Pipe joint sealing performance monitoring method and device and medium

Abstract

The invention discloses a pipe joint sealing performance monitoring method and device and a medium, and relates to the technical field related to pipe joints, and the method comprises the steps: carrying out the dynamic sensing of a connecting area of a pipe joint, obtaining a first sensing data set, and carrying out the dynamic sensing of a connecting area of a connecting piece, and obtaining a second sensing data set; carrying out deviation calculation and carrying out leakage analysis on the pipe joint according to a deviation value; mapping the first leakage feature vector group to a pipe joint three-dimensional structure model to generate a second leakage feature vector group; leakage repositioning is carried out, a leakage risk alarm signal is constructed and sent to a remote monitoring terminal to carry out sealing diagnosis on the pipe joint, and a sealing diagnosis report of the pipe joint is generated. The technical problems that in the prior art, due to the fact that the anti-interference capacity is insufficient and multi-source data fusion analysis is insufficient, the real-time performance of pipe joint sealing performance monitoring is poor, and the leakage positioning precision is low are solved, and the technical effects of high-sensitivity and high-precision leakage detection, dynamic real-time monitoring and intelligent sealing diagnosis are achieved.

Description

Technical Field

[0001] This application relates to the technical field of pipe fittings, and in particular to methods, equipment and media for monitoring the sealing performance of pipe fittings. Background Technology

[0002] The sealing performance of pipe joints directly affects the safety, reliability, and efficiency of fluid transportation. Because pipeline systems often operate under complex conditions such as high pressure, high temperature, or corrosive media, pipe joints are prone to leakage due to mechanical vibration, material aging, or improper installation, leading to safety hazards, environmental pollution, or energy waste. Existing pipe joint sealing monitoring systems are insufficient to comprehensively characterize leakage features, cannot dynamically monitor minute leaks, and lack robust anti-interference capabilities under complex conditions, multi-source data fusion analysis, and spatial mapping accuracy of leakage characteristics. This results in poor real-time performance, location accuracy, and automation levels in pipe joint sealing monitoring and early warning, hindering early warning and precise leak location.

[0003] Current technologies suffer from insufficient anti-interference capabilities and inadequate multi-source data fusion analysis, resulting in poor real-time monitoring of pipe joint sealing and low accuracy in leak location. Summary of the Invention

[0004] This application provides a method, equipment, and medium for monitoring the sealing performance of pipe joints, which solves the technical problems of insufficient anti-interference ability and inadequate multi-source data fusion analysis in the prior art, resulting in poor real-time monitoring and low leakage location accuracy of pipe joint sealing performance. It achieves the technical effects of high sensitivity, high precision leakage detection, dynamic real-time monitoring, and intelligent sealing diagnosis.

[0005] This application provides a method for monitoring the sealing performance of pipe joints. The method includes: dynamically sensing the connection area of ​​the pipe joint to obtain a first sensing dataset; dynamically sensing the connection area of ​​the connector to obtain a second sensing dataset; calculating the deviation between the first sensing dataset and the second sensing dataset; performing leakage analysis on the pipe joint based on the deviation value to construct a first leakage feature vector group; mapping the first leakage feature vector group to a three-dimensional structural model of the pipe joint; determining multiple leakage coordinate points for spatial verification; performing inversion based on the verification results and the deviation value; updating the first leakage feature vector group based on the inversion results to generate a second leakage feature vector group; relocating the leakage based on the second leakage feature vector group; constructing a leakage risk alarm signal and sending it to a remote monitoring terminal for sealing diagnosis of the pipe joint; and generating a sealing performance diagnosis report for the pipe joint.

[0006] In one possible implementation, the pipe joint sealing performance monitoring method is further configured to perform the following processing: retrieve a first historical area data change log of the pipe joint sealing connection area, perform change frequency analysis based on the first historical area data change log, and determine a first acquisition cycle; retrieve a second historical area data change log of the connector sealing connection area, perform change frequency analysis based on the second historical area data change log, and determine a second acquisition cycle; perform pressure change sensing acquisition on the sealing connection area of ​​the pipe joint according to the first acquisition cycle to obtain first pressure fluctuation data, and add the first pressure fluctuation data to the first sensor dataset; perform pressure change sensing acquisition on the connection area of ​​the connector according to the second acquisition cycle to obtain second pressure fluctuation data, and add the second pressure fluctuation data to the second sensor dataset.

[0007] In one possible implementation, the pipe joint sealing monitoring method is further configured to perform the following processing: performing spatiotemporal alignment based on the first sensor dataset and the second sensor dataset to construct a joint sensing matrix; calculating pressure field deviations on the first pressure fluctuation data and the second pressure fluctuation data based on the joint sensing matrix to construct a pressure deviation topology map; calculating deformation field deviations based on the joint sensing matrix to construct a deformation mismatch topology map; dynamically aligning and fusing the pressure deviation topology map and the deformation mismatch topology map in spatiotemporal alignment to construct a dynamic deviation field; performing anomaly analysis based on the dynamic deviation field to determine anomaly deviation patterns; and performing leakage feature analysis through the anomaly deviation patterns to generate the first leakage feature vector group.

[0008] In one possible implementation, the pipe joint sealing monitoring method is further configured to perform the following processing: perform coupling anomaly identification based on the dynamic deviation field to determine a first abnormal event feature; perform maximum value analysis based on the pressure deviation topology map and the deformation mismatch topology map to extract the pressure deviation peak value and the deformation mismatch peak value; perform phase lag calculation based on the pressure deviation peak value and the deformation mismatch peak value to determine a second abnormal event feature; normalize the first abnormal event feature and the second abnormal event feature, and perform leakage feature matching by traversing the abnormal deviation pattern according to the normalization result to construct the first leakage feature vector group.

