Visual monitoring method and device and storage medium

Abstract

The invention discloses a visual monitoring method and device and a storage medium, and belongs to the technical field of chemical engineering. The method comprises the following steps: receiving multi-dimensional data of a pipeline operation state collected by a multi-source sensing device; processing the multi-dimensional data to generate a pipeline state feature set; dynamically associating the pipeline state feature set to a pipeline GIS coordinate, and generating a pipeline operation model; identifying an abnormal state of the pipeline operation model to obtain a graded early warning instruction; and sending the grading early warning instruction to a target terminal according to the user authority, so that a user executes corresponding operation according to the grading early warning instruction acquired by the target terminal.

Description

Technical Field

[0001] This application relates to the field of chemical technology, and in particular to a visual monitoring method, device, and storage medium. Background Technology

[0002] With the continuous growth of energy demand, pipelines transporting fluids such as oil and natural gas are becoming increasingly large and complex, leading to frequent pipeline leaks that pose a serious threat to the environment, economy, and public safety. Therefore, efficient and accurate pipeline leak monitoring technology has become crucial for ensuring the safe operation of pipelines.

[0003] The mainstream pipeline leak monitoring technology on the market is mainly based on the principle of fiber optic vibration data monitoring. By laying fiber optic sensors along the pipeline, vibration signals caused by leaks are captured, and complex algorithm models are used to analyze and judge whether a leak has occurred.

[0004] However, existing technologies rely heavily on data analysis in monitoring pipeline leaks and lack intuitive visualization methods, making it difficult for emergency responders to quickly obtain basic information about the leak site. This delays the best rescue opportunity and increases the difficulty and cost of handling the accident. Summary of the Invention

[0005] To address the aforementioned technical problems, this application provides a visual monitoring method, apparatus, and storage medium.

[0006] The technical solution provided in this application is described below: The first aspect of this application provides a visual monitoring method, the method comprising: Receive multi-dimensional data on pipeline operating status collected by multi-source sensing devices; The multidimensional data is processed to generate a pipeline status feature set; The pipeline status feature set is dynamically associated with the pipeline GIS coordinates to generate a pipeline operation model. Identify abnormal states in the pipeline operation model to obtain tiered early warning instructions; The tiered warning instructions are sent to the target terminal according to user permissions, so that the user can perform corresponding operations based on the tiered warning instructions obtained by the target terminal.

[0007] Optionally, the processing of the multidimensional data to generate a pipeline status feature set includes: Data type identification is performed on the multidimensional data; If the data type is image data, then noise reduction processing is performed to obtain a pipeline surface feature image; If the data type is temperature data, then baseline calibration is performed to obtain a temperature gradient sequence; If the data type is gas concentration data, then spatial interpolation analysis is performed to obtain the diffusion distribution parameters of hazardous substances; The pipeline surface feature image, the temperature gradient sequence, and the hazardous substance diffusion distribution parameters are associated with identifiers to generate a pipeline state feature set.

[0008] Optionally, before dynamically associating the pipeline status feature set with the pipeline GIS coordinates to generate the pipeline operation model, the method further includes: Obtain pipeline GIS coordinates from geographic information data; A mapping table is constructed based on the pipeline GIS coordinates and the identifier; Construct an initial 3D model; The initial 3D model is reconstructed based on the pipeline GIS coordinates and the mapping table to generate a 3D GIS model of the utility tunnel.

[0009] Optionally, the step of dynamically associating the pipeline status feature set with pipeline GIS coordinates to generate a pipeline operation model includes: Extract the target feature data and corresponding unique identifier from the pipeline status feature set; The corresponding target pipeline GIS coordinates are retrieved from the mapping table based on the unique identifier. Based on the target pipeline's GIS coordinates, the pipeline location nodes are obtained in the 3D GIS model of the utility tunnel. The target feature data is stored in the pipeline location node to generate a pipeline operation model.

[0010] Optionally, identifying the abnormal state of the pipeline operation model to obtain tiered early warning instructions includes: The pipeline operation model is compared with a preset risk threshold range to obtain abnormal data points and corresponding risk levels; Retrieve an early warning instruction template from the early warning instruction template library based on the aforementioned risk level; The abnormal data points are filled into the early warning instruction template to obtain tiered early warning instructions.

[0011] Optionally, filling the abnormal data points into the early warning instruction template to obtain tiered early warning instructions includes: Parse the warning instruction template to obtain placeholders and data format requirements; Replace the placeholders with the abnormal data points according to the data format requirements to generate a tiered early warning instruction.

[0012] Optionally, after sending the tiered warning instruction to the target terminal according to user permissions, so that the user can perform corresponding operations according to the tiered warning instruction obtained by the target terminal, the method further includes: Receive the execution result returned by the target terminal, and update the multidimensional data according to the execution result.

[0013] A second aspect of this application provides a visual monitoring device, the device comprising: The receiving unit is used to receive multi-dimensional data on pipeline operating status collected by multi-source sensing devices; The processing unit is used to process the multidimensional data and generate a pipeline status feature set; The association unit is used to dynamically associate the pipeline status feature set with the pipeline GIS coordinates to generate the pipeline operation status. The identification unit is used to identify abnormal states of the pipeline operation model in order to obtain graded early warning instructions; The sending unit is used to send the graded warning instruction to the target terminal according to the user's permissions, so that the user can perform corresponding operations according to the graded warning instruction obtained by the target terminal.

[0014] A third aspect of this application provides a visual monitoring device, the device comprising: Processor, memory, input / output units, and bus; The processor is connected to the memory, the input / output unit, and the bus; The memory stores a program, which the processor invokes to execute the first aspect and any one of the optional methods in the first aspect.

