, provide accurate data basis for subsequent links such as the impact inference of key anomaly patterns and the determination of anomaly risk level, and ensure the scientificity and effectiveness of the risk assessment link in the online inspection of the quality of low-voltage pipeline laying.

[0071] S06. Based on the aforementioned abnormal risk level, the key abnormal patterns and the quality parameter data are comprehensively integrated to obtain an online inspection report of the target project's installation. In this embodiment of the invention, the step of comprehensively integrating the key anomaly patterns and the quality parameter data based on the anomaly risk level to obtain an online inspection report for the target project's installation includes: Based on the aforementioned anomaly risk level, the key anomaly patterns are dynamically prioritized to obtain a priority sequence for the key anomaly patterns. Based on the priority sequence, the quality parameter data is correlated and labeled to obtain labeled data of the quality parameter data; Based on the priority sequence, the labeled data is subjected to temporal attention weighting to obtain the time-weighted sequence of the labeled data; The time-weighted sequence is organized into an itemized structure to obtain an online inspection report of the target project's installation.

[0072] When dynamically prioritizing key anomaly patterns based on their risk levels, the risk level of each key anomaly pattern is first determined. The key anomaly patterns are then initially sorted in descending order of risk level. For key anomaly patterns at the same risk level, a secondary sort is performed based on the scope of their impact on the overall quality of low-voltage pipeline laying. Patterns with a scope of impact involving core functional areas are given priority. This results in a clear priority sequence of key anomaly patterns with a well-defined order and hierarchy.

[0073] When performing correlation annotation on quality parameter data based on priority sequences, the core feature information of each key anomaly pattern in the priority sequence is extracted one by one, including the anomaly type and the laying links involved. Then, the quality parameter data is traversed to find the data items that match the core feature information of each key anomaly pattern. The corresponding key anomaly pattern name, priority level, and core features associated with the anomaly are marked on the successfully matched data items. After all data items are marked, the labeled data of the quality parameter data is obtained.

[0074] When performing time-series attention weighting on labeled data based on priority sequences, the construction time nodes corresponding to the labeled data are first sorted out and the labeled data are arranged in chronological order. Then, the weight of each labeled data is determined according to the priority sequence. The higher the priority of the key anomaly pattern, the greater the weight of the labeled data associated with it. The specific content of each labeled data is combined with the corresponding weight. At the same time, special attention is paid to the labeled data that is close to the current inspection stage in time to ensure that its weight role is fully reflected, and finally a time-weighted sequence of labeled data is formed.

[0075] When organizing the time-weighted sequence into structured items, each data item in the time-weighted sequence is classified and integrated in descending order of priority. Each item clearly includes the name of the key anomaly pattern, priority level, details of the associated quality parameters, weighting result, time node of the anomaly, and description of the scope of impact. A unified item format and layout specification are adopted, and all integrated content is presented in modules and chapters to ensure that the overall structure is clear, the information is complete, and it is easy to look up, ultimately resulting in an online inspection report of the target project's installation.

[0076] The beneficial effects include: prioritizing key anomaly patterns according to their risk level and the scope of their association with core functional areas, forming a clear and hierarchical priority sequence that provides clear guidance for subsequent processing; extracting the core features of key anomaly patterns and matching them with quality parameter data to accurately associate the anomaly information with priority, clarifying the correspondence between data and anomalies; sorting the labeled data by construction time, allocating weights according to priority and highlighting the weight of recent data, resulting in a time-weighted sequence that reflects both priority differences and timeliness; and classifying and integrating the time-weighted sequence items by priority, clarifying the core information of each item and presenting them in modules using a unified format, ultimately yielding a clear, complete, and easy-to-read online inspection report. This provides an intuitive and reliable basis for the quality assessment and rectification decisions of low-voltage pipeline laying, improving the practicality and efficiency of quality inspection.

[0077] like Figure 2 The diagram shown is a functional module diagram of an online inspection system for the quality of low-voltage pipeline laying provided in an embodiment of the present invention.