[0009] In one possible implementation, the pipe joint sealing monitoring method is further configured to perform the following processing: introducing a three-dimensional structural model of the pipe joint, and discretizing the three-dimensional structural model of the pipe joint into multiple voxel grids; performing leakage impact analysis based on the first leakage feature vector group, generating an impact factor, and assigning weights to the first leakage feature vector group according to the impact factor to determine multiple weight coefficients; mapping the first leakage feature vector group to the multiple voxel grids to determine the multiple leakage coordinate points; superimposing the multiple leakage coordinate points with the multiple weight coefficients to construct an initial leakage probability distribution map; traversing the initial leakage probability distribution map to perform multimodal spatial verification, generating a physical leakage evidence dataset, and adding the physical leakage evidence dataset to the verification result.

[0010] In one possible implementation, the pipe joint sealing monitoring method is further configured to perform the following processing: traversing the initial leakage probability distribution map to determine the leakage probability of the multiple voxel grids and identify multiple probability peak regions; locating based on the multiple probability peak regions to identify multiple high-risk regions; determining whether the multiple high-risk regions are adjacent continuous regions; if the multiple high-risk regions are adjacent continuous regions, extracting the continuous high-position regions and performing morphological dilation processing to generate a target cluster to be verified; calculating the centroid based on the target cluster to be verified, generating centroid coordinates as priority verification points, and performing multimodal spatial verification based on the priority verification points to generate the physical leakage evidence dataset.

[0011] In one possible implementation, the pipe joint sealing monitoring method further includes performing the following processes: extracting multi-level verification tags based on the second leakage feature vector group, relocating the leakage according to the multi-level verification tags, and determining the target leakage coordinate set; connecting the leakage trajectories according to the target leakage coordinate set to construct a leakage path topology map; retrieving the operating parameters of the pipe joint, mapping the operating parameters to the leakage path topology map for change analysis, and generating change trend results; performing a decreasing analysis according to the change trend results to identify the main leakage channel and the seal failure point on the leakage path topology map; performing leakage risk analysis on the pipe joint based on the main leakage channel to construct a first leakage risk level; performing leakage risk analysis on the pipe joint based on the seal failure point to construct a second leakage risk level; performing overlapping leakage risk analysis on the pipe joint based on the main leakage channel and the seal failure point to construct a third leakage risk level; and adding the first leakage risk level, the second leakage risk level, and the third leakage risk level to the leakage risk alarm signal.

[0012] In one possible implementation, the pipe joint sealing monitoring method is further configured to perform the following processing: sending the leakage risk alarm signal to a remote monitoring terminal to mark the leakage in the three-dimensional structural model of the pipe joint based on the first leakage risk level, the second leakage risk level, and the third leakage risk level, and constructing a leakage location map; performing sealing failure diagnosis based on the leakage location map to determine the root cause data of the failure; and performing a sealing safety assessment of the pipe joint based on the root cause data of the failure to generate a sealing performance diagnosis report for the pipe joint.

[0013] This application also provides an electronic device, including: a memory for storing executable instructions; and a processor for implementing a pipe joint sealing monitoring method when executing the executable instructions stored in the memory.

[0014] This application also provides a computer-readable storage medium, including: a computer program stored thereon, which, when executed by a processor, implements a method for monitoring the sealing performance of pipe joints.

[0015] This application proposes a pipe joint sealing monitoring method, equipment, and medium. The method involves dynamically sensing the connection area of ​​the pipe joint to obtain a first sensing dataset, and then dynamically sensing the connection area of ​​the connector to obtain a second sensing dataset. Deviation calculations are performed, and leakage analysis is conducted on the pipe joint based on the deviation values. The first leakage feature vector set is mapped to the three-dimensional structural model of the pipe joint to generate a second leakage feature vector set. Leakage relocation is performed, and a leakage risk alarm signal is constructed and sent to a remote monitoring terminal for sealing diagnosis of the pipe joint, generating a sealing diagnosis report. This method solves the technical problems of insufficient anti-interference capability and inadequate multi-source data fusion analysis in existing technologies, which lead to poor real-time monitoring and low leakage location accuracy in pipe joint sealing performance. It achieves high-sensitivity, high-precision leakage detection, dynamic real-time monitoring, and intelligent sealing diagnosis. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings of the embodiments of the present invention will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0017] Figure 1 This is a schematic flowchart of the pipe joint sealing monitoring method provided in the embodiments of this application.

[0018] Figure 2This is a schematic flowchart illustrating the leakage analysis process in the pipe joint sealing monitoring method provided in this application embodiment.

[0019] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0020] Explanation of reference numerals in the attached drawings: Input device 401, processor 402, memory 403, output device 404. Detailed Implementation

[0021] The above description is merely an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below.

[0022] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application will be provided in conjunction with the accompanying drawings. The described embodiments should not be considered as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0023] In the following description, references to "some embodiments" describe a subset of all possible embodiments. However, it is understood that "some embodiments" can be the same or different subsets of all possible embodiments and can be combined with each other without conflict. The terms "first" and "second" are used merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, product, or server that includes a series of steps is not necessarily limited to those steps explicitly listed, but may include other steps not explicitly listed or inherent to such processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only.

[0024] This application provides a method for monitoring the sealing performance of pipe joints, such as... Figure 1 As shown, the method includes: Step S100: Dynamically sense the connection area of ​​the pipe fitting to obtain a first sensing dataset, and dynamically sense the connection area of ​​the connector to obtain a second sensing dataset.