[0015] A fourth aspect of this application provides a computer-readable storage medium on which a program is stored, which, when executed on a computer, performs the methods of the first aspect and any one of the first aspects.

[0016] As can be seen from the above technical solutions, this application has the following advantages: This application overcomes the limitations of single monitoring methods by receiving multi-dimensional data collected from multiple source sensors, achieving comprehensive perception of the overall operational status of pipelines. The multi-dimensional data is then processed to generate a pipeline status feature set, providing a high-quality data foundation for subsequent analysis. Next, the pipeline status feature set is dynamically correlated with the pipeline's GIS coordinates to generate a pipeline operation model, accurately locating the pipeline. This overcomes the limitations of traditional monitoring methods, which suffer from missing spatial information and difficulty in intuitively presenting data, achieving deep integration of data and geospatial data. Furthermore, abnormal states in the pipeline operation model are identified, and tiered early warning instructions are obtained, enabling precise early warning and allowing administrators to take targeted measures based on different risk levels. Finally, tiered early warning instructions are sent to target terminals according to user permissions, ensuring that pipeline anomaly information is accurately delivered to relevant administrators, improving response efficiency, effectively reducing the difficulty and cost of accident handling, and guaranteeing the safe and stable operation of pipelines. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A schematic flowchart of an embodiment of the visualization monitoring method provided in this application; Figure 2 A schematic flowchart of another embodiment of the visualization monitoring method provided in this application; Figure 3 A schematic flowchart of another embodiment of the visualization monitoring method provided in this application; Figure 4 A schematic flowchart of another embodiment of the visualization monitoring method provided in this application; Figure 5 A schematic flowchart of another embodiment of the visualization monitoring method provided in this application; Figure 6 A schematic diagram of an embodiment of the visual monitoring device provided in this application; Figure 7 This is a schematic diagram of another embodiment of the visualization monitoring device provided in this application. Detailed Implementation

[0019] It should be noted that the visualization monitoring method provided in this application can be applied to terminals or systems, and can also be applied to systems. For example, a terminal can be a smartphone, computer, tablet, smart TV, smartwatch, portable computer terminal, or a desktop computer or other fixed terminal. For ease of explanation, this application uses a terminal as the implementing entity for illustration.

[0020] Please see Figure 1 This application first provides an embodiment of a visualization monitoring method, which includes: S101, Receive multi-dimensional data on pipeline operating status collected by multi-source sensing devices; In this embodiment, multi-source sensing devices refer to various types of sensors deployed at and around key parts of the pipeline, including but not limited to high-definition cameras, gas sensors, and infrared sensors, used to collect pipeline operation data from different dimensions. Multi-dimensional data: Various types of data collected by multi-source sensing devices, covering video image data, gas concentration data, and temperature data, reflecting pipeline operation status data from multiple perspectives.

[0021] Before receiving multidimensional data, multi-source sensors need to be strategically deployed at key locations and in the surrounding area based on pipeline characteristics and monitoring requirements. For example, gas sensors can be installed at leak-prone connections, and high-definition cameras and infrared imagers can be installed in open pipeline areas. Next, a communication link is established between the multi-source sensors and the system via wired or wireless communication technology. Then, the multi-source sensors collect multidimensional data in real time according to a preset sampling frequency. For example, gas sensors collect gas concentration data once per second, high-definition cameras collect 25 frames of image data per second, and infrared sensors detect surface heat radiation at a far-infrared frequency of 30THz to collect operating temperature data. Finally, the multi-source sensors transmit the collected multidimensional data to the system via the communication link.

[0022] S102. Process the multidimensional data to generate a pipeline status feature set; In this embodiment, the received multidimensional data undergoes data type identification. By analyzing the data's characteristics and format, it is determined whether the data is image data, temperature data, or gas concentration data, and a classification label is output for it. Next, for the portion identified as image data, a filtering algorithm can be used for noise reduction processing (no specific limitation is made here) to remove noise interference from the image, thereby obtaining a clear image of the pipeline surface features. For temperature data, baseline calibration is performed. Based on historical temperature data or theoretical temperature values, a temperature baseline model is constructed. The collected temperature data is compared and calibrated with the baseline model to obtain an accurate temperature gradient sequence. For gas concentration data, spatial interpolation analysis is used, combined with gas concentration measurements at different locations, to calculate the hazardous substance diffusion distribution parameters throughout the space. Furthermore, unique identifiers are assigned to these processed data and then bound with these unique identifiers. For example, the unique identifier for a temperature gradient sequence is PIPE-001-TEMP1-202505121030, containing information such as pipeline number PIPE-001, sensor type and number TEMP1, and timestamp 202505121030. Finally, the pipeline surface feature images, temperature gradient sequences, and hazardous substance diffusion distribution parameters, all bound with unique identifiers, are aggregated into a unified data structure to generate a pipeline status feature set.

[0023] S103. Dynamically associate the pipeline status feature set with the pipeline GIS coordinates to generate a pipeline operation model; In this embodiment, pipeline GIS coordinates refer to the coordinates of the geographical location of the pipeline corresponding to the multi-source sensing device, which can accurately determine the position of the sensor in the pipeline geographic model space. The pipeline operation model is the overall operation status of the pipeline presented in a dynamic visualization manner in a 3D GIS model after associating the pipeline status feature set with the pipeline GIS coordinates.

