Oil and gas pipeline monitoring method and monitoring system based on unmanned aerial vehicle inspection

Abstract

The invention discloses an oil and gas pipeline monitoring method and monitoring system based on unmanned aerial vehicle routing inspection, and relates to the technical field of oil and gas pipeline detection. Scene abnormality is generated according to collected scene data, and if the scene abnormality exceeds an abnormality threshold value, routing inspection is performed on an oil and gas pipeline by using a routing inspection unmanned aerial vehicle according to an optimized routing inspection path; identifying abnormal points on the oil and gas pipeline according to the obtained detection data, generating a maintenance value according to the abnormal state data of each abnormal point, determining a corresponding maintenance priority according to the maintenance value, and planning a corresponding maintenance path for each abnormal point according to the maintenance priority and the position information; and matching a corresponding maintenance scheme for the abnormal point by the oil and gas pipeline maintenance knowledge graph according to the fault feature, maintaining the abnormal point of the oil and gas pipeline, and polling the oil and gas pipeline at the time node conforming to the constraint condition. Abnormalities and faults existing in the oil and gas pipeline are overhauled and maintained, and the maintenance and overhaul efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of oil and gas pipeline detection, and in particular to an oil and gas pipeline monitoring method and a monitoring system based on unmanned aerial vehicle inspection. Background Art

[0002] Oil and gas pipeline inspection is a comprehensive technical activity that relies on a variety of high-tech means, such as ultrasonic testing, radiographic testing, magnetic flux leakage testing, electromagnetic acoustic wave testing, eddy current testing, and mechanical contact testing, etc., to conduct comprehensive and detailed inspections and assessments of the internal structure, external coating, weld quality, corrosion conditions, etc. of oil and gas pipelines. The purpose of this process is to promptly discover and accurately locate defects, damage or potential risks in the pipeline, and provide a scientific basis for subsequent maintenance, replacement or strengthening measures, thereby ensuring that the oil and gas pipeline maintains a safe, reliable and efficient operation state under complex natural environments and operating conditions. Through oil and gas pipeline inspection, not only can the occurrence of safety accidents such as pipeline leakage and explosion be effectively prevented, but also the service life of the pipeline can be extended, operating costs can be reduced, and public safety, environmental protection and the stability of energy supply can be guaranteed.

[0003] In the Chinese invention patent with application publication number CN118243782A, a weld detection method for oil and gas pipelines is provided, which can solve the technical problem of low defect detection rate of girth welds of oil and gas pipelines. The weld detection method for oil and gas pipelines includes confirming multiple first partitions of the first test area; adjusting the amplitude of the flat-bottom hole corresponding to each first partition according to the first preset condition to obtain the debugging result; the flat-bottom hole is set on the first detection comparison test block; according to the debugging result, each first partition is detected to obtain the first detection result; the first detection result is used to characterize the position of the signal relative to the groove; according to the first detection result and the second preset condition, the second detection result of the first test area is obtained; according to the second detection comparison test block, the third detection result of the second test area is confirmed; according to the third detection comparison test block, the fourth detection result of the third test area is confirmed. By covering the entire weld and the heat-affected zone, more defects are detected, thereby improving the defect detection rate of the girth weld of the oil and gas pipeline.

[0004] Combined with the above application and the contents of the prior art:

[0005] Considering the safety of oil and gas transportation, it is necessary to monitor the status of oil and gas pipelines regularly to verify whether there are any abnormalities or even failures in the oil and gas pipelines, such as deformation, damage or even leakage. If the safety hazards in the oil and gas pipelines cannot be discovered and repaired in time, when large-scale leakage of oil and gas in the oil and gas pipelines occurs, serious accidents and damage will occur.

[0006] The existing oil and gas pipeline monitoring methods are mainly based on manual periodic inspections, as well as the detection of key parts, such as pipeline connection parts, in conjunction with monitoring equipment and sensor networks. This monitoring method is less comprehensive and tends to neglect non-critical parts. Therefore, in order to reduce the neglect of non-critical parts, drone inspections are introduced on the basis of existing monitoring methods. The sensor network carried by the drone conducts a comprehensive inspection of the oil and gas pipeline. However, when the working scene of the oil and gas pipeline undergoes major changes, such as large changes in external temperature conditions and large changes in internal oil and gas pressure, the oil and gas pipeline is more likely to be affected by the properties of the material and is more prone to failures and abnormalities. At this time, the efficiency of comprehensive inspections of the oil and gas pipeline is low, and it is difficult to effectively reduce the risks that the oil and gas pipeline may face.