[0025] Preferably, pipe fittings are components used to connect fluid or gas transmission pipelines. Different structures enable pipeline connections, diversions, or mergings, ensuring system sealing and stability. Pipe fittings include connecting, turning, diverting, or merging pipes, facilitating the transfer of fluid / gas between different pipelines. For example, in a hydraulic system, they connect pipelines and components; or in a water pipe system, they change the direction of water flow or connect pipes of different diameters. The connection area of ​​a pipe fitting refers to the physical contact surface between the pipe and flanges, threaded joints, clamps, etc., and its adjacent area, typically sealing rings, gaskets, or welded areas, which are prone to leakage due to mechanical stress, corrosion, or aging. The connection area of ​​a connector refers to the component that mates with the pipe fitting, such as the corresponding contact surface of another section of pipe, valve, or equipment interface; loosening or deformation of this area may affect the overall sealing performance.

[0026] Preferably, dynamic sensing is performed on the connection areas of the pipe fitting and the connector. Dynamic sensing refers to real-time, continuous monitoring of parameter changes in the connection area. This typically involves detecting abnormal vibration signals using vibration sensors; capturing high-frequency sound waves during leakage using acoustic emission sensors, such as ultrasonic waves from gas leaks or noise from liquid leaks; monitoring deformation or stress distribution of the joint due to pressure or temperature changes using strain / stress sensors; identifying localized temperature anomalies caused by medium leakage using temperature sensors, such as heat absorption during gas leaks or frictional heat generation; and monitoring strain, temperature, or acoustic signals using distributed optical fibers, covering a large connection area. This results in a first sensing dataset and a second sensing dataset. The first sensing dataset contains dynamic signals from the pipe fitting side, such as vibration spectra, acoustic emission energy, and strain changes; the second sensing dataset contains synchronous monitoring data from the connector side, used for comparative analysis to identify abnormal deviations between the two datasets and improve the reliability of leak detection.

[0027] Furthermore, step S200 also includes step S210, retrieving the first historical area data change log of the pipe joint sealing connection area, performing change frequency analysis based on the first historical area data change log, and determining the first acquisition period; step S220, retrieving the second historical area data change log of the connector sealing connection area, performing change frequency analysis based on the second historical area data change log, and determining the second acquisition period; step S230, performing pressure change sensing acquisition on the sealing connection area of ​​the pipe joint according to the first acquisition period, obtaining first pressure fluctuation data, and adding the first pressure fluctuation data to the first sensor dataset; step S240, performing pressure change sensing acquisition on the connection area of ​​the connector according to the second acquisition period, obtaining second pressure fluctuation data, and adding the second pressure fluctuation data to the second sensor dataset.

[0028] Preferably, the sensor acquisition cycle is dynamically adjusted based on historical data to optimize leakage monitoring efficiency and enhance the accuracy of leakage characteristic analysis using pressure fluctuation data. Specifically, the first and second historical data change logs for the pipe joint sealing connection area and the connector sealing connection area are retrieved respectively. The first historical data change log records the time-series changes in parameters such as pressure, vibration, and temperature in the pipe joint sealing connection area over a period of time, while the second historical data change log records similar historical data for the corresponding sealing area of ​​the connector. Then, the change frequency analysis is performed on the first and second historical data change logs respectively, that is, the fluctuation period and frequency of abnormal peaks of parameters in the historical data are statistically analyzed to determine the typical change pattern of the area. For example, if a pipe joint experiences periodic pressure fluctuations during the high-pressure transmission phase every day, the acquisition frequency needs to be increased during that period. If a connector rarely shows fluctuations in the historical data, the acquisition frequency can be reduced to save resources. Then, for the pipe joint area, the sampling interval is dynamically set according to its historical fluctuation characteristics, that is, the first acquisition cycle is determined. For the connector area, a sampling frequency matching its historical behavior is independently set, that is, the second acquisition cycle is determined. To avoid data redundancy or missed detections caused by a fixed sampling frequency, the system adapts to different operating conditions of pipe fittings and improves monitoring efficiency.

[0029] Preferably, pressure change sensing is performed on the sealing connection area of ​​the pipe joint according to the first acquisition cycle to obtain the transient pressure value as the first pressure fluctuation data. Pressure change sensing is then performed on the connection area of ​​the connector according to the second acquisition cycle to obtain the synchronous pressure data of the connector area as the second pressure fluctuation data. The first and second pressure fluctuation data are then added to the first and second sensor datasets, respectively, together with vibration, temperature, and other sensor signals to form multi-dimensional monitoring data, including temperature field distribution and acoustic emission signals. Through dynamic adaptive data acquisition, the accuracy and response speed of sealing performance monitoring are significantly improved.

[0030] Step S200: Calculate the deviation between the first sensor dataset and the second sensor dataset, perform leakage analysis on the pipe joint based on the deviation value, and construct a first leakage feature vector group.

[0031] Furthermore, such as Figure 2As shown, step S200 further includes step S210, performing spatiotemporal alignment based on the first sensor dataset and the second sensor dataset to construct a joint sensing matrix; step S220, calculating pressure field deviation on the first pressure fluctuation data and the second pressure fluctuation data based on the joint sensing matrix to construct a pressure deviation topology map; step S230, calculating deformation field deviation based on the joint sensing matrix to construct a deformation mismatch topology map; step S240, dynamically aligning and fusing the pressure deviation topology map and the deformation mismatch topology map to construct a dynamic deviation field; step S250, performing anomaly analysis based on the dynamic deviation field to determine anomaly deviation patterns, performing leakage feature analysis through the anomaly deviation patterns, and generating the first leakage feature vector group.