[0024] Before generating the pipeline operation model, it is necessary to obtain overall geographic information data of the pipeline from reliable data sources such as data from on-site pipeline surveys, professional geographic information databases, and high-precision maps. Then, based on the location of the multi-source sensors in the overall geographic information data of the pipeline, the GIS coordinates of the multi-source sensors on the pipeline are extracted. Next, the unique identifiers of the target feature data and the pipeline GIS coordinates within the pipeline status feature set are traversed. During this traversal, the unique identifiers and pipeline GIS coordinates in the pipeline status feature set are structured and parsed to extract key information as matching keys, establishing key-value pair mappings to generate a mapping table. Then, using 3D modeling software, an initial 3D model of the pipeline and its surrounding environment is created based on the actual design layout of the pipeline. The obtained pipeline GIS coordinates are marked on the initial 3D model to obtain the 3D GIS model of the utility tunnel. Subsequently, the target feature data and corresponding unique identifiers in the pipeline status feature set are extracted in a first-in-first-out order, and the target pipeline GIS coordinates corresponding to the unique identifiers are found in the established mapping table. Finally, the corresponding pipeline location nodes are found in the 3D GIS model of the utility tunnel based on the target pipeline GIS coordinates. Finally, by replacing the original data of each target feature data in sequence and storing it on the pipeline location node, a pipeline operation model that can be dynamically presented in a visual manner in the 3D model can be generated. It should be noted that the pipeline location node is the data storage location of the target feature data in the 3D GIS model of the utility tunnel.

[0025] S104. Identify abnormal states in the pipeline operation model to obtain tiered early warning instructions; In this embodiment, the tiered early warning instruction is an early warning message generated based on different risk levels, used to guide relevant personnel to take corresponding measures. The risk threshold range is a pre-set range of normal data values; data exceeding this range is considered abnormal data.

[0026] Before obtaining tiered early warning instructions, it is necessary to first set risk threshold ranges for different types of risk indicators. These threshold ranges can be determined based on historical data and industry standards. Next, the pipeline operating status is compared with the risk threshold ranges, analyzing each data point in the pipeline operating status to see if it exceeds the threshold range. If it does, it is marked as an abnormal data point, and the corresponding risk level is determined based on the degree of exceedance, such as minor, moderate, or severe. Then, based on the determined risk level, a suitable early warning instruction template is selected from the early warning instruction template library. This library stores early warning instruction templates corresponding to different risk levels, and these templates contain the basic format and content framework of the early warning information. Further, according to the placeholders and data format requirements in the early warning instruction template, the abnormal data points are replaced with placeholders according to the data format requirements to generate a complete tiered early warning instruction. The placeholders are used to fill in the positions of specific abnormal data.

[0027] S105. Send the graded warning instructions to the target terminal according to the user's permissions, so that the user can perform corresponding operations according to the graded warning instructions obtained by the target terminal; In this embodiment, before sending the tiered warning instructions to the target terminal, the system assigns a specific permission level to each user's terminal device, such as ordinary user or administrator. Different permission levels correspond to different permissions for viewing and processing warning information. Next, the warning instructions are filtered and processed according to user permissions. For example, ordinary users may only see some key information, while administrators can see detailed warning reports. Then, the tiered warning instructions are sent to the target terminal via push notification services, such as SMS, app push notifications, and email. The target terminal can be a mobile phone, tablet, computer, or other device. Furthermore, after receiving the tiered warning instructions, the target terminal will prominently remind the user, such as through pop-ups or sound alerts, allowing the user to perform corresponding operations based on the content of the warning instructions, such as inspecting or repairing pipelines.

[0028] S106. Receive the execution result returned by the target terminal and update the multidimensional data according to the execution result.

[0029] In this embodiment, before receiving the execution result returned by the target terminal, a data receiving channel needs to be established to receive the execution result returned by the target terminal. This channel can be a network-based interface, such as an HTTP interface or a WebSocket interface. Then, after the user completes the maintenance, the execution result after maintenance can be sent to this data receiving channel through the target terminal according to a preset data format. After receiving the execution result, the system verifies and parses it to ensure the accuracy and integrity of the data. Then, the system evaluates the pipeline's operating status after maintenance. If the operation is successful and the pipeline status improves, the relevant parameters in the multidimensional data, such as temperature, gas concentration, and image data, are adjusted accordingly. If the operation is unsuccessful or the pipeline status still has problems, these problems are marked, and the causes are further analyzed. The updated multidimensional data is stored in the system's database for subsequent analysis and processing. Simultaneously, this updated data also participates in the next round of monitoring and early warning processes, ensuring that the system can reflect the overall operating status of the pipeline in real time.

[0030] This embodiment overcomes the limitations of single monitoring methods by receiving multi-dimensional data collected from multiple source sensors, achieving comprehensive perception of the overall pipeline operating status. The multi-dimensional data is then processed to generate a pipeline status feature set, providing a high-quality data foundation for subsequent analysis. Next, the pipeline status feature set is dynamically correlated with the pipeline's GIS coordinates to generate a pipeline operation model, accurately locating the pipeline. This overcomes the limitations of traditional monitoring methods, which suffer from missing spatial information and difficulty in intuitively presenting data, achieving deep integration of data and geospatial data. Furthermore, abnormal states in the pipeline operation model are identified, and tiered early warning instructions are obtained, enabling precise early warning and allowing administrators to take targeted measures based on different risk levels. Finally, tiered early warning instructions are sent to target terminals according to user permissions, ensuring that pipeline anomaly information is accurately delivered to relevant administrators, improving response efficiency, effectively reducing the difficulty and cost of accident handling, and guaranteeing the safe and stable operation of the pipeline.