[0078] The online inspection system 10 for the quality of low-voltage cable laying described in this invention can be installed in an electronic device. Depending on the functions implemented, the online inspection system 10 may include a quality parameterization and analysis module 11, a spatiotemporal reference alignment module 12, a compliance dynamic judgment module 13, an anomaly feature extraction module 14, a topology risk assessment module 15, and a comprehensive report generation module 16. The modules described in this invention can also be referred to as units, which are a series of computer program segments that can be executed by the processor of an electronic device and perform a fixed function, stored in the memory of the electronic device.

[0079] In this embodiment, the functions of each module / unit are as follows: The quality parameterization analysis module 11 is used to perform structured information analysis on the real-time construction status of the target project to obtain the quality parameter data of the real-time construction status. The spatiotemporal reference alignment module 12 is used to perform spatiotemporal coordinate mapping between the quality parameter data and the design specifications for the laying of the target project, so as to obtain the dynamic anchoring relationship of the quality parameter data. The compliance dynamic judgment module 13 is used to perform continuous frame compliance evaluation on the quality parameter data based on the dynamic anchoring relationship, and obtain the spatiotemporal deviation of the quality parameter data. The anomaly feature extraction module 14 is used to extract key features from the quality parameter data based on the spatiotemporal deviation to obtain the key anomaly patterns of the quality parameter data. The topology risk assessment module 15 is used to construct a logical topology diagram of the target project based on the connection dependencies of the design specifications, and to perform impact deduction on the key anomaly mode based on the logical topology diagram to obtain the anomaly risk level of the key anomaly mode. The comprehensive report generation module 16 is used to perform comprehensive quality integration of the key anomaly patterns and the quality parameter data based on the anomaly risk level to obtain an online inspection report of the target project's installation.

[0080] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0081] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0082] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0083] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0084] This application embodiment can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0085] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for online inspection of the quality of low-voltage wiring installation, characterized in that, The method includes: S01. Perform structured information parsing on the real-time construction status of the target project to obtain the quality parameter data of the real-time construction status. S02. Perform spatiotemporal coordinate mapping between the quality parameter data and the design specifications for the target project to obtain the dynamic anchoring relationship of the quality parameter data; S03. Based on the dynamic anchoring relationship, perform continuous frame compliance assessment on the quality parameter data to obtain the spatiotemporal deviation of the quality parameter data; S04. Based on the spatiotemporal deviation, key features are extracted from the quality parameter data to obtain the key anomaly patterns of the quality parameter data; S05. Based on the connection dependencies of the design specifications, construct the logical topology diagram of the target project, and based on the logical topology diagram, perform impact deduction on the key anomaly mode to obtain the anomaly risk level of the key anomaly mode. S06. Based on the aforementioned abnormal risk level, the key abnormal patterns and the quality parameter data are comprehensively integrated to obtain an online inspection report of the target project's installation.

2. The method for online inspection of the quality of low-voltage wiring as described in claim 1, characterized in that, The structured information parsing of the real-time construction status of the target project yields quality parameter data for the real-time construction status, including: Multi-source status acquisition is performed on the real-time construction status of the target project to obtain the status data of the real-time construction status. The state data is time-series aligned, and redundant information is cleaned from the aligned data to obtain a regular state sequence of the real-time construction state. The regular state sequence is deconstructed in a multidimensional way to obtain the quality feature sequence of the regular state sequence. The quality feature sequence is normalized to obtain the quality parameter data of the real-time construction status.