[0007] To this end, the present invention provides an oil and gas pipeline monitoring method and a monitoring system based on drone inspection. Summary of the invention

[0008] 1. Technical issues to be solved

[0009] In view of the deficiencies of the prior art, the present invention provides an oil and gas pipeline monitoring method and monitoring system based on drone inspection, which generates a scene abnormality from the collected scene data. If the scene abnormality exceeds the abnormal threshold, the inspection drone is used to inspect the oil and gas pipeline according to the optimized inspection path, and the abnormal points on the oil and gas pipeline are identified according to the acquired detection data. The inspection value is generated from the abnormal state data of each abnormal point, and the corresponding maintenance priority is determined according to the inspection value. The corresponding maintenance path is planned for each abnormal point according to the maintenance priority and location information; the oil and gas pipeline maintenance knowledge graph matches the abnormal point with the corresponding maintenance plan according to the fault characteristics, and the abnormal point of the oil and gas pipeline is maintained, and the oil and gas pipeline is inspected at the time node that meets the constraint conditions. The abnormalities and faults existing in the oil and gas pipeline are inspected and maintained to improve the efficiency of maintenance and inspection; thereby solving the technical problems recorded in the background technology.

[0010] (II) Technical solution

[0011] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0012] The oil and gas pipeline monitoring method based on drone inspection includes: collecting scene data in the installation area, generating a scene abnormality Htp from the collected scene data, and issuing a path planning instruction to the outside if the scene abnormality Htp exceeds an abnormal threshold;

[0013] Use inspection drones to inspect oil and gas pipelines according to optimized inspection routes, identify abnormal points on the oil and gas pipelines based on the acquired inspection data, and issue maintenance and evaluation instructions to external parties if the abnormality of the abnormal point exceeds expectations;

[0014] The maintenance value Jtp is generated from the abnormal state data of each abnormal point, the corresponding maintenance priority is determined according to the maintenance value Jtp, and the corresponding maintenance path is planned for each abnormal point according to the maintenance priority and location information; wherein, the abnormal score Yp and the scene abnormality Htp of the abnormal point are dimensionlessly processed, and the maintenance value Jtp is generated according to the following method:

[0015]

[0016] Weight coefficient: 0≤ρ≤1, 0≤ζ≤1, and ρ+ζ=1;

[0017] The oil and gas pipeline maintenance knowledge graph matches the corresponding maintenance plan for the abnormal point according to the fault characteristics. After iterative optimization of the maintenance plan, the maintenance plan is executed to maintain the abnormal point of the oil and gas pipeline.

[0018] The inspection drone is enabled to inspect the oil and gas pipeline at a time node that meets the constraints, predict the abnormal points of the oil and gas pipeline, and insert the predicted abnormal points into the maintenance path at the next inspection node.

[0019] Furthermore, an electronic map covering the installation area is constructed, and scene data is collected in the installation area, including temperature, humidity data and fluid pressure fluctuation data in the oil and gas pipeline, and the detection scene data set is summarized and generated. The scene anomaly degree Htp is generated from the scene data in the scene data set.

[0020] Furthermore, the temperature Tp and fluid pressure Sp are linearly normalized, and the corresponding data values ​​are mapped to the interval [0,1], and the scene abnormality Htp is generated according to the following formula:

[0021]

[0022] Weight coefficient: 0≤β≤1, 0≤α≤1, and α+β=1, i=2,…,k, where k is the number of time nodes; is the mean value of the fluid pressure, is the qualified standard value of fluid pressure, is the humidity of the i-th fluid pressure; is the mean temperature, is the qualified standard value of temperature, is the temperature at the i-th monitoring point.

[0023] Furthermore, the pipeline location information, terrain information and building information in the installation area are marked on the electronic map. After receiving the path planning instruction, the pre-trained path planning algorithm plans the inspection path for the inspection drone;

[0024] Build a digital twin model for drone inspection, use the digital twin model for drone inspection to test the inspection path, and obtain corresponding test data.

[0025] Furthermore, the inspection path is iteratively optimized with the test data to obtain the optimized inspection path; a sensor network is deployed on the inspection drone, and the inspection drone is used to inspect the oil and gas pipeline according to the optimized inspection path. The sensor network collects the inspection data in real time and aggregates it to generate a pipeline inspection data set;

[0026] Taking pipeline inspection data as input, the trained anomaly detection model is used to perform anomaly detection on the oil and gas pipeline, identify abnormal points on the oil and gas pipeline and provide anomaly scores.

[0027] Furthermore, the position of each abnormal point is marked on the oil and gas pipeline model. After receiving the maintenance evaluation instruction, the abnormal status data of each abnormal point is summarized to generate an abnormal status data set, and the maintenance value Jtp is generated from the abnormal status data set; the obtained maintenance values ​​are arranged in order from large to small to obtain the corresponding maintenance sequence; the corresponding maintenance priority is determined by the order of the abnormal points in the maintenance sequence, and the pre-trained path planning algorithm plans the corresponding maintenance path for each abnormal point according to the maintenance priority and location information of the abnormal point.