[0031] Please see Figure 2 , Figure 2 Another embodiment of the visualization monitoring method provided in this application includes: S201. Identify data types for multidimensional data; In this embodiment, before performing data type identification on multidimensional data, a modality recognition model needs to be constructed using convolutional neural networks and recurrent neural networks in deep learning. Convolutional neural networks are used to extract local features of data, such as capturing the edges and textures of image data; for other types of data, they can also extract some basic and representative feature patterns. Recurrent neural networks are used to process the temporal relationship of data, because different types of data may have different patterns of change over time.

[0032] Next, the model training phase begins. A large dataset with labeled data types is collected, containing different modal data samples from various real-world pipeline scenarios. This dataset is then divided into training, validation, and test sets. The model is trained using the training set, with backpropagation continuously adjusting its parameters to enable it to learn the features of different modalities. During training, the validation set is used to monitor model performance and prevent overfitting. Finally, after training is complete, the test set is used to evaluate metrics such as accuracy and recall, resulting in a well-trained modality recognition model.

[0033] Next, the received multidimensional data is input into the trained modality recognition model. The model outputs a data type label for each data sample, such as "image," "temperature," or "gas concentration." Finally, the data is classified according to the labels and marked according to time sequence or data source, forming a multi-source sensor data type with classification labels, so that subsequent processing can accurately identify and process different types of data.

[0034] S202. If the data type is image data, then noise reduction processing is performed to obtain the surface feature image of the pipeline. In this embodiment, when the data type is image data, a noise reduction scheme combining Gaussian filtering and adaptive median filtering can be adopted. First, Gaussian filtering is applied. Gaussian filtering is based on a weighted average of the image using a Gaussian function, which effectively suppresses Gaussian noise. For pipeline images, a suitable Gaussian kernel size, such as 5×5, and a standard deviation, such as 1.5, are set. Then, the Gaussian kernel is slid across the image, and the new value of each pixel is the sum of the products of its neighboring pixel values ​​and the Gaussian kernel weights, making the overall image smoother and initially reducing noise interference. Next, adaptive median filtering is performed. Median filtering dynamically adjusts the filtering window size according to the noise density in the pixel's neighborhood. It first checks whether the current pixel is noise; if it is noise, it replaces it with the median value of non-noise pixels in the neighborhood; if it is not noise, the original value is retained. In pipeline image processing, the initial value of the filtering window is set to 3×3. If the proportion of noise pixels in the neighborhood exceeds a certain percentage, such as 40%, the window is expanded to 5×5 for further evaluation. This method removes impulse noise, such as salt-and-pepper noise, that Gaussian filtering fails to eliminate, while also preserving key edge details on the pipeline surface, such as cracks and corrosion pits. After these two filtering processes, a clear pipeline surface feature image with minimal noise interference is obtained.

[0035] S203. If the data type is temperature data, then perform baseline calibration to obtain the temperature gradient sequence; In this embodiment, the temperature gradient sequence refers to a sequence composed of temperature differences between adjacent time points, reflecting the rate of temperature change over time.

[0036] When the data type is temperature data and baseline calibration is being performed, the first step is to collect historical temperature data over a period of time. This data can come from measurements taken by the same pipeline under normal operating conditions, or it can be referenced from standard temperature data of similar pipelines. Based on this historical data, a temperature baseline model is established. Methods such as polynomial fitting and linear regression can be used to build the model. For example, a linear regression model can be used to fit the temperature change trend over time. After obtaining the baseline model, the currently collected temperature data is compared with the baseline model. If a deviation is found, calibration is required. The calibration method can be to add or subtract a correction value for each temperature data point; this correction value is calculated based on the difference between the baseline model and the current data. After calibration, the temperature difference between adjacent time points is calculated to obtain the temperature gradient sequence.

[0037] S204. If the data type is gas concentration data, then perform spatial interpolation analysis to obtain the diffusion distribution parameters of hazardous substances. In this embodiment, the hazardous substance diffusion distribution parameter refers to the parameter that describes the diffusion range, concentration distribution, and other characteristics of hazardous substances in space.

[0038] When the data type is gas concentration data, a reasonable concentration threshold must first be set, such as 50% of the lower explosive limit of flammable gases. Next, the collected gas concentration data is compared with the threshold to perform threshold segmentation, initially screening out high-risk areas exceeding the safety threshold. Subsequently, Kriging spatial interpolation is used to perform spatial interpolation analysis on the limited gas concentration monitoring point data, calculating the semi-variogram and selecting a suitable theoretical model, such as an exponential model, for fitting. By solving the Kriging equations, the gas concentration distribution throughout the monitoring area is estimated, generating a continuous concentration distribution map. Further, based on the interpolation results, the maximum concentration value and its coordinates are extracted, concentration contour lines are drawn to divide areas into different risk levels, and the diffusion radius is calculated. The diffusion radius is the distance from the leak source to a specific concentration contour line, and the diffusion direction is corrected by incorporating meteorological data such as wind direction and speed. Finally, by integrating the maximum concentration value, concentration contour line distribution, diffusion radius, and risk area division results, the hazardous substance diffusion distribution parameters can be obtained.

[0039] S205. Associate the pipeline surface feature image, temperature gradient sequence, and hazardous substance diffusion distribution parameters with identifiers to generate a pipeline status feature set.

[0040] In this embodiment, when performing tag association, firstly, an identifier is designed and assigned to each type of data. The identifier adopts a format such as "PIPE-001-TEMP1-202505121030", where "PIPE-001" clearly defines the pipeline number, "TEMP1" represents the sensor type and number, and "202505121030" records the data acquisition timestamp. Next, this identifier is used as the core tag and associated with the corresponding data. For example, for pipeline surface feature images, the identifier is embedded in the image file metadata to ensure that the image is bound to a specific pipeline, sensor, and acquisition time; for temperature gradient sequences and hazardous substance diffusion distribution parameters, an identifier field is added to the stored data table to correspond the data one-to-one with the tag. Furthermore, by associating and integrating the data, redundancy and contradictions between data are eliminated, and various types of tagged data are summarized into a unified data structure to construct a pipeline status feature set.