[0032] Preferably, a dynamic deviation field is constructed through spatiotemporal alignment and joint analysis of multi-source sensor data to accurately identify leakage characteristics and improve the reliability and accuracy of sealing diagnosis. Specifically, spatiotemporal alignment is performed on the first and second sensor datasets, including time alignment and spatial alignment. The sensor data of pipe fittings and connectors may come from different sampling devices, and the time axis is aligned through interpolation or synchronization algorithms. The monitoring points of pipe fittings and connectors may be distributed in different locations, and they are unified to the same reference coordinates through coordinate mapping. Then, a joint sensing matrix is ​​constructed with the first and second sensor datasets as inputs. Each row of the matrix represents a time point, and each column represents a sensing parameter, such as pipe fitting / connector pressure, pipe fitting / connector vibration, pipe fitting / connector strain, etc.

[0033] Preferably, pressure field deviation is calculated based on the joint sensing matrix for the first and second pressure fluctuation data. Specifically, for each time point, the difference between the pipe joint pressure and the connector pressure is calculated. If the pipe joint is well-sealed, theoretically the pressure field deviation is 0, indicating that the pressure at both ends is balanced. If leakage exists, the pressure field deviation exhibits a specific fluctuation pattern, such as a continuous negative deviation indicating internal leakage, and periodic oscillations indicating external leakage. Then, the monitoring points of the pipe joint and connector are mapped onto a three-dimensional model, and different colors and contour lines are used to represent the magnitude of the pressure field deviation, determining the pressure deviation topology map, such as red representing high deviation and blue representing low deviation. This displays the trend of pressure deviation at different locations over time, intuitively showing the spatial location where leakage may occur and distinguishing between normal pressure fluctuations and actual leakage signals.

[0034] Preferably, deformation field deviation is calculated based on a joint sensing matrix, i.e., the mismatch between the deformation of the pipe joint and the deformation of the connecting parts is calculated. When the seal fails, the deformation mismatch will increase abnormally, such as when loose flange bolts cause uneven local deformation. Then, a deformation mismatch topology map is constructed based on the deformation field deviation calculation results to show the spatial distribution of the deformation mismatch, so as to identify potential leakage risk points caused by uneven mechanical stress. Next, the pressure deviation topology map and the deformation mismatch topology map are dynamically spatiotemporally aligned and fused, i.e., weighted fusion is performed in time and space to generate a comprehensive dynamic deviation field. Then, anomaly analysis is performed based on the dynamic deviation field to determine the abnormal deviation mode, including distinguishing between normal and abnormal modes based on machine learning classification models such as SVM and random forest, and then determining the abnormal deviation mode. Among them, when the medium leaks into the pipeline, it causes a continuous negative pressure deviation, which is internal leakage; when the medium leaks into the external environment, it causes periodic oscillations, which is external leakage; when the bolts are loose, it causes a concentrated area of ​​deformation mismatch, which is mechanical connection failure. Finally, the abnormal patterns are encoded into feature vectors to form the first leakage feature vector group, which is used for subsequent leakage type identification, risk assessment or predictive maintenance.

[0035] Furthermore, step S250 also includes step S251, performing coupling anomaly identification based on the dynamic deviation field to determine the first abnormal event feature; step S252, performing maximum value analysis based on the pressure deviation topology map and the deformation mismatch topology map to extract the pressure deviation peak value and the deformation mismatch peak value; step S253, performing phase lag calculation based on the pressure deviation peak value and the deformation mismatch peak value to determine the second abnormal event feature; step S254, normalizing the first abnormal event feature and the second abnormal event feature, and performing leakage feature matching by traversing the abnormal deviation pattern according to the normalization result to construct the first leakage feature vector group.

[0036] Preferably, a spatiotemporal clustering algorithm or deep learning model is used to couple anomaly identification of the dynamic deviation field. This includes using DBSCAN to analyze the evolution pattern of the dynamic deviation field and identify typical characteristics of abnormal events, such as a sudden drop in pressure and a sudden increase in deformation, indicating a ruptured seal; and a slow decrease in pressure and a continuous increase in deformation, indicating loose bolts. Key parameters of the abnormal events are then extracted to form a structured feature vector, thus determining the first abnormal event characteristic. Maximum value analysis is performed based on the pressure deviation topology map and the deformation mismatch topology map. Specifically, local maxima are extracted from the pressure deviation topology map as the pressure deviation peak value, and local maxima are extracted from the deformation mismatch topology map as the deformation mismatch peak value. These pressure deviation peak values ​​and deformation mismatch peak values ​​are used as strong indicators of leakage. Phase lag calculation is performed based on the pressure deviation peak value and the deformation mismatch peak value, i.e., the time difference between the pressure peak value and the deformation peak value is calculated. If the pressure change leads the deformation, it may be an internal leak; if the deformation leads the pressure change, it may be an external leak. The phase relationship is then quantified as an abnormal feature, and the second abnormal event characteristic is determined. Next, the features of the first and second abnormal events are normalized, that is, the dynamic deviation field anomaly and the peak phase relationship are standardized to ensure the comparability of data from different sensors. Then, based on the normalization results, the abnormal deviation patterns are traversed to match leakage features, that is, the closest leakage type is matched by similarity calculation such as Euclidean distance. Finally, the multi-dimensional features are integrated to form a first leakage feature vector group that can be used for diagnosis, ensuring high-precision and low-false-alarm seal failure diagnosis in complex industrial environments.

[0037] Step S300: Map the first leakage feature vector group to the three-dimensional structural model of the pipe joint, determine multiple leakage coordinate points for spatial verification, perform inversion based on the verification results and deviation values, update the first leakage feature vector group based on the inversion results, and generate the second leakage feature vector group.