[0028] Furthermore, the corresponding fault features are obtained after feature extraction of the detection data of the abnormal points, and the oil and gas pipeline maintenance is used as the target word to pre-construct the oil and gas pipeline maintenance knowledge graph;

[0029] The oil and gas pipeline maintenance knowledge graph matches the corresponding maintenance plan for the abnormal point, and the oil and gas pipeline maintenance digital twin model tests the maintenance plan for the abnormal point and obtains the corresponding test data. The maintenance plan is iteratively optimized based on the test data, and the maintenance plan is executed to maintain the abnormal point of the oil and gas pipeline.

[0030] Furthermore, after executing the maintenance plan to maintain the abnormal store, the inspection interval Pa of the drone inspection is constrained, and the inspection is carried out on the inspection nodes that meet the constraints. After determining the next inspection node of the drone, the trained fault prediction model is used to predict the fault of the oil and gas pipeline at the inspection node.

[0031] Furthermore, the inspection interval Pa is constrained as follows:

[0032]

[0033] Weight coefficient: 0≤β≤1, 0≤α≤1, and α+β=1, where n is the number of outliers, U(i,j) is the distance from the i-th outlier to the j-th outlier, and Ut a is the average value of the distance, Htpmax 、Htp min They are respectively the maximum and minimum values ​​of the scene abnormality in the previous inspection cycle.

[0034] The oil and gas pipeline monitoring system based on drone inspection includes a scene analysis unit, which collects scene data in the installation area, generates a scene abnormality Htp from the collected scene data, and sends a path planning instruction to the outside if the scene abnormality Htp exceeds the abnormality threshold;

[0035] The abnormality analysis unit uses inspection drones to inspect oil and gas pipelines according to the optimized inspection path, identifies abnormal points on the oil and gas pipelines based on the acquired inspection data, and issues maintenance and evaluation instructions to the outside if the abnormality of the abnormal point exceeds expectations;

[0036] The maintenance path planning unit generates a maintenance value Jtp from the abnormal state data of each abnormal point, determines the corresponding maintenance priority according to the maintenance value Jtp, and plans the corresponding maintenance path for each abnormal point according to the maintenance priority and location information;

[0037] The maintenance plan matching unit matches the corresponding maintenance plan for the abnormal point based on the oil and gas pipeline maintenance knowledge graph according to the fault characteristics. After iterative optimization of the maintenance plan, the maintenance plan is executed to maintain the abnormal point of the oil and gas pipeline.

[0038] The pre-maintenance unit enables the inspection drone to inspect the oil and gas pipeline at a time node that meets the constraint conditions, predict the abnormal points of the oil and gas pipeline, and insert the predicted abnormal points into the maintenance path at the next inspection node.

[0039] (III) Beneficial effects

[0040] The present invention provides an oil and gas pipeline monitoring method and monitoring system based on drone inspection, which has the following beneficial effects:

[0041] 1. Generate the scene abnormality Htp from the collected data, analyze and judge whether there is any abnormality in the environment in the installation area based on the scene abnormality Htp, and predict the risk of abnormality in the oil and gas pipeline.

[0042] 2. The drone plans the inspection route and tests it to ensure the effectiveness and safety of the inspection route and improve the efficiency of the inspection. The abnormality of the oil and gas pipeline is evaluated and judged based on the acquired detection data. Under the current abnormal scenario conditions, the identified abnormal points can be processed in a targeted manner.

[0043] 3. By sorting several test values, determining the maintenance order and corresponding maintenance priority, the abnormalities and faults in the oil and gas pipelines can be inspected and maintained, and the efficiency of maintenance and repair can be improved. By planning the maintenance path, the efficiency of maintenance can be improved when the oil and gas pipelines are maintained according to the maintenance path.

[0044] 4. Test, optimize and improve the maintenance plan to improve its reliability and effectiveness, so as to achieve better maintenance results when maintaining oil and gas pipelines.

[0045] 5. Constrain and adjust the inspection frequency and time interval according to the current scene condition changes and the density of existing abnormal points, so that the inspection cycle is more consistent with the current actual situation and status of the oil and gas pipeline, which can improve the overall inspection efficiency.

