Risk monitoring and early warning method and system for long-distance oil and gas pipelines

Abstract

The present disclosure relates to the field of pipeline safety monitoring and provides a method and system for monitoring and early warning risks in long-distance oil and gas pipelines. The method comprises: collecting satellite imagery of long-distance oil and gas pipelines and identifying potential geological hazard areas; collecting surface imagery of potential geological hazard areas and identifying potential risk areas; collecting actual geological environmental information of potential risk areas; predicting environmental impact values using a trained neural network model based on the potential geological hazard areas, potential risk areas, and actual geological environmental information; acquiring pipeline monitoring data; determining a risk assessment value for the potential risk area based on the environmental impact value and pipeline monitoring data; generating a Class I or Class II risk signal for the potential risk area based on the risk assessment value, and determining and issuing an early warning signal corresponding to the potential risk area. The present disclosure can quickly identify and locate potential geological hazard areas, enabling efficient monitoring and early warning of geological risks in long-distance oil and gas pipelines.

Description

Technical Field

[0001] The present disclosure relates to the technical field of pipeline safety monitoring, and in particular to a risk monitoring and early warning method and system for long-distance oil and gas pipelines. Background Art

[0002] Long-distance oil and gas pipelines are a vital means of energy transportation. However, they often traverse complex terrain and geological environments, exposing them to various geological hazards such as landslides, debris flows, earthquakes, and subsidence. These hazards can not only cause pipeline ruptures or leaks but can also lead to major environmental pollution and safety accidents. Therefore, timely and accurate monitoring and early warning of geological risks are crucial to ensuring the safe operation of long-distance oil and gas pipelines.

[0003] Traditional methods for monitoring long-distance oil and gas pipelines rely primarily on ground inspections and fixed sensors. Ground inspections involve regular manual inspections along the pipeline to identify potential safety hazards, leaks, or other anomalies. Fixed sensors are various types of sensors installed at key locations along the pipeline to monitor the pipeline's operating status and surrounding environmental changes in real time.

[0004] However, fixed sensors have limited monitoring areas, making it difficult to fully cover the complex terrain and geological environment along the entire pipeline route. Ground inspections are limited in frequency, making it difficult to achieve real-time monitoring and rapid response to sudden geological disasters. Inadequate integration and analysis between disparate data sources prevents the full utilization of comprehensive information from multiple sources to improve the accuracy of risk identification. Summary of the Invention

[0005] The present disclosure aims to solve at least one of the problems existing in the prior art and provide a method and system for monitoring and early warning of risks in long-distance oil and gas pipelines. By collecting and processing data from different sensors and monitoring stations in real time, it is possible to monitor a wider area along the oil and gas pipeline, overcoming the limitations of traditional monitoring methods in geographical coverage; improving the real-time nature of monitoring, and through rapid analysis and processing of data, timely identifying early signs of geological disasters, thereby achieving a rapid response to sudden geological disaster events; by integrating data from different sources, enhancing the comprehensive analysis capability of data and improving the accuracy of risk assessment; through advanced algorithms and models, improving the ability to identify and warn of geological disasters in the early stage, reducing the risk of pipeline rupture and leakage caused by geological disasters, ensuring the safe operation of oil and gas pipelines, and supporting energy security and stable economic development.

[0006] One aspect of the present disclosure provides a method for monitoring and early warning risks in a long-distance oil and gas pipeline, the method comprising:

[0007] Collect satellite image information of each transport sub-area of the long-distance oil and gas pipeline, and determine potential geological disaster areas based on the satellite image information; collect surface image information of the potential geological disaster areas, and determine potential risk areas based on the surface image information; collect actual geological environment information of the potential risk areas;

[0008] Based on the potential geological disaster area, the potential risk area and the actual geological environment information, using the trained neural network model to predict the environmental impact value of the potential risk area;

[0009] Obtaining pipeline monitoring data in the potential risk area;

[0010] Determine a risk assessment value for the potential risk area based on the environmental impact value and the pipeline monitoring data; determine whether the risk assessment value is greater than a preset risk threshold; if so, generate a Class I risk signal for the potential risk area; otherwise, generate a Class II risk signal for the potential risk area; wherein the Class I risk signal is used to indicate that a risk is clearly present in the potential risk area, and the Class II risk signal is used to indicate that a potential risk hazard may exist in the potential risk area;

[0011] Based on the first-class risk signal and the second-class risk signal, a warning signal corresponding to the potential risk area is determined and issued.