[0038] Step S300 further includes step S310, introducing a three-dimensional structural model of the pipe joint, and discretizing the three-dimensional structural model of the pipe joint into multiple voxel meshes; step S310, performing leakage impact analysis based on the first leakage feature vector group, generating an impact factor, and assigning weights to the first leakage feature vector group according to the impact factor to determine multiple weight coefficients; step S320, mapping the first leakage feature vector group to the multiple voxel meshes to determine the multiple leakage coordinate points; step S330, superimposing the multiple leakage coordinate points with the multiple weight coefficients to construct an initial leakage probability distribution map; step S340, traversing the initial leakage probability distribution map to perform multimodal spatial verification, generating a physical leakage evidence dataset, and adding the physical leakage evidence dataset to the verification result.

[0039] Preferably, the first leakage feature vector set is mapped to the three-dimensional structural model of the pipe joint, and multiple leakage coordinate points are determined for spatial verification. Through three-dimensional voxel modeling and multimodal spatial verification, the abstract leakage feature vector is transformed into a visualized leakage probability distribution, and the credibility of the diagnosis is improved by combining physical evidence. Specifically, a three-dimensional structural model of the pipe joint is introduced, and the CAD model or three-dimensional scanning data of the pipe joint is divided into tiny cubic units, i.e., voxel meshes, such as 1mm×1mm×1mm meshes. Each voxel contains initial attributes such as spatial coordinates, material type, and stress distribution. Leakage impact analysis is performed based on the first leakage feature vector set, that is, analyzing its potential impact on different areas. For example, high-frequency vibration characteristics indicate a higher weight for the bolted connection area, and pressure attenuation characteristics indicate a higher weight for the gasket area. Then, influence factors, such as values ​​from 0 to 1, are generated through expert experience or machine learning models. The weight allocation of the first leakage feature vector set is then dynamically adjusted according to the influence factors to obtain multiple weight coefficients.

[0040] Preferably, the first leakage feature vector group is mapped to multiple voxel grids, including vibration-dominant features mapped to voxels around bolt holes and pressure-dominant features mapped to voxels on the sealing surface, so as to associate the weighted feature vectors with three-dimensional spatial locations. Each voxel records the intensity of its associated leakage feature, thereby determining multiple leakage coordinate points. Then, multiple leakage coordinate points are combined with multiple weight coefficients and superimposed, that is, all weighted feature values ​​mapped to that location are accumulated for each voxel to generate a three-dimensional probability heat map, i.e., the initial leakage probability distribution map, where high-probability areas represent areas of concentrated leakage risk. Finally, the initial leakage probability distribution map is traversed for multimodal spatial verification. Specifically, temperature anomalies near the leakage point are detected by infrared thermal imaging, the spatial distribution of leakage ultrasonic signals is captured by acoustic emission detection, and surface oil stains or bubbles are identified by an industrial camera, thereby generating a physical leakage evidence dataset. Finally, the physical leakage evidence dataset is added to the verification results.

[0041] Preferably, the inversion is performed based on the verification results and the deviation value. The deviation value is the quantified difference between the first leakage feature vector and the physical evidence, such as probability error or spatial position offset. If the predicted location of the leakage coordinate point differs from the actual detection location by 2 mm, the spatial deviation is recorded. Specifically, the parameter with the greatest impact on the deviation in the first leakage feature vector is identified. For example, if the vibration frequency weight is too high, it may lead to false alarms. Gradient descent or genetic algorithm is used to adjust the weight coefficients of the feature vector to minimize the deviation between the prediction result and the physical evidence. That is, the parameters of the initial leakage feature vector are corrected through reverse reasoning, and the updated feature parameters are output as the inversion result. Then, the first leakage feature vector is updated based on the inversion result, including adjusting the weight allocation of the original vector based on the inversion result, such as reducing the weight of noise-sensitive features. If the physical evidence reveals an unmodeled leakage mode, such as a sudden drop in temperature, a new dimension is added to the first leakage feature vector group, and finally, the updated leakage feature vector group is obtained, that is, the second leakage feature vector group is generated, thereby ensuring the accuracy and reliability of the pipe joint sealing diagnosis.

[0042] Furthermore, step S340 also includes step S341, traversing the initial leakage probability distribution map to determine the leakage probability of the multiple voxel grids and identifying multiple probability peak regions; step S342, locating based on the multiple probability peak regions to identify multiple high-risk regions; step S343, determining whether the multiple high-risk regions are adjacent continuous regions. If the multiple high-risk regions are adjacent continuous regions, extracting continuous high-position regions for morphological dilation processing to generate a target cluster to be verified; step S344, calculating the centroid based on the target cluster to be verified, generating centroid coordinates as priority verification points, and performing multimodal spatial verification based on the priority verification points to generate the physical leakage evidence dataset.

[0043] Preferably, by using probability peak clustering and morphological optimization, discrete high-risk voxels are transformed into logically coherent suspected leakage areas, and the highest priority verification targets are intelligently screened to improve detection efficiency. Specifically, the leakage probability value of each voxel grid is threshold filtered, local maximum detection is used on the initial leakage probability distribution map, each voxel is compared with its three-dimensional neighboring voxels, the peak point in the probability distribution map is determined, and multiple discrete high-risk voxel sets are output, that is, multiple probability peak regions are determined. Each region contains spatial coordinate range and probability intensity. Then, based on the multiple probability peak regions, multiple high-risk areas are located and determined. Then, it is determined whether the multiple high-risk areas are adjacent and continuous regions, that is, the spatial position relationship of the high-risk areas is checked. If two regions share a surface or edge, such as a distance ≤ 1 voxel unit, they are determined to be continuous. Otherwise, they are isolated regions. For example, three adjacent high-probability voxels on a flange sealing ring are merged into a continuous leakage zone.