[0046] 6. By predicting possible anomalies and faults in oil and gas pipelines, adding the predicted abnormal points to the next inspection node, and realizing early detection, it is possible to avoid the current possible faults from becoming a reality. On the basis of inspecting the oil and gas pipelines, prevention of oil and gas pipelines can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a flow chart of the oil and gas pipeline monitoring method based on drone inspection of the present invention;

[0048] Figure 2 The figure is a schematic diagram of the structure of the oil and gas pipeline monitoring system based on drone inspection of the present invention. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] See also Figure 1 The present invention provides an oil and gas pipeline monitoring method based on drone inspection, comprising:

[0051] Step 1: collect scene data in the installation area, generate scene abnormality Htp from the collected scene data, and if the scene abnormality Htp exceeds the abnormal threshold, issue a path planning instruction to the outside;

[0052] The step 1 includes the following contents:

[0053] Step 101: After the oil and gas pipeline is located, the area is taken as the installation area and an electronic map covering the installation area is constructed. Scene data is collected in the installation area, including temperature, humidity data and fluid pressure fluctuation data in the oil and gas pipeline, and the detection scene data set is generated by summarizing;

[0054] Step 102: After presetting corresponding thresholds for various scene condition parameters, the scene abnormality Htp is generated from the scene data in the scene data set, wherein the temperature Tp of the installation area where the oil and gas pipeline is located and the fluid pressure Sp inside the oil and gas pipeline are linearly normalized, and the corresponding data values ​​are mapped to the interval [0,1] according to the following formula:

[0055]

[0056] Weight coefficient: 0≤β≤1, 0≤α≤1, and α+β=1. The weight coefficient can be obtained by referring to the hierarchical analysis method;

[0057] i=2,…,k,k is the number of time nodes, is the mean value of the fluid pressure, is the qualified standard value of fluid pressure, is the humidity of the i-th fluid pressure; is the mean temperature, is the qualified standard value of temperature, is the temperature at the i-th monitoring point;

[0058] Based on historical data and expectations for oil and gas pipeline fault management, an abnormal threshold is pre-set; if the scene abnormality Htp exceeds the abnormal threshold, it means that the current environmental scene of the oil and gas pipeline may have a certain degree of impact on the oil and gas pipeline, for example, causing deformation or corrosion of the oil and gas pipeline, etc., and the oil and gas pipeline needs to be monitored and inspected to prevent the oil and gas pipeline from malfunctioning and causing safety hazards. At this time, a path planning instruction is issued to the outside;

[0059] When using, combine the contents in steps 101 and 102:

[0060] Before fault and anomaly detection of oil and gas pipelines, the environmental condition data of the area where they are located is first collected and analyzed, and the scene anomaly degree Htp is generated from the collected data. Based on the scene anomaly degree Htp, it is analyzed and judged whether there is any abnormality in the environment in the installation area. Considering that abnormal environmental conditions will affect and damage oil and gas pipelines, the collection scene and environmental conditions are used as the starting point to predict the risk of abnormalities in oil and gas pipelines.

[0061] The existing oil and gas pipeline monitoring methods are mainly based on manual periodic inspections, as well as the detection of key parts, such as pipeline connection parts, in conjunction with monitoring equipment and sensor networks. This monitoring method is less comprehensive and tends to neglect non-critical parts. Therefore, in order to reduce the neglect of non-critical parts, drone inspections are introduced on the basis of existing monitoring methods. The sensor network carried by the drone conducts a comprehensive inspection of the oil and gas pipeline. However, when the working scene of the oil and gas pipeline undergoes major changes, such as large changes in external temperature conditions and large changes in internal oil and gas pressure, the oil and gas pipeline is more likely to be affected by the properties of the material and is more prone to failures and abnormalities. At this time, the efficiency of comprehensive inspections of the oil and gas pipeline is low, and it is difficult to effectively reduce the risks that the oil and gas pipeline may face.

[0062] Step 2: Use the inspection drone to inspect the oil and gas pipeline according to the optimized inspection route, identify abnormal points on the oil and gas pipeline based on the detection data, and issue an inspection and evaluation instruction to the outside if the abnormality of the abnormal point exceeds expectations;

[0063] The step 2 includes the following contents:

[0064] Step 201: Mark the pipeline location information, terrain information, and building information in the installation area on the electronic map. After receiving the path planning instruction, the pre-trained path planning algorithm plans the inspection path for the inspection drone.

[0065] The drone inspection digital twin model is constructed based on the inspection drone’s performance and flight data, terrain and building data in the installation area, etc. The drone inspection digital twin model is used to test the inspection path and obtain corresponding test data, such as the inspection path time consumption and flight feasibility.

[0066] When in use, when the inspection conditions are met, planning the inspection path for the drone and testing it can ensure the effectiveness and safety of the inspection path and improve the efficiency of the inspection;

[0067] Step 202: iteratively optimize the inspection path using the test data to obtain an optimized inspection path;

[0068] A sensor network is deployed on the inspection drone, including a multispectral camera, ultrasonic sensors, and high-sensitivity gas detectors. The inspection drone is used to inspect the oil and gas pipeline according to the optimized inspection path. The sensor network collects detection data in real time, such as image data of the oil and gas pipeline, pipeline material data, ultrasonic alarm data, and gas leakage data. The pipeline detection data set is aggregated and generated.