[0012] Optionally, determining a potential geological disaster area based on the satellite image information includes:

[0013] Determine, based on the satellite image information, an actual abnormal data group for each geological disaster unit corresponding to each of the transport sub-areas; the actual abnormal data group includes actual data values of a plurality of preset abnormal indicators corresponding to the geological disaster unit;

[0014] For each of the transport sub-areas, respectively calculating the similarity between the actual abnormal data group and the corresponding standard abnormal data group of each geological hazard monomer; the standard abnormal data group includes preset standard data values of a plurality of preset abnormal indicators corresponding to the geological hazard monomer;

[0015] If the similarity is greater than or equal to a preset first similarity threshold, the corresponding transport sub-area is used as the geological disaster area corresponding to the geological disaster monomer.

[0016] Optionally, determining the potential risk area according to the surface image information includes:

[0017] Determine, based on the surface image information, an actual hidden danger data group for each geological disaster unit corresponding to each of the transport sub-areas included in the potential geological disaster area; the actual hidden danger data group includes actual data values of multiple preset hidden danger indicators corresponding to the geological disaster units;

[0018] For each of the transport sub-areas included in the potential geological hazard area, respectively calculating the distance between the actual data value of each of the preset hidden danger indicators under each corresponding geological hazard monomer and the corresponding preset standard value, and comparing the distance with the corresponding preset distance threshold to obtain a corresponding comparison result;

[0019] If the comparison result shows that the distance under a certain geological hazard monomer corresponding to a certain transport sub-area is greater than or equal to the corresponding preset distance threshold and the number of preset hidden danger indicators is greater than or equal to the preset number threshold, then the geological hazard monomer is taken as the target geological hazard monomer, and the transport sub-area is taken as the potential risk area of the target geological hazard monomer.

[0020] Optionally, the predicting the environmental impact value of the potential risk area using a trained neural network model based on the potential geological hazard area, the potential risk area and the actual geological environment information includes:

[0021] Acquire actual geological data values of various geological indicators under the target geological hazard monomer corresponding to the potential risk area from the actual geological environment information;

[0022] According to the following formula, the disaster value of the target geological disaster unit corresponding to the potential risk area is calculated: :

[0023] ;

[0024] Wherein, i represents the number of the geological indicator under the target geological hazard unit and the value range of i is 1 to n, and n represents the total number of geological indicators under the target geological hazard unit; Indicates the preset proportional coefficient corresponding to the i-th geological indicator; represents the actual geological data value corresponding to the i-th geological indicator; represents the standard geological data value corresponding to the i-th geological indicator;

[0025] The potential geological disaster area, the potential risk area and the disaster value of the target geological disaster unit corresponding to the potential risk area are , input the trained neural network model to obtain the environmental impact value of the potential risk area output by the neural network model.

[0026] Optionally, determining the risk assessment value of the potential risk area according to the environmental impact value and the pipeline monitoring data includes:

[0027] Obtaining an actual monitoring value corresponding to each target monitoring indicator from the pipeline monitoring data;

[0028] Count the total time that the actual monitoring value corresponding to each target monitoring indicator falls into the corresponding standard value interval within the preset time window;

[0029] Calculate the time ratio of the total time value corresponding to each target monitoring indicator to the preset time window respectively;

[0030] Calculate the risk assessment value of the potential risk area according to the following formula: :

[0031] ;

[0032] in, 、 are all preset evaluation coefficients; j represents the number of the target monitoring indicator and the value range of j is 1 to m, and m represents the total number of target monitoring indicators; Indicates the preset risk coefficient corresponding to the jth target monitoring indicator; Represents the time ratio of the total time value corresponding to the j-th target monitoring indicator to the preset time window; The environmental impact value representing the potential risk area.