[0041] This embodiment identifies and categorizes multi-source sensor data types by performing data type identification on multi-dimensional data. This allows for precise differentiation of different data types, such as images, temperatures, and gas concentrations, laying the foundation for subsequent targeted processing, avoiding data confusion, and significantly improving data processing efficiency and accuracy. Furthermore, specific processing measures for different data types yield significant results. Noise reduction processing of image data effectively eliminates noise interference, obtaining clear images of pipeline surface features, facilitating accurate identification of pipeline surface damage, corrosion, and other defects. Baseline calibration of temperature data eliminates systematic errors, obtaining accurate temperature gradient sequences that precisely reflect pipeline temperature changes. Spatial interpolation analysis of gas concentration data compensates for sensor placement limitations, obtaining hazardous substance diffusion distribution parameters and visually displaying the diffusion range and concentration distribution, providing a basis for subsequent risk assessment. Further, these processed data are tagged and associated to generate a pipeline status feature set, achieving multi-dimensional data integration. This provides complete data support for comprehensive and in-depth analysis of the overall pipeline operating status and helps ensure the safe and stable operation of the pipeline.

[0042] Please see Figure 3 , Figure 3 Another embodiment of the visualization monitoring method provided in this application includes: S301. Obtain the GIS coordinates of the pipeline based on geographic information data; In this embodiment, when acquiring the GIS coordinates of the pipeline, it is necessary to collect comprehensive geographic information data, including pipeline route, burial depth, and connection relationships, from data sources such as pipeline field survey data, professional geographic information databases, and high-precision maps, to construct a comprehensive geographic information archive for the pipeline. Then, based on the fact that the multi-source sensor device has been installed on the pipeline, it is mapped to the overall geographic information data according to its relative position in the actual pipeline geographic location. Next, by analyzing the spatial relationship between the sensor device and various nodes and feature points of the pipeline, and utilizing the correspondence between the data, the GIS coordinates of the multi-source sensor device on the pipeline can be calculated.

[0043] S302. Construct a mapping table based on the pipeline GIS coordinates and identifiers; In this embodiment, when constructing the mapping table, the system first iterates through the identifiers and GIS coordinates of the data within the pipeline status feature set. During the traversal, string parsing technology is used to extract information such as the pipeline number PIPE-001, sensor type and number TEMP1, and data generation timestamp 202505121030 from the identifier, such as PIPE-001-TEMP1-202505121030. Simultaneously, the pipeline GIS coordinates, such as PIPE-001-TEMP1-116.3-39.9, are formatted and parsed to obtain the pipeline number, sensor type and number, longitude, and latitude contained within them. By comparing the fields of the two, the system identifies key information that is the same in the identifier and GIS coordinates, such as the pipeline number and sensor type and number, and uses this as the matching basis. Then, based on these matching key information, the system establishes a key-value pair mapping relationship between the identifier and the corresponding GIS coordinate. After each set of data is matched, the key-value pair is stored in the mapping table of the relational database, and the table structure includes identifier and GIS coordinate fields. Finally, during the mapping process, a data verification mechanism is set up to perform a second check on the matched key-value pairs to ensure that the pipeline number and sensor type are completely consistent, thus avoiding mismatches and completing the construction of a complete and accurate mapping table.

[0044] S303. Construct the initial 3D model; In this embodiment, the three-dimensional model is a basic three-dimensional geometric structure that has not yet incorporated specific geographic information and monitoring data. It uses a three-dimensional spatial coordinate system as a reference to initially outline the approximate shape, size, and layout of the utility tunnel, providing a framework for adding detailed information later.

[0045] When building a 3D model, the first step is to accurately draw the various parts of the utility tunnel, such as pipes, supports, and valves, according to scale and dimensions, using 3D modeling software such as 3ds Max, based on the collected detailed design data. Next, materials and textures are added to the 3D model to make it more closely resemble the appearance of a real utility tunnel, thus obtaining the initial 3D model.

[0046] S304. Reconstruct the initial 3D model based on the pipeline GIS coordinates and mapping table to generate a 3D GIS model of the utility tunnel.

[0047] In this embodiment, during the 3D model reconstruction, the system first uses a coordinate transformation algorithm to convert the pipeline GIS coordinates from the geographic coordinate system to the local coordinate system of the initialized 3D model, achieving a precise mapping of coordinates from real geographic space to virtual 3D space. Next, based on the pipeline type, preset 3D model components, such as straight lines, bends, and valves, are invoked. By connecting adjacent coordinate points of linear pipelines, a continuous 3D pipeline form is constructed, and visual annotations are added at the nodes. Then, the system binds the objects corresponding to the pipeline GIS coordinates in the 3D model with identifiers in the mapping table, establishing a correspondence between model objects and pipeline GIS coordinates, enabling bidirectional interactive querying of geographic information, monitoring data, and the 3D model. Finally, through lighting rendering, shadow processing, and layer optimization, the realism and readability of the model are enhanced, thereby generating a complete, accurate, and interactive 3D GIS model of the utility tunnel.

[0048] This embodiment acquires pipeline GIS coordinates through geographic information data, integrating multi-source information such as on-site surveying and professional databases to provide a precise spatial benchmark for pipeline positioning, ensuring the reliability and comprehensiveness of data sources. Next, a mapping table is constructed based on the pipeline GIS coordinates and identifiers, establishing a two-way association between data and spatial location, enabling efficient data retrieval and accurate matching, facilitating subsequent data analysis and management. Furthermore, an initial 3D model is constructed and reconstructed based on the pipeline GIS coordinates and mapping table, achieving deep integration of geographic information and the 3D model. It also establishes a link between pipeline status feature sets and spatial location, providing an efficient visual interactive platform for subsequent data analysis and operation and maintenance management.