[0069] Step 203: training a machine learning algorithm with the labeled sample data to obtain a trained anomaly detection model;

[0070] Taking real-time pipeline inspection data as input, the trained anomaly detection model is used to perform anomaly detection on the oil and gas pipeline, identify the abnormal points on the oil and gas pipeline, and the abnormal state of the abnormal points, and output an anomaly score describing the corresponding degree of abnormality. If the anomaly score exceeds the preset score threshold, an external maintenance assessment instruction is issued;

[0071] When using, combine the contents in steps 201 to 203:

[0072] As further feedback on the inspection route, drones equipped with sensor networks can inspect and collect data on the oil and gas pipelines along the planned inspection route. The abnormalities of the oil and gas pipelines can be evaluated and judged based on the acquired detection data, completing the preliminary inspection process of the oil and gas pipelines. As corresponding feedback, targeted processing can be performed on the identified abnormal points under the current abnormal scenario conditions.

[0073] Step 3: Generate a maintenance value Jtp from the abnormal state data of each abnormal point, determine the corresponding maintenance priority according to the maintenance value Jtp, and plan a corresponding maintenance path for each abnormal point according to the maintenance priority and location information;

[0074] The step three includes the following contents:

[0075] Step 301: construct an oil and gas pipeline model in the installation area, mark the location of each abnormal point on the oil and gas pipeline model, and after receiving the maintenance evaluation instruction, summarize the abnormal state data of each abnormal point, including the abnormal degree of the abnormal point and the corresponding scene data, to generate an abnormal state data set;

[0076] Step 302: Generate a maintenance value Jtp from the abnormal state data set, wherein the abnormal score Yp of the abnormal point and the scene abnormality Htp are dimensionlessly processed according to the following method:

[0077]

[0078] Weight coefficient: 0≤ρ≤1, 0≤ζ≤1, and ρ+ζ=1; the obtained maintenance values ​​are arranged in order from large to small to obtain the corresponding maintenance sequence;

[0079] When in use, based on the analysis and identification of the detection data, the maintenance value is generated by the corresponding abnormal status data set. According to the maintenance value, a comprehensive judgment can be made on the abnormal degree of the abnormal point and the risk of further abnormalities. By sorting several detection values, the maintenance sequence and the corresponding maintenance priority are determined. The abnormalities and faults in the oil and gas pipelines can be inspected and maintained, thereby improving the efficiency of maintenance and repair.

[0080] Step 303: Determine the corresponding maintenance priority according to the order of the abnormal points in the maintenance sequence, plan the corresponding maintenance path for each abnormal point according to the maintenance priority and location information of the abnormal point by using the pre-trained path planning algorithm, and mark the corresponding maintenance path on the oil and gas pipeline model;

[0081] When using, combine the contents in steps 301 to 303:

[0082] As a further content, after obtaining the location information and maintenance priority of each abnormal point, by planning the maintenance path, the maintenance efficiency of the oil and gas pipeline can be improved according to the maintenance path.

[0083] Step 4: The oil and gas pipeline maintenance knowledge graph matches the corresponding maintenance plan for the abnormal point according to the fault characteristics. After iterative optimization of the maintenance plan, the maintenance plan is executed to maintain the abnormal point of the oil and gas pipeline;

[0084] The step 4 includes the following contents:

[0085] Step 401: extract features from the detection data of abnormal points to obtain corresponding fault features, take oil and gas pipeline inspection and maintenance as the target word, and pre-build the oil and gas pipeline maintenance knowledge graph after deep retrieval and entity relationship building;

[0086] According to the correspondence between the fault characteristics of the abnormal point and the inspection and maintenance plan, the oil and gas pipeline maintenance knowledge map matches the corresponding maintenance plan for the abnormal point;

[0087] When used, as further feedback, by extracting features from the detection data at the abnormal points, the oil and gas pipeline maintenance knowledge graph matches the corresponding maintenance plan for the abnormal points, so that targeted maintenance of the oil and gas pipelines can be achieved when abnormalities and failures occur in the oil and gas pipelines;

[0088] Step 402: constructing a digital twin model of oil and gas pipeline maintenance based on relevant data of the oil and gas pipeline, such as material structure data, specification data, and operation status data, and using the digital twin model of oil and gas pipeline maintenance to test the maintenance plan of the abnormal point and obtain corresponding test data, such as the time consumption, effectiveness, and reliability of the inspection and maintenance;

[0089] If it is not feasible, the maintenance plan is iteratively optimized based on the test data as feedback to obtain a feasible maintenance plan, and the maintenance plan is executed to maintain the abnormal points of the oil and gas pipeline;

[0090] When using, combine the contents in steps 401 and 402:

[0091] As a further content, when a maintenance plan for an oil and gas pipeline is obtained, the maintenance plan can be tested, optimized and improved to further improve the reliability and effectiveness of the maintenance plan, and better maintenance results can be achieved when maintaining the oil and gas pipeline.