[0033] Optionally, determining and issuing a warning signal corresponding to the potential risk area based on the first-type risk signal and the second-type risk signal includes:

[0034] Determine the total number of risk signals of the type corresponding to the potential risk area in the preset warning window and the total number of the two types of risk signals ;

[0035] According to the following formula, the warning value corresponding to the potential risk area is calculated: :

[0036] ;

[0037] Where, e represents the base of natural logarithm; 、 All are preset proportional coefficients;

[0038] Determine the warning value The corresponding target warning interval, the warning level corresponding to the target warning interval is used as the warning value The corresponding warning level generates and issues the warning signal corresponding to the warning level.

[0039] Optionally, the method further includes:

[0040] Obtaining a preset patrol frequency value PL along the potential risk area within the adjustment window;

[0041] According to the following formula, the patrol frequency value PP in the next adjustment window is calculated, so that the inspection personnel can patrol the pipeline along the potential risk area according to the patrol frequency value PP when the next adjustment window arrives:

[0042] ;

[0043] Wherein, D represents the warning level corresponding to the potential risk area. The larger the value, the more serious the risk of the potential risk area.

[0044] Optionally, the neural network model is trained by the following steps:

[0045] Obtaining the target potential geological hazard area of the oil and gas long-distance pipeline, the target potential risk area within the target potential geological hazard area, and the environmental impact value corresponding to the target potential risk area, and constructing a training data set, a validation data set, and a test data set;

[0046] The target potential geological hazard area, the target potential risk area, and the hazard value of the training target geological hazard monomer corresponding to the target potential risk area are used as inputs of the neural network model, and the environmental impact value of the target potential risk area is used as output of the neural network model. The neural network model is iteratively trained using the training data set and the validation data set until a preset model performance index is met or the training converges;

[0047] The trained neural network model is evaluated using the test data set. If all the model performance indicators of the neural network model meet the corresponding preset standards, the training of the neural network model is determined to be complete.

[0048] Another aspect of the present disclosure provides a risk monitoring and early warning system for long-distance oil and gas pipelines, the system comprising an acquisition module, a matching module, a neural network module, a pipeline monitoring module, an evaluation and analysis module, and an early warning module;

[0049] The acquisition module is used to collect satellite image information of each transportation sub-area of the long-distance oil and gas pipeline; collect surface image information of the potential geological disaster area fed back by the matching module; and collect actual geological environment information of the potential risk area fed back by the matching module;

[0050] The matching module is used to determine potential geological disaster areas based on the satellite image information and feed the potential geological disaster areas back to the acquisition module; and to determine potential risk areas based on the surface image information and feed the potential risk areas back to the acquisition module;

[0051] The neural network module is used to predict the environmental impact value of the potential risk area using a trained neural network model based on the satellite image information, the surface image information, and the actual geological environment information;

[0052] The pipeline monitoring module is used to obtain pipeline monitoring data of the potential risk area;

[0053] The assessment and analysis module is configured to determine a risk assessment value for the potential risk area based on the environmental impact value and the pipeline monitoring data; determine whether the risk assessment value is greater than a preset risk threshold; if so, generate a Class I risk signal for the potential risk area; otherwise, generate a Class II risk signal for the potential risk area; wherein the Class I risk signal is used to indicate that a risk is clearly present in the potential risk area, and the Class II risk signal is used to indicate that a potential risk hazard may exist in the potential risk area;

[0054] The early warning module is used to determine and issue an early warning signal corresponding to the potential risk area based on the first type of risk signal and the second type of risk signal.

[0055] Optionally, the system further comprises:

[0056] The adjustment module is used to obtain the preset patrol frequency value PL of the potential risk area in the adjustment window; and calculate the patrol frequency value PP of the next adjustment window according to the following formula, so that the inspection personnel can conduct pipeline patrols in the potential risk area according to the patrol frequency value PP when the next adjustment window arrives:

[0057] ;

[0058] Wherein, D represents the warning level corresponding to the potential risk area. The larger the value, the more serious the risk of the potential risk area.