[0049] Please see Figure 4 , Figure 4 Another embodiment of the visualization monitoring method provided in this application includes: S401. Extract the target feature data and corresponding unique identifier from the pipeline status feature set; In this embodiment, during data extraction, the system loads the pipeline status feature set into a memory queue and sorts it naturally according to the data generation timestamp, constructing a first-in-first-out (FIFO) data processing queue. Then, the system traverses the data items starting from the head of the queue using a queue pointer, extracting one complete pipeline status data item at a time. For each extracted pipeline status data item, regular expression matching and string segmentation techniques are used to parse its structure, separating target feature data such as temperature, image, gas concentration, deformation, and corresponding unique identifiers. Taking temperature data as an example, its identifier format is “PIPE-001-TEMP1-202505121030”, where “PIPE-001” represents the pipeline number, “TEMP1” represents temperature sensor number 1, and “202505121030” is the data acquisition time. Furthermore, the system performs format validation on the parsed target feature data and unique identifiers to obtain the target feature data and their corresponding unique identifiers.

[0050] S402. Find the corresponding target pipeline GIS coordinates in the mapping table based on the unique identifier; In this embodiment, during the process of finding the GIS coordinates of the target pipeline, the system uses the unique identifier of the target feature data as a precise query keyword and performs a fast retrieval in the pre-constructed mapping table using database indexing technology. Since the identifier and the pipeline GIS coordinates have a many-to-one relationship, the mapping table uses key information of the pipeline GIS coordinates, such as pipeline number and sensor type, as index keys to associate multiple identifiers with different timestamps or parameter types with the pipeline GIS coordinates. Therefore, by parsing the key information in the unique identifier and matching it with the index keys, the system quickly locates the corresponding target pipeline GIS coordinates.

[0051] S403. Obtain pipeline location nodes in the 3D GIS model of the utility tunnel based on the GIS coordinates of the target pipeline. In this embodiment, after obtaining the GIS coordinates of the target pipeline, the system performs a spatial index query to generate a buffer range with a dynamic radius centered on the GIS coordinates of the target pipeline. Then, it uses spatial topology relationships to filter out the set of 3D GIS model nodes of the pipe gallery that intersect with it. When a unique node is directly matched, the pipeline location node is directly obtained. If there are multiple candidate nodes, such as at branch intersections, a secondary verification is performed using the pipeline spatial topology network. The pipeline flow direction is analyzed to determine whether the pipe segment to which the candidate node belongs is associated with the target pipeline ID, and the 3D spatial distance from each node to the target coordinates is calculated. The node with consistent topological connectivity and the shortest spatial distance is selected first. If no node is matched, path deduction is initiated based on the pipeline network spatial connectivity model. The Dijkstra algorithm is used to calculate the shortest path from the nearest known node to the theoretical location, and a temporary virtual node is generated and marked as to be mapped, thus obtaining the pipeline location node.

[0052] S404. Store the target feature data in the pipeline location node to generate the pipeline operation model.

[0053] In this embodiment, after determining the pipeline location nodes, the system parses the storage space structure of the pipeline location nodes. If the storage space structure already contains feature data, a rolling update mechanism is used to retain the three most recent data records. If the storage space structure does not contain feature data, data is stored directly at the pipeline location node. Next, a visualization update is triggered. If the target feature data is a temperature gradient sequence, it is directly mapped to the surface of the 3D GIS model of the pipeline corridor using red, yellow, and blue colors (e.g., 55 is displayed as red, 30 as green). If the target feature data is a hazardous substance diffusion distribution parameter, a semi-transparent overlay layer is generated, with the color darker as the concentration increases. If the target feature data is a pipeline surface feature image, the system compresses the denoised pipeline surface image into a thumbnail and generates a clickable floating icon at the corresponding pipeline location node. Clicking this icon pops up a comparison window displaying the difference between the current image and the historical baseline image. Furthermore, according to a preset global update rule, such as updating and rendering the target feature data within all pipeline location nodes in real time every 5 minutes, a dynamic pipeline operation model with low spatial deviation rate can be obtained.

[0054] In this embodiment, by extracting target feature data and corresponding unique identifiers from the pipeline status feature set, the system can accurately filter out key information and reduce redundant data interference. Next, the unique identifiers are used to quickly locate the target pipeline's GIS coordinates in the mapping table, establishing a precise association between data and spatial location, enabling spatial traceability of dispersed monitoring data. Subsequently, based on the target pipeline's GIS coordinates, pipeline location nodes are determined in the 3D GIS model of the utility tunnel, and the target feature data is stored in the corresponding pipeline location nodes. This generates a pipeline operation model that not only intuitively presents the current overall real-time operating status of the pipeline but also displays data change trends through a dynamic update mechanism, helping maintenance personnel to quickly locate abnormal points and predict potential risks.

[0055] Please see Figure 5 , Figure 5 Another embodiment of the visualization monitoring method provided in this application includes: S501. Compare the pipeline operation model with the risk threshold range to obtain abnormal data points and corresponding risk levels; In this embodiment, before acquiring abnormal data points, threshold ranges need to be pre-defined for different types of risks in the pipeline operation model. These thresholds are based on historical data statistics and industry standards. Next, the system compares each data point in the real-time updated pipeline operation model with its corresponding threshold range. For each data point, if its value exceeds the upper limit of the threshold, the system marks it as an abnormal data point. Simultaneously, based on the degree to which the threshold is exceeded, risk levels are classified according to preset rules. For example, exceeding the threshold by less than 10% is considered a minor risk, exceeding by 10% to 30% is a moderate risk, and exceeding by more than 30% is a severe risk. This quantifies the degree of risk and provides a clear basis for subsequent early warnings.