[0044] Preferably, if multiple high-risk areas have adjacent continuous regions, the continuous high-level regions are extracted and subjected to morphological dilation processing. Specifically, three-dimensional morphological dilation is performed on the continuous high-risk regions, that is, expanding outward by 1-2 voxel units from the original region as the center to fill the small gaps caused by discretization, thereby outputting the target cluster to be verified, that is, the smoothed continuous region, covering the potential leakage range; the geometric center, i.e., the centroid, is calculated based on the spatial coordinates of all voxels in the target cluster, including calculating the centroid coordinates as priority verification points using the leakage probability value of each voxel as a weight, where the centroid usually corresponds to the location of the most severe leakage; then, multimodal spatial verification is performed based on the priority verification points, including detecting temperature anomalies at the centroid point through infrared thermal imaging, capturing local high-frequency vibration signals using laser vibrometer, and directly observing surface damage or media seepage through a high-definition endoscope to cross-verify the centroid points, generating structured record verification results, i.e., physical leakage evidence dataset, thereby ensuring the reliability of the verification results.

[0045] Step S400: Based on the second leakage feature vector group, the leakage is relocated, a leakage risk alarm signal is constructed and sent to the remote monitoring terminal to perform a sealing diagnosis on the pipe joint and generate a sealing diagnosis report of the pipe joint.

[0046] Step S400 further includes step S410, extracting multi-level verification labels based on the second leakage feature vector group, performing leakage relocation according to the multi-level verification labels, and determining the target leakage coordinate set; step S420, connecting leakage trajectories according to the target leakage coordinate set to construct a leakage path topology map; step S430, retrieving the operating parameters of the pipe joint, mapping the operating parameters to the leakage path topology map for change analysis, and generating change trend results; step S440, performing a decreasing analysis according to the change trend results to identify the main leakage channel and seal failure point on the leakage path topology map; step S450, performing leakage risk analysis on the pipe joint based on the main leakage channel to construct a first leakage risk level, performing leakage risk analysis on the pipe joint based on the seal failure point to construct a second leakage risk level, and performing overlapping leakage risk analysis on the pipe joint based on the main leakage channel and the seal failure point to construct a third leakage risk level; step S460, adding the first leakage risk level, the second leakage risk level, and the third leakage risk level to the leakage risk alarm signal.

[0047] Preferably, the second leakage feature vector group (optimized features) is used to perform hierarchical verification of leakage points to improve positioning accuracy. That is, multi-level verification labels are extracted based on the second leakage feature vector group, including S-level labels, A-level labels, and B-level labels. Among them, S-level labels represent basic physical quantity verification, A-level labels represent temporal feature verification, and B-level labels represent temporal feature verification. Specifically, S-level label points are verified first. For example, a sphere search space with a radius of 5mm is created with the point as the center, and the spatial derivatives of the acoustic energy value and the pressure gradient value are superimposed to calculate the probability density peak point in the sphere. Then, A / B-level label points are verified, that is, the branch direction of the leakage path topology is extrapolated by 20mm to detect the temperature drop point on the extrapolated path. Then, high-confidence leakage points are selected according to the label level, false alarms are eliminated, and the target leakage coordinate set is output, including the three-dimensional coordinates of the leakage points verified by multiple levels, which is used to determine the real leakage point.

[0048] Preferably, the leakage trajectory is connected based on the target leakage coordinate set, that is, the spatial correlation of leakage points is analyzed to infer the leakage diffusion path. Specifically, based on the spatial proximity and temporal continuity of the coordinate set, discrete leakage points are connected, with leakage points as nodes and connection relationships as edges, to construct a leakage path topology map, that is, to represent the leakage channel with a graph structure, thereby revealing the direction of leakage diffusion and distinguishing between primary and secondary leaks. Then, real-time operating parameters such as pressure, temperature, and vibration of the pipe joint are retrieved and mapped onto the leakage path topology map. The parameter values ​​of each node are marked and then change analysis is performed, that is, the expansion rate of the leakage path is predicted through time series analysis, and the change trend results are output, such as the primary leakage channel extending by 2 mm every 10 minutes. Then, according to the change trend results, a decreasing analysis is performed, that is, the parameter decay gradient along the leakage path is calculated, such as the pressure drop rate, and the primary leakage channel and the seal failure point are marked on the leakage path topology map. Among them, the primary leakage channel represents the core path with the fastest parameter decay, and the seal failure point refers to the branch coordinate of the effective seal time, representing the bifurcation point of the secondary leakage path.

[0049] Preferably, leakage risk analysis is performed on the pipe joint based on the main leakage channel, i.e., calculating the direct risk based on leakage flow and pressure loss to construct a first leakage risk level, such as high risk – potentially leading to pipe burst; leakage risk analysis is also performed on the pipe joint based on the seal failure point, i.e., calculating the potential risk based on the number of branches and their timeliness to construct a second leakage risk level, such as medium risk – potentially expanding within 3 months; overlapping leakage risk analysis is performed on the pipe joint based on the main leakage channel and the seal failure point, i.e., comprehensively considering the interaction effect between the main channel and the branch point to construct a third leakage risk level, such as extremely high risk – the main channel accelerates the failure of the branch point. Finally, the first, second, and third leakage risk levels are encoded as alarm signals, added to the leakage risk alarm signal list, and pushed to the operation and maintenance terminal, realizing closed-loop management of the entire chain from micro-leakage points to macro-risk prediction.

[0050] Furthermore, step S400 also includes step S470, sending the leakage risk alarm signal to the remote monitoring terminal to mark the leakage on the three-dimensional structural model of the pipe joint based on the first leakage risk level, the second leakage risk level, and the third leakage risk level, and constructing a leakage location map; step S480, performing a seal failure diagnosis based on the leakage location map to determine the root cause data of the failure; step S490, performing a seal safety assessment on the pipe joint based on the root cause data of the failure, and generating a seal performance diagnosis report for the pipe joint.