[0092] Step 5: Make the inspection drone inspect the oil and gas pipeline at a time node that meets the constraint conditions, predict the abnormal points of the oil and gas pipeline, and insert the predicted abnormal points into the maintenance path at the next inspection node;

[0093] The step five includes the following contents:

[0094] Step 501: After executing the maintenance plan to maintain the abnormal store, under dimensionless conditions, the inspection interval Pa of the drone inspection is constrained, and the inspection is performed on the inspection nodes that meet the constraint conditions. The constraint method of the inspection interval Pa is as follows:

[0095]

[0096] Weight coefficient: 0≤β≤1, 0≤α≤1, and α+β=1, the weight coefficient is consistent with the previous value;

[0097] Where n is the number of outliers, U(i,j) is the distance from the i-th outlier to the j-th outlier, and Ut a is the average value of the distance, Htp max 、Htp min They are the maximum and minimum values ​​of the scene abnormality in the previous inspection cycle respectively;

[0098] When using it, considering that continuous inspections are difficult, the periodic use of drone inspections is more in line with the use of oil and gas pipelines. Therefore, as a further content, the frequency and time interval of inspections are constrained and adjusted according to the current changes in scene conditions and the density of existing abnormal points, so that the inspection cycle is more in line with the current actual situation and status of the oil and gas pipeline, which can improve the overall efficiency of the inspection.

[0099] Step 502: train a convolutional neural network with the labeled sample data to obtain a trained fault prediction model; after determining the next inspection node of the drone, use the scene condition data and operation data of the oil and gas pipeline as input, use the trained fault prediction model to predict the fault of the oil and gas pipeline at the inspection node, obtain the predicted fault, insert the possible predicted fault into the maintenance path at the next inspection node, and update the maintenance path in advance;

[0100] When using, combine the contents in steps 501 and 502:

[0101] As further feedback, based on the acquisition of various real-time and historical data, the possible anomalies and faults of the oil and gas pipelines are predicted, and comprehensive prediction results are obtained. The prediction results are evaluated in the same logical way. By adding the predicted abnormal points to the next inspection node, the oil and gas pipelines can be inspected in the next inspection cycle to achieve early detection, which can avoid the current possible faults from becoming a reality. On the basis of the inspection of the oil and gas pipelines, prevention of the oil and gas pipelines can be achieved, completing the closed-loop construction.

[0102] The analytic hierarchy process (AHP) is a structured and systematic decision support technology. It decomposes a much more complex standard decision problem into a series of interrelated and hierarchical sub-problems or factors, uses expert judgment to compare and quantitatively evaluate the importance of factors at each level, and finally obtains the relative advantages and disadvantages of each decision plan through synthetic calculation, providing decision makers with a clear and quantitative decision basis. This method combines qualitative and quantitative analysis, effectively improving the scientificity and accuracy of decision-making.

[0103] The construction method of the oil and gas pipeline maintenance knowledge graph can refer to the following:

[0104] Extensive data collection: Data related to oil and gas pipeline maintenance is collected from various documents, technical reports, accident cases, industry specifications, expert interviews, and real-time monitoring systems. This includes pipeline materials, design parameters, operating environment, historical fault records, maintenance history, etc. Data cleaning and integration: Clean the collected data to remove duplicate, incomplete or erroneous information. Then, integrate the data according to its nature and purpose to form a unified data format and standard.

[0105] Domain ontology development: Based on the expertise of oil and gas pipeline maintenance, develop a domain ontology containing key concepts, attributes, relationships, and constraints. This helps to clarify the structure and hierarchy of knowledge and lays the foundation for subsequent knowledge extraction and storage. Knowledge model design: Design a knowledge representation model suitable for oil and gas pipeline maintenance scenarios, such as a graph-based data model (such as RDF, OWL, etc.) to effectively store and query complex knowledge relationships.

[0106] Automatic and semi-automatic extraction: Use natural language processing (NLP) technology, such as named entity recognition (NER) and relationship extraction, to automatically extract entities and relationships from text data. At the same time, combine manual review and annotation to improve the accuracy and completeness of extraction. Knowledge fusion and verification: Integrate the extracted knowledge with the existing knowledge base to solve knowledge conflicts and redundancy problems. At the same time, ensure the accuracy and consistency of knowledge through logical reasoning and verification.