[0056] S502. Obtain the early warning instruction template from the early warning instruction template library according to the risk level; In this embodiment, after determining the risk level, the system first accesses a pre-stored warning instruction template library. This library is categorized by risk level, with each level corresponding to a standardized set of warning instruction templates, covering warning information frameworks for different scenarios. Next, the system uses the determined risk level as an index to quickly retrieve and extract the corresponding template from the warning instruction template library. For example, if the risk level is severe, the system will retrieve the template corresponding to severe risk, which includes emergency response requirements, key data display formats, and other content frameworks, providing standardized format support for generating specific warning instructions and ensuring the standardized and efficient output of warning information.

[0057] S503. Parse the warning instruction template to obtain placeholders and data format requirements; In this embodiment, after obtaining the warning instruction template, the template is parsed. The template contains placeholders, such as "[abnormal location]" and "[exceeding standard value]", which are used to fill in specific abnormal data later. Next, by scanning the warning instruction template to identify specific symbols and markers, all placeholders are quickly located. For placeholders of different data types, the system analyzes their contextual logic and clarifies the format requirements: for temperature data, two decimal places must be retained and the unit "" must be included. Image data must provide a web access link and specify the resolution; gas concentration data must be accurate to an integer and the unit "ppm" must be indicated. The system records these format requirements in detail to provide clear guidance for replacing placeholders in subsequent abnormal data, ensuring that the generated warning instructions are formatted correctly and accurate.

[0058] S504. Replace placeholders with abnormal data points according to data format requirements to generate hierarchical early warning instructions.

[0059] In this embodiment, before generating the tiered early warning instruction, the placeholders in the template are replaced sequentially for the marked abnormal data. Specifically, the coordinates of the anomaly are first filled into "[Abnormal Location]", and then abnormal temperature and gas concentration data are filled into "[Exceeding Value]". For image data, the system extracts network access links that meet the 1920×1080 resolution requirement from the storage path and replaces the "[Abnormal Image Link]" placeholder in the template.

[0060] After the replacement is completed, the system integrates all information and generates a complete graded early warning instruction that includes anomaly details, risk level, and handling suggestions, such as "[XX location] temperature reaches [85.00℃], gas concentration [200ppm], the abnormal situation has triggered a serious risk warning, see relevant on-site pictures [https: / / xxx.com / xxx.jpg], please evacuate surrounding personnel immediately, close nearby valves and arrange maintenance", so that it can be pushed to the terminals of relevant management personnel.

[0061] This embodiment compares pipeline operating status with risk thresholds, enabling accurate identification of abnormal data points and risk level classification based on historical data and industry standards. This avoids the subjectivity and lag of human judgment, allowing for timely detection of potential risks. Next, it retrieves early warning instruction templates based on risk levels, ensuring that different levels of risk have corresponding standardized information frameworks, improving the standardization and consistency of early warnings. Furthermore, it parses the template to obtain placeholders and data format requirements, ensuring that abnormal data is integrated into the template in a unified and accurate format, guaranteeing clear early warning instruction content. Finally, the generated tiered early warning instructions cover anomaly details, risk level, and handling suggestions, helping managers quickly grasp the core of the problem and take corresponding measures, effectively shortening risk response time, improving the efficiency and reliability of operation and maintenance management, and providing strong support for the safe and stable operation of pipelines.

[0062] Please see Figure 6 , Figure 6 One embodiment of the visual monitoring device provided in this application includes: The receiving unit 601 is used to receive multi-dimensional data on pipeline operating status collected by the multi-source sensing device; The processing unit 602 is used to process multidimensional data and generate a pipeline status feature set; The association unit 603 is used to dynamically associate the pipeline status feature set with the pipeline GIS coordinates to generate a pipeline operation model. The identification unit 604 is used to identify abnormal states of the pipeline operation model in order to obtain graded early warning instructions; The sending unit 605 is used to send the graded warning instruction to the target terminal according to the user's permissions, so that the user can perform corresponding operations according to the graded warning instruction obtained by the target terminal.

[0063] Optionally, the processing unit 602 is specifically used for: Data type identification for multidimensional data; If the data type is image data, then noise reduction processing is performed to obtain the pipeline surface feature image; If the data type is temperature data, then baseline calibration is performed to obtain the temperature gradient sequence; If the data type is gas concentration data, then spatial interpolation analysis is performed to obtain the diffusion distribution parameters of hazardous substances; By associating pipeline surface feature images, temperature gradient sequences, and hazardous substance diffusion distribution parameters with identifiers, a pipeline state feature set is generated.

[0064] Optionally, it also includes building unit 606, specifically used for: Obtain pipeline GIS coordinates from geographic information data; Construct a mapping table based on pipeline GIS coordinates and identifiers; Construct an initial 3D model; The initial 3D model is reconstructed based on the pipeline GIS coordinates and mapping table to generate a 3D GIS model of the utility tunnel.

[0065] Optionally, the associated unit 603 is also used for: Extract the target feature data and corresponding unique identifier from the pipeline status feature set; The corresponding target pipeline GIS coordinates are found in the mapping table based on the unique identifier. Obtain pipeline location nodes in the 3D GIS model of the utility tunnel based on the GIS coordinates of the target pipeline. The target feature data is stored in the pipeline location node to generate the pipeline operation model.