[0051] Preferably, through 3D visualization annotation and root cause analysis, leakage risk alarms are transformed into operable sealing diagnostic reports. This involves sending leakage risk alarm signals to a remote monitoring terminal and annotating the 3D structural model of the pipe joint according to the first, second, and third leakage risk levels. Specifically, through-cracks or main leakage paths are highlighted in red on the 3D structural model, with annotation information including leakage flow rate and pressure attenuation gradient. Potential failure points, such as loose bolt areas, are marked in yellow, with the failure probability and expected failure time indicated. High-risk interactive areas, such as the intersection of the main channel and branch points, are displayed with a purple flashing effect, and the risk superposition coefficient is indicated. Finally, a rotatable and scalable 3D model is constructed as a leakage location map, integrating all risk annotations and real-time data.

[0052] Preferably, sealing failure diagnosis is based on leakage location maps. This involves combining leakage location maps with historical data to analyze the root cause of sealing failure. Specifically, the leakage path topology is compared with a database of typical failure cases, such as gasket aging and bolt fatigue. Then, pattern matching and parameter tracing are used to analyze the root cause of sealing failure, including tracing abnormal operating parameters, such as pressure peak records and sudden temperature changes. For example, if the leakage path shows a circumferential distribution and is accompanied by abnormal bolt vibration, it is determined that insufficient flange bolt preload leads to gasket sealing failure. If the leakage point is concentrated in the weld heat-affected zone and there is a sudden temperature gradient change, it is attributed to fatigue cracks caused by welding residual stress. The failure type, direct causes, and secondary factors are recorded in a structured manner to form root cause data.

[0053] Preferably, the pipe joint's sealing safety is assessed based on failure root cause data, including remaining life prediction, compliance checks, and risk quantification. Specifically, the safety window is calculated based on the leakage propagation rate and operating parameters to assess the remaining life, such as a current crack propagation rate of 0.1 mm / day and a remaining life of 20 days; violations are identified by comparing with industry standards; risk quantification refers to determining the consequences of failure through failure mode and effects analysis; finally, a comprehensive sealing performance diagnostic report for the pipe joint is generated, including a risk heatmap, root cause analysis tree, maintenance priority recommendations, and a spare parts recommendation list, thereby achieving an organic combination of predictive maintenance and precise repair, ensuring highly sensitive and accurate pipe joint leak detection, dynamic real-time monitoring and early warning, and intelligent sealing performance diagnosis.

[0054] Based on the foregoing embodiments, this application also provides an electronic device and a computer-readable storage medium storing a computer program. When the computer program is executed by the processor of the electronic device, it can implement the methods described in any of the preceding embodiments.

[0055] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention, showing a block diagram of an exemplary electronic device suitable for implementing the embodiments of the present invention. Figure 3 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments of the present invention. This electronic device is in the form of a general-purpose computing device, and its components may include, but are not limited to, an input device 401, a processor 402, a memory 403, and an output device 404. The processor 402 may be one or more; the memory 403 may include a computer-readable medium and at least one program product having a set (at least one) of program modules configured to perform the functions of the embodiments of this application.

[0056] The memory 403 shown in this embodiment of the invention can be any combination of one or more computer-readable media. The computer-readable storage medium can be, but is not limited to, an infrared, semiconductor system, device or apparatus, or any combination thereof, used to store software programs, computer-executable programs and modules, such as the program instructions / modules corresponding to the pipe joint sealing monitoring method in this embodiment of the invention. The processor 402 executes various functional applications and data processing of the computer device by running the software programs, instructions and modules stored in the memory 403, thereby realizing the above-mentioned pipe joint sealing monitoring method.

[0057] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application. In some cases, the actions or steps described in this application can be performed in a different order than that shown in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A method for monitoring the sealing performance of pipe joints, characterized in that, The method includes: Dynamic sensing is performed on the connection area of ​​the pipe fitting to obtain the first sensing dataset, and dynamic sensing is performed on the connection area of ​​the connector to obtain the second sensing dataset. Based on the deviation calculation between the first sensor dataset and the second sensor dataset, leakage analysis is performed on the pipe joint according to the deviation value, and a first leakage feature vector group is constructed. The first leakage feature vector group is mapped to the three-dimensional structural model of the pipe joint, multiple leakage coordinate points are determined for spatial verification, and inversion is performed based on the verification results and deviation values. The first leakage feature vector group is updated based on the inversion results to generate the second leakage feature vector group. Based on the second leakage feature vector group, the leakage is relocated, a leakage risk alarm signal is constructed and sent to the remote monitoring terminal to perform a sealing diagnosis on the pipe joint and generate a sealing diagnosis report of the pipe joint.

2. The pipe joint sealing performance monitoring method as described in claim 1, characterized in that, The connection area of ​​the pipe fitting is dynamically sensed to obtain a first sensing dataset, and the connection area of ​​the connector is dynamically sensed to obtain a second sensing dataset. The method includes: Retrieve the first historical area data change log of the pipe joint sealing connection area, perform change frequency analysis based on the first historical area data change log, and determine the first collection period; Retrieve the second historical area data change log of the sealing connection area of ​​the connector, perform change frequency analysis based on the second historical area data change log, and determine the second collection period; According to the first acquisition cycle, pressure change sensing is performed on the sealing connection area of ​​the pipe joint to obtain the first pressure fluctuation data, and the first pressure fluctuation data is added to the first sensing dataset. Pressure change sensing is performed on the connection area of ​​the connector according to the second acquisition cycle to obtain second pressure fluctuation data, and the second pressure fluctuation data is added to the second sensing dataset.