[0107] Graph database selection and configuration: Select a graph database suitable for oil and gas pipeline maintenance scenarios (such as Neo4j, Virtuoso, etc.), and configure and optimize it according to the data scale and query requirements. Knowledge graph construction: Import the extracted and fused knowledge into the graph database in the form of triples to build the basic framework of the knowledge graph. Optimization and expansion: According to actual application requirements, the knowledge graph is optimized for performance (such as indexing, caching, etc.) and expanded (such as adding new entity types, relationship types, etc.).

[0108] Application scenario development: According to the actual needs of oil and gas pipeline maintenance, develop application scenarios based on knowledge graphs, such as fault diagnosis, risk assessment, maintenance strategy optimization, etc. User interface and interaction design: Design intuitive and easy-to-use user interfaces and interaction methods so that non-professionals can easily use knowledge graphs for decision-making and query. Effect evaluation and feedback: Through actual applications and user feedback, evaluate the accuracy and practicality of knowledge graphs, and make necessary adjustments and optimizations based on the evaluation results.

[0109] See also Figure 2 The present invention provides an oil and gas pipeline monitoring system based on drone inspection, comprising:

[0110] The scene analysis unit collects scene data in the installation area, generates a scene abnormality Htp from the collected scene data, and issues a path planning instruction to the outside if the scene abnormality Htp exceeds an abnormal threshold;

[0111] The abnormality analysis unit uses inspection drones to inspect oil and gas pipelines according to the optimized inspection path, identifies abnormal points on the oil and gas pipelines based on the acquired inspection data, and issues maintenance and evaluation instructions to the outside if the abnormality of the abnormal point exceeds expectations;

[0112] The maintenance path planning unit generates a maintenance value Jtp from the abnormal state data of each abnormal point, determines the corresponding maintenance priority according to the maintenance value Jtp, and plans the corresponding maintenance path for each abnormal point according to the maintenance priority and location information;

[0113] The maintenance plan matching unit matches the corresponding maintenance plan for the abnormal point based on the oil and gas pipeline maintenance knowledge graph according to the fault characteristics. After iterative optimization of the maintenance plan, the maintenance plan is executed to maintain the abnormal point of the oil and gas pipeline.

[0114] The pre-maintenance unit enables the inspection drone to inspect the oil and gas pipeline at a time node that meets the constraint conditions, predict the abnormal points of the oil and gas pipeline, and insert the predicted abnormal points into the maintenance path at the next inspection node.

[0115] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0116] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0117] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only some logical function divisions. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0118] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0119] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. The oil and gas pipeline monitoring method based on drone inspection is characterized by: include, Scene data is collected in the installation area, and the scene abnormality Htp is generated from the collected scene data. If the scene abnormality Htp exceeds the abnormality threshold, a path planning instruction is issued to the outside; Use inspection drones to inspect oil and gas pipelines according to optimized inspection routes, identify abnormal points on the oil and gas pipelines based on the acquired inspection data, and issue maintenance and evaluation instructions to external parties if the abnormality of the abnormal point exceeds expectations; The maintenance value Jtp is generated from the abnormal state data of each abnormal point, the corresponding maintenance priority is determined according to the maintenance value Jtp, and the corresponding maintenance path is planned for each abnormal point according to the maintenance priority and location information; wherein, the abnormal score Yp and the scene abnormality Htp of the abnormal point are dimensionlessly processed, and the maintenance value Jtp is generated according to the following method: Weight coefficient: 0≤ρ≤1, 0≤ζ≤1, and ρ+ζ=1; The oil and gas pipeline maintenance knowledge graph matches the corresponding maintenance plan for the abnormal point according to the fault characteristics. After iterative optimization of the maintenance plan, the maintenance plan is executed to maintain the abnormal point of the oil and gas pipeline. The inspection drone is enabled to inspect the oil and gas pipeline at a time node that meets the constraints, predict the abnormal points of the oil and gas pipeline, and insert the predicted abnormal points into the maintenance path at the next inspection node.

2. The oil and gas pipeline monitoring method based on drone inspection according to claim 1 is characterized in that: Construct an electronic map covering the installation area, collect scene data in the installation area, including temperature, humidity data and fluid pressure fluctuation data in the oil and gas pipeline, summarize and generate a detection scene data set, and generate the scene anomaly degree Htp from the scene data in the scene data set.

3. The oil and gas pipeline monitoring method based on drone inspection according to claim 2 is characterized in that: Perform linear normalization on the temperature Tp and fluid pressure Sp, and map the corresponding data values ​​to the interval [0,1], and generate the scene abnormality Htp according to the following formula: Weight coefficient: 0≤β≤1, 0≤α≤1, and α+β=1, i=2,…,k, k is the number of time nodes; Sp a is the mean value of the fluid pressure, is the qualified standard value of fluid pressure, Sp i is the humidity of the i-th fluid pressure; Tp a is the mean temperature, is the qualified standard value of temperature, Tp i is the temperature at the i-th monitoring point.