[0059] Compared with the existing technology, the present invention comprehensively collects geological information of the oil and gas long-distance pipeline area, realizes multi-dimensional data acquisition in a coordinated manner between space, air and ground, and provides more comprehensive and accurate monitoring data. Through the comprehensive application and intelligent analysis of multi-dimensional data, it can quickly identify and locate potential geological disaster areas, improve the accuracy and timeliness of risk identification, realize efficient monitoring and early warning of geological risks of long-distance oil and gas pipelines, and greatly improve the safety management level and operational reliability of the pipeline. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] One or more embodiments are exemplarily illustrated by pictures in the corresponding drawings, and these exemplifications do not constitute limitations on the embodiments. Elements with the same reference numerals in the drawings represent similar elements, and unless otherwise stated, the figures in the drawings do not constitute proportional limitations.

[0061] Figure 1 This is a flow chart of a method for risk monitoring and early warning of long-distance oil and gas pipelines provided in one embodiment of the present disclosure;

[0062] Figure 2 This is a structural diagram of a long-distance oil and gas pipeline risk monitoring and early warning system provided by another embodiment of the present disclosure. DETAILED DESCRIPTION

[0063] In order to make the purpose, technical solutions and advantages of the embodiments of the present disclosure clearer, the embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings. However, it will be understood by those skilled in the art that in each embodiment of the present disclosure, many technical details are provided to enable readers to better understand the present disclosure. However, even without these technical details and various changes and modifications based on the following embodiments, the technical solutions claimed in the present disclosure can be implemented. The division of the following embodiments is for the convenience of description and should not constitute any limitation on the specific implementation of the present disclosure. The various embodiments can be combined and referenced with each other under the premise that there is no contradiction.

[0064] One embodiment of the present disclosure relates to a method for monitoring and early warning of risks in long-distance oil and gas pipelines, the process of which is as follows: Figure 1 As shown, it includes steps S110 to S150.

[0065] Step S110: collect satellite image information of each transportation sub-area of the long-distance oil and gas pipeline, and determine the potential geological disaster area based on the satellite image information; collect surface image information of the potential geological disaster area, and determine the potential risk area based on the surface image information; collect actual geological environment information of the potential risk area.

[0066] Specifically, long-distance oil and gas pipelines typically cover vast geographical areas. To ensure high-precision early warning of pipeline risks, they are typically divided into multiple sub-areas, allowing for separate management and maintenance of the pipeline within each sub-area.

[0067] For example, step S110 may first collect satellite image information of each transmission sub-area of a long-distance oil and gas pipeline via satellite, and use the satellite image information as the "sky" collection information set. Then, based on the "sky" collection information set, a potential geological disaster area is determined as a first target area. Surface image information of the first target area is obtained by aerial surveying using a drone, and the surface image information is used as the "air" collection information set. Then, based on the "air" collection information set, a potential risk area is determined as a second target area. Actual geological environment information of the second target area is verified through ground collection, and the actual geological environment information is used as the "ground" collection information set. By separately collecting satellite image information, surface image information, and actual geological environment information, step S110 can ensure the comprehensiveness and accuracy of the data.

[0068] In particular, step S110 can obtain large-scale image information of each transmission sub-region of the long-distance oil and gas pipeline using satellite imagery. This image information covers the pipeline and its surrounding environment in each transmission sub-region, forming a preliminary data set as a "daily" collection information set for monitoring surface changes in each transmission sub-region and large-scale geological disaster risks. For example, step S110 can use satellite imagery to display geological changes and their time-series deformation information, such as landslides, ground fissures, or river diversions, around the pipeline in each transmission sub-region, and detect surface deformation of the mountain near the pipeline.

[0069] Compared to satellites, drones can provide images with higher resolution and flexibility. Therefore, step S110 can use the drone to conduct detailed aerial surveys of the potential geological disaster area, i.e., the first target area, determined based on satellite image information. High-resolution, high-precision images and videos of the first target area can be obtained as surface image information of the first target area, forming an "empty" collection of information to provide more detailed information about the first target area and assist in the precise identification of geological anomalies within a small area. For example, if the first target area is a landslide area, step S110 can use the surface image information collected by the drone to display a clearer image of the landslide area. Based on this, the specific location, surface features, and relative relationship of the landslide to the pipeline can be confirmed, and the direct impact of the landslide on the pipeline can be assessed.