[0066] Optionally, the identification unit 604 is specifically used for: The pipeline operation model is compared with the preset risk threshold range to obtain abnormal data points and corresponding risk levels; Retrieve an early warning instruction template from the early warning instruction template library based on the risk level; Fill the abnormal data points into the early warning instruction template to obtain tiered early warning instructions.

[0067] Optionally, the identification unit 604 is further used for: Parse the warning instruction template to obtain placeholders and data format requirements; Replace placeholders with abnormal data points according to data format requirements to generate tiered early warning instructions.

[0068] Optionally, it also includes update unit 607, specifically used for: Receive the execution result returned by the target terminal and update the multidimensional data based on the execution result.

[0069] Please see Figure 7 , Figure 7 Another embodiment of a visual monitoring device provided in this application includes: Processor 701, memory 702, input / output unit 703, bus 704; The processor 701 is connected to the memory 702, the input / output unit 703, and the bus 704; The memory 702 stores a program, and the processor 701 calls the program to execute any of the methods described above.

[0070] This application also relates to a computer-readable storage medium on which a program is stored, which, when run on a computer, causes the computer to perform any of the methods described above.

[0071] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0072] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between apparatuses or units through some interfaces, and may be electrical, mechanical, or other forms.

[0073] The units described as separate components may or may not be physically separate. The components shown as units 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 units can be selected to achieve the purpose of this embodiment according to actual needs.

[0074] Furthermore, the functional units in the various embodiments of this application 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 as a software functional unit.

[0075] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, a system, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A visual monitoring method, characterized in that, The method includes: Receive multi-dimensional data on pipeline operating status collected by multi-source sensing devices; The multidimensional data is processed to generate a pipeline status feature set; The pipeline status feature set is dynamically associated with the pipeline GIS coordinates to generate a pipeline operation model. Identify abnormal states in the pipeline operation model to obtain tiered early warning instructions; The tiered warning instructions are sent to the target terminal according to user permissions, so that the user can perform corresponding operations based on the tiered warning instructions obtained by the target terminal.

2. The visualization monitoring method according to claim 1, characterized in that, The process of processing the multidimensional data to generate a pipeline status feature set includes: Data type identification is performed on the multidimensional data; If the data type is image data, then noise reduction processing is performed to obtain a pipeline surface feature image; If the data type is temperature data, then baseline calibration is performed to obtain a temperature gradient sequence; If the data type is gas concentration data, then spatial interpolation analysis is performed to obtain the diffusion distribution parameters of hazardous substances; The pipeline surface feature image, the temperature gradient sequence, and the hazardous substance diffusion distribution parameters are associated with identifiers to generate a pipeline state feature set.

3. The visualization monitoring method according to claim 2, characterized in that, Before dynamically associating the pipeline status feature set with pipeline GIS coordinates to generate a pipeline operation model, the method further includes: Obtain pipeline GIS coordinates from geographic information data; A mapping table is constructed based on the pipeline GIS coordinates and the identifier; Construct an initial 3D model; The initial 3D model is reconstructed based on the pipeline GIS coordinates and the mapping table to generate a 3D GIS model of the utility tunnel.

4. The visualization monitoring method according to claim 3, characterized in that, The step of dynamically associating the pipeline status feature set with pipeline GIS coordinates to generate a pipeline operation model includes: Extract the target feature data and corresponding unique identifier from the pipeline status feature set; The corresponding target pipeline GIS coordinates are retrieved from the mapping table based on the unique identifier. Based on the target pipeline's GIS coordinates, the pipeline location nodes are obtained in the 3D GIS model of the utility tunnel. The target feature data is stored in the pipeline location node to generate a pipeline operation model.

5. The visualization monitoring method according to claim 1, characterized in that, The process of identifying abnormal states of the pipeline operation model to obtain tiered early warning instructions includes: The pipeline operation model is compared with a preset risk threshold range to obtain abnormal data points and corresponding risk levels; Retrieve an early warning instruction template from the early warning instruction template library based on the aforementioned risk level; The abnormal data points are filled into the early warning instruction template to obtain tiered early warning instructions.

6. The visualization monitoring method according to claim 5, characterized in that, The step of filling the abnormal data points into the early warning instruction template to obtain tiered early warning instructions includes: Parse the warning instruction template to obtain placeholders and data format requirements; Replace the placeholders with the abnormal data points according to the data format requirements to generate a tiered early warning instruction.

7. The visualization monitoring method according to claim 1, characterized in that, After sending the tiered warning instruction to the target terminal according to user permissions, so that the user can perform corresponding operations according to the tiered warning instruction obtained by the target terminal, the method further includes: Receive the execution result returned by the target terminal, and update the multidimensional data according to the execution result.

8. A visual monitoring device, characterized in that, The device includes: The receiving unit is used to receive multi-dimensional data on pipeline operating status collected by multi-source sensing devices; The processing unit is used to process the multidimensional data and generate a pipeline status feature set; The association unit is used to dynamically associate the pipeline status feature set with the pipeline GIS coordinates to generate the pipeline operation status. The identification unit is used to identify abnormal states of the pipeline operation model in order to obtain graded early warning instructions; The sending unit is used to send the graded warning instruction to the target terminal according to the user's permissions, so that the user can perform corresponding operations according to the graded warning instruction obtained by the target terminal.

9. A visual monitoring device, characterized in that, The device includes: Processor, memory, input / output units, and bus; The processor is connected to the memory, the input / output unit, and the bus; The memory stores a program, which the processor invokes to perform the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium contains a program that, when executed on a computer, performs the method as described in any one of claims 1 to 7.

Images