3. The method for online inspection of the quality of low-voltage wiring as described in claim 1, characterized in that, The step of mapping the quality parameter data to the design specifications for the target project's laying in a spatiotemporal coordinate system to obtain the dynamic anchoring relationship of the quality parameter data includes: The design specifications for the target project are divided into spatiotemporal grids to obtain the reference spatiotemporal grid for the design specifications. Based on the reference spatiotemporal grid, the matching degree of the quality parameter data is verified to obtain the spatiotemporal deviation of the quality parameter data; Based on the spatiotemporal deviation, the mass parameter data is interpolated and fitted to obtain the continuous spatiotemporal trajectory of the mass parameter data; Based on the construction load data of the target project, the continuous spatiotemporal trajectory is subjected to load compensation correction to obtain the corrected spatiotemporal trajectory of the continuous spatiotemporal trajectory. The corrected spatiotemporal trajectory is associated and bound point by point with the reference spatiotemporal grid to obtain the dynamic anchoring relationship of the quality parameter data.

4. The method for online inspection of the quality of low-voltage wiring as described in claim 1, characterized in that, The step of performing continuous frame compliance assessment on the quality parameter data based on the dynamic anchoring relationship to obtain the spatiotemporal deviation of the quality parameter data includes: Based on the dynamic anchoring relationship, the quality parameter data is segmented for process consistency to obtain segmented quality data of the quality parameter data. A two-way compliance cross-validation is performed on the overlapping areas of the segmented quality data to obtain the judgment result of the segmented quality data. Based on the determination result, the segmented quality data is extrapolated within the segment to obtain the predicted compliance threshold of the segmented quality data. Based on the predicted compliance threshold, a forward-looking anomaly detection is performed on the quality parameter data to obtain parameter deviation markers for the quality parameter data; The spatiotemporal weighted integral of the parameter deviation marker is performed to obtain the spatiotemporal deviation of the quality parameter data.

5. The method for online inspection of the quality of low-voltage wiring as described in claim 4, characterized in that, Based on the determination result, the segmented quality data is extrapolated intra-segmentally to obtain the predicted compliance threshold of the segmented quality data, including: The segmented quality data is time-ordered to obtain a time-domain aligned data sequence of the segmented quality data. Based on the determination result, a nonlinear weight allocation is performed on the time-domain aligned data sequence to obtain the quality weight sequence of the time-domain aligned data sequence. Based on the quality weight sequence, the time-domain aligned data sequence is synthesized by weighted average to obtain the weighted average value of the time-domain aligned data sequence; The trend change rate of the time-domain aligned data sequence is analyzed to obtain the overall change rate of the time-domain aligned data sequence. Based on the weighted average, the overall rate of change, and the judgment result, the predicted compliance threshold for the segmented quality data is calculated, wherein the formula for calculating the predicted compliance threshold is: ； In the formula, The predicted compliance threshold, The index of the data point in the time-domain aligned data sequence. The number of data points in the time-domain aligned data sequence. The determination result is as follows. For index The corresponding quality parameter data, The quality parameter data for the last data point in the time-domain aligned data sequence. The quality parameter data is for the first data point in the time-domain aligned data sequence.

6. The method for online inspection of the quality of low-voltage pipeline laying as described in claim 1, characterized in that, The step of extracting key features from the quality parameter data based on the spatiotemporal deviation to obtain key anomaly patterns in the quality parameter data includes: Based on the spatiotemporal deviation, the anomaly persistence of the quality parameter data is evaluated to obtain the process persistence coefficient of the quality parameter data; Based on the process persistence coefficient, intermittent anomaly patterns are identified in the quality parameter data to obtain candidate anomaly patterns of the quality parameter data. Based on the preset construction tolerance deviation, the candidate anomaly patterns are filtered by engineering semantic constraints to obtain the engineering anomaly patterns of the candidate anomaly patterns. The abnormal patterns in the engineering are categorized and merged to obtain the key abnormal patterns in the quality parameter data.