[0070] Ground acquisition can be performed directly in the potential risk area by obtaining geological samples, monitoring surface movement, soil moisture, climate conditions, and other actual on-site conditions. Therefore, step S110 can obtain more detailed geological details and environmental data of the second target area as actual geological environmental information by performing ground acquisition in the potential risk area, i.e., the second target area. The actual geological environmental information is then used to verify and supplement the satellite image information and surface image information corresponding to the second target area. For example, when the second target area is determined to be a landslide area based on the surface image information collected by the drone, step S110 can conduct an on-site inspection of the second target area through ground acquisition, recording the actual geological environmental information such as the geological characteristics, inducing mechanism, movement direction, and evolution history of the landslide in the second target area, so as to confirm the development trend of the landslide and its specific threat level to the pipeline.

[0071] Taking abnormal surface changes as an example, assuming that the route of a long-distance oil and gas pipeline is divided into multiple transmission sub-regions, step S110 can obtain a "sky" collection information set through satellite acquisition. If the "sky" collection information set reveals that a region within these transmission sub-regions has experienced abnormal surface changes, such as vegetation loss or significant temporal deformation changes, then this region is identified as a potential geological disaster area, namely the first target area. Subsequently, step S110 can plan a drone route and use the drone to conduct high-resolution aerial surveys of the first target area, obtaining more detailed surface images and videos of the first target area as the surface image information of the first target area, forming an "air" collection information set. Based on the "air" collection information set, the area within the first target area with the most abnormal surface changes is identified as a potential risk area, namely the second target area. Subsequently, step S110 can conduct a field survey of the second target area through ground acquisition, collecting more detailed geological samples and environmental data from the second target area as actual geological environmental information, forming a "ground" collection information set.

[0072] It should be noted that step S110 can also perform data processing operations on the "sky" collection information set, the "air" collection information set, and the "ground" collection information set respectively to obtain the corresponding "sky" processed data set, the "air" processed data set, and the "ground" processed data set, so as to extract the required image features and data features of different data types, and screen out the required images and related geological information for subsequent steps.

[0073] Exemplarily, in step S110, the potential geological disaster area is determined based on the satellite image information, including: determining the actual abnormal data group of each geological disaster unit corresponding to each transport sub-area based on the satellite image information; the actual abnormal data group includes the actual data values of multiple preset abnormal indicators corresponding to the geological disaster unit; for each transport sub-area, calculating the similarity between the actual abnormal data group of each geological disaster unit corresponding to it and the corresponding standard abnormal data group; the standard abnormal data group includes the preset standard data values of multiple preset abnormal indicators corresponding to the geological disaster unit; if the similarity is greater than or equal to the preset first similarity threshold, the corresponding transport sub-area is used as the geological disaster area of the corresponding geological disaster unit.

[0074] Specifically, the operation of determining potential geological disaster areas based on satellite image information can be called a first-level matching operation. Through the first-level matching operation, potential geological disaster areas can be preliminarily identified based on satellite image information, i.e., the "day" collection information set or the "day" processed data set obtained after data processing of the "day" collection information set, and each transportation sub-area of the oil and gas long-distance pipeline can be divided into potential geological disaster areas and non-potential geological disaster areas according to the multiple preset abnormal indicators corresponding to each geological disaster monomer, which serve as the first target area and the non-first target area respectively. Among them, the various geological disaster monomers may include landslides, collapses, mudslides, ground subsidence, ground fissures, etc. The multiple preset abnormal indicators corresponding to the geological disaster monomer refer to a series of pre-set risk indicators for judging whether the geological disaster monomer is in an active state or may occur, such as surface deformation, vegetation changes, etc.

[0075] When performing a first-level matching operation, the actual abnormal data group corresponding to each geological hazard unit in each transport sub-region can be recorded as vector A, and the standard