4. The oil and gas pipeline monitoring method based on drone inspection according to claim 3 is characterized in that: The pipeline location information, terrain information and building information in the installation area are marked on the electronic map. After receiving the path planning instruction, the pre-trained path planning algorithm plans the inspection path for the inspection drone. Build a digital twin model for drone inspection, use the digital twin model for drone inspection to test the inspection path, and obtain corresponding test data.

5. The oil and gas pipeline monitoring method based on drone inspection according to claim 4 is characterized in that: Iteratively optimize the inspection path using test data to obtain the optimized inspection path; A sensor network is deployed on the inspection drone, and the inspection drone is used to inspect the oil and gas pipeline according to the optimized inspection path. The sensor network collects the inspection data in real time and aggregates it to generate a pipeline inspection data set. Taking pipeline inspection data as input, the trained anomaly detection model is used to perform anomaly detection on the oil and gas pipeline, identify abnormal points on the oil and gas pipeline and provide anomaly scores.

6. The oil and gas pipeline monitoring method based on drone inspection according to claim 5 is characterized in that: The positions of each abnormal point are marked on the oil and gas pipeline model. After receiving the maintenance evaluation instruction, the abnormal state data of each abnormal point is summarized to generate an abnormal state data set, and the maintenance value Jtp is generated from the abnormal state data set; Arrange the obtained maintenance values ​​in order from large to small to obtain the corresponding maintenance sequence; The corresponding maintenance priority is determined by the order of the abnormal points in the maintenance sequence. According to the maintenance priority and location information of the abnormal points, the pre-trained path planning algorithm plans the corresponding maintenance path for each abnormal point.

7. The oil and gas pipeline monitoring method based on drone inspection according to claim 6 is characterized by: After extracting features from the detection data of abnormal points, the corresponding fault features are obtained. With oil and gas pipeline inspection and maintenance as the target word, a knowledge graph of oil and gas pipeline maintenance is pre-built. The oil and gas pipeline maintenance knowledge graph matches the corresponding maintenance plan for the abnormal point, and the oil and gas pipeline maintenance digital twin model tests the maintenance plan for the abnormal point and obtains the corresponding test data. The maintenance plan is iteratively optimized based on the test data, and the maintenance plan is executed to maintain the abnormal point of the oil and gas pipeline.

8. The oil and gas pipeline monitoring method based on drone inspection according to claim 7 is characterized in that: After executing the maintenance plan to maintain the abnormal store, the inspection interval Pa of the drone inspection is constrained, and the inspection is carried out on the inspection nodes that meet the constraints. After determining the next inspection node of the drone, the trained fault prediction model is used to predict the fault of the oil and gas pipeline at the inspection node.

9. The oil and gas pipeline monitoring method based on drone inspection according to claim 8 is characterized in that: The inspection interval Pa is constrained as follows: Weight coefficient: 0≤β≤1, 0≤α≤1, and α+β=1, where n is the number of outliers, U(i,j) is the distance from the i-th outlier to the j-th outlier, and Ut a is the average value of the distance, Htp max 、Htp min They are respectively the maximum and minimum values ​​of the scene abnormality in the previous inspection cycle.

10. The oil and gas pipeline monitoring system based on drone inspection is characterized by: include, The scene analysis unit collects scene data in the installation area, generates a scene abnormality Htp from the collected scene data, and issues a path planning instruction to the outside if the scene abnormality Htp exceeds an abnormal threshold; The abnormality analysis unit uses inspection drones to inspect oil and gas pipelines according to the optimized inspection path, identifies abnormal points on the oil and gas pipelines based on the acquired inspection data, and issues maintenance evaluation instructions to the outside if the abnormality of the abnormal point exceeds expectations; The maintenance path planning unit generates a maintenance value Jtp from the abnormal state data of each abnormal point, determines the corresponding maintenance priority according to the maintenance value Jtp, and plans the corresponding maintenance path for each abnormal point according to the maintenance priority and location information; The maintenance plan matching unit matches the corresponding maintenance plan for the abnormal point based on the oil and gas pipeline maintenance knowledge graph according to the fault characteristics. After iterative optimization of the maintenance plan, the maintenance plan is executed to maintain the abnormal point of the oil and gas pipeline. The pre-maintenance unit enables the inspection drone to inspect the oil and gas pipeline at a time node that meets the constraint conditions, predict the abnormal points of the oil and gas pipeline, and insert the predicted abnormal points into the maintenance path at the next inspection node.

Images