7. The method for online inspection of the quality of low-voltage wiring as described in claim 1, characterized in that, Based on the connection dependencies of the design specifications, a logical topology diagram of the target project is constructed. Based on this logical topology diagram, the impact of the key anomaly patterns is analyzed to obtain the anomaly risk level of each key anomaly pattern, including: Semantic dependency destructuring is performed on the connection dependencies of the design specification to obtain the connection rules of the design specification; Based on the connection rules and the location information of the key anomaly pattern, neighborhood topology induction is performed on the key anomaly pattern to obtain a local logical topology subgraph of the key anomaly pattern. Based on the local logic topology subgraph, signal flow resistance analysis is performed on the key anomaly mode to obtain the local blocking characteristics of the key anomaly mode. Based on the local blocking characteristics and the connection rules, the key dependency path is inverted to obtain the logical topology diagram of the key anomaly pattern. Based on the logical topology graph, the structural vulnerability of the key anomaly pattern is quantified to obtain the path influence weight of the key anomaly pattern. Based on the path influence weight, a multi-level threshold adaptive determination is performed on the key anomaly pattern to obtain the anomaly risk level of the key anomaly pattern.

8. The method for online inspection of the quality of low-voltage wiring as described in claim 7, characterized in that, The step of quantifying the structural vulnerability of the key anomaly patterns based on the logical topology graph to obtain the path influence weights of the key anomaly patterns includes: The node betweenness centrality and edge clustering coefficient of the logical topology graph are extracted in parallel in two dimensions to obtain the topological structural feature parameters of the logical topology graph. Based on the logical topology diagram, spatial statistical characterization is performed on the key anomaly pattern to obtain the anomaly distribution characteristic parameters of the key anomaly pattern. Based on the topological structure feature parameters and the abnormal distribution feature parameters, the vulnerability score of each edge in the logical topology graph is calculated, wherein the formula for calculating the vulnerability score is: ； In the formula, For the edges in the logical topology graph. For the edge Vulnerability score, For node variables, For the edge endpoint nodes, For nodes betweenness centrality For nodes The edge aggregation coefficient, It is an exponential function. For nodes The topological shortest distance to the key anomaly pattern. The mean of the topological distribution among the abnormal distribution characteristic parameters is... The variance of the topological distribution in the abnormal distribution characteristic parameters; The vulnerability scores are mapped using a probability distribution to obtain the path influence weights of the key anomaly patterns.

9. The method for online inspection of the quality of low-voltage wiring as described in claim 1, characterized in that, Based on the anomaly risk level, the key anomaly patterns and the quality parameter data are comprehensively integrated to obtain an online inspection report for the target project's installation, including: Based on the aforementioned anomaly risk level, the key anomaly patterns are dynamically prioritized to obtain a priority sequence for the key anomaly patterns. Based on the priority sequence, the quality parameter data is correlated and labeled to obtain labeled data of the quality parameter data; Based on the priority sequence, the labeled data is subjected to temporal attention weighting to obtain the time-weighted sequence of the labeled data; The time-weighted sequence is organized into an itemized structure to obtain an online inspection report of the target project's installation.

10. An online inspection system for the quality of low-voltage cable laying, characterized in that, The system for implementing the online inspection method for the quality of low-voltage pipeline laying as described in claim 1 includes: The quality parameterization analysis module is used to perform structured information analysis on the real-time construction status of the target project to obtain the quality parameter data of the real-time construction status. The spatiotemporal reference alignment module is used to perform spatiotemporal coordinate mapping between the quality parameter data and the design specifications for the laying of the target project, so as to obtain the dynamic anchoring relationship of the quality parameter data. The compliance dynamic judgment module is used to perform continuous frame compliance evaluation on the quality parameter data based on the dynamic anchoring relationship, and obtain the spatiotemporal deviation of the quality parameter data. An anomaly feature extraction module is used to extract key features from the quality parameter data based on the spatiotemporal deviation to obtain the key anomaly patterns of the quality parameter data. The topology risk assessment module is used to construct a logical topology diagram of the target project based on the connection dependencies of the design specifications, and to perform impact deduction on the key anomaly modes based on the logical topology diagram to obtain the anomaly risk level of the key anomaly modes. The comprehensive report generation module is used to perform comprehensive quality integration of the key anomaly patterns and the quality parameter data based on the anomaly risk level, and obtain an online inspection report of the target project's installation.

Images