3. The pipe joint sealing performance monitoring method as described in claim 2, characterized in that, Based on the deviation calculation between the first sensor dataset and the second sensor dataset, leakage analysis is performed on the pipe joint according to the deviation value, and a first leakage feature vector group is constructed. The method includes: Spatiotemporal alignment is performed based on the first and second sensor datasets to construct a joint sensing matrix; Based on the joint sensing matrix, pressure field deviation is calculated for the first pressure fluctuation data and the second pressure fluctuation data, and a pressure deviation topology map is constructed. Deformation field deviation is calculated based on the joint sensing matrix, and a deformation mismatch topology map is constructed. The pressure deviation topology map and the deformation mismatch topology map are dynamically spatiotemporally aligned and fused to construct a dynamic deviation field. Anomaly analysis is performed based on the dynamic deviation field to determine the abnormal deviation pattern. Leakage feature analysis is then performed using the abnormal deviation pattern to generate the first leakage feature vector group.

4. The pipe joint sealing performance monitoring method as described in claim 3, characterized in that, Anomaly analysis is performed based on the dynamic deviation field to determine the abnormal deviation pattern. Leakage feature analysis is then performed using the abnormal deviation pattern to generate the first leakage feature vector set. The method includes: Based on the dynamic deviation field, coupled anomaly identification is performed to determine the characteristics of the first anomalous event; Based on the pressure deviation topology map and the deformation mismatch topology map, maximum value analysis is performed to extract the pressure deviation peak value and the deformation mismatch peak value. Phase lag calculations are performed based on the pressure deviation peak value and the deformation mismatch peak value to determine the characteristics of the second abnormal event. The first abnormal event feature and the second abnormal event feature are normalized, and the leakage feature is matched by traversing the abnormal deviation pattern according to the normalization result to construct the first leakage feature vector group.

5. The pipe joint sealing performance monitoring method as described in claim 1, characterized in that, The first leakage feature vector group is mapped to the three-dimensional structural model of the pipe joint, and multiple leakage coordinate points are determined for spatial verification. The method includes: A three-dimensional structural model of the pipe joint is introduced, and the three-dimensional structural model of the pipe joint is discretized into multiple voxel meshes; Based on the first leakage feature vector group, leakage impact analysis is performed to generate impact factors. Based on the impact factors, the first leakage feature vector group is weighted and multiple weight coefficients are determined. The first leakage feature vector group is mapped to the plurality of voxel grids to determine the plurality of leakage coordinate points; An initial leakage probability distribution map is constructed by superimposing the multiple leakage coordinate points and the multiple weighting coefficients. The initial leakage probability distribution map is traversed to perform multimodal spatial verification, generating a physical leakage evidence dataset, which is then added to the verification result.

6. The pipe joint sealing performance monitoring method as described in claim 5, characterized in that, The method involves traversing the initial leakage probability distribution map to perform multimodal spatial verification and generate a physical leakage evidence dataset. The leakage probability of the multiple voxel grids is determined by traversing the initial leakage probability distribution map, and multiple probability peak regions are identified. Based on the multiple probability peak regions, multiple high-risk areas are identified; Determine whether the multiple high-risk areas are adjacent and continuous. If the multiple high-risk areas are adjacent and continuous, extract the continuous high-level areas and perform morphological dilation processing to generate a target cluster to be verified. Centroids are calculated based on the target cluster to be verified, and centroid coordinates are generated as priority verification points. Multimodal spatial verification is then performed based on the priority verification points to generate the physical leakage evidence dataset.

7. The pipe joint sealing performance monitoring method as described in claim 1, characterized in that, Leak relocation is performed based on the second leakage feature vector group, and a leakage risk alarm signal is constructed. The method includes: Based on the second leakage feature vector group, multi-level verification labels are extracted, and leakage is relocated according to the multi-level verification labels to determine the target leakage coordinate set; Based on the target leakage coordinate set, leakage trajectories are connected to construct a leakage path topology map; The operating parameters of the pipe joint are retrieved, and the operating parameters are mapped to the leakage path topology map for change analysis to generate change trend results; Based on the changing trend results, a decreasing analysis is performed to identify the main leakage channel and the seal failure point in the leakage path topology map. Based on the main leakage channel, a leakage risk analysis is performed on the pipe joint to construct a first leakage risk level. Based on the seal failure point, a leakage risk analysis is performed on the pipe joint to construct a second leakage risk level. Based on the main leakage channel and the seal failure point, an overlapping leakage risk analysis is performed on the pipe joint to construct a third leakage risk level. Add the first leakage risk level, the second leakage risk level, and the third leakage risk level to the leakage risk alarm signal.

8. The pipe joint sealing performance monitoring method as described in claim 7, characterized in that, The method for generating a leak risk alarm signal and sending it to a remote monitoring terminal to perform a seal diagnosis on the pipe joint and generate a pipe joint seal diagnosis report includes: The leakage risk alarm signal is sent to the remote monitoring terminal to mark the leakage in the three-dimensional structural model of the pipe joint based on the first leakage risk level, the second leakage risk level, and the third leakage risk level, and to construct a leakage location map; Based on the leak location map, seal failure diagnosis is performed to determine the root cause data of the failure. Based on the failure root cause data, a sealing safety assessment of the pipe joint is performed, and a sealing performance diagnostic report of the pipe joint is generated.

9. An electronic device, characterized in that, The electronic device includes: Memory, used to store executable instructions; The processor, when executing executable instructions stored in the memory, implements the pipe joint sealing monitoring method according to any one of claims 1-8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the pipe joint sealing monitoring method as described in any one of claims 1-8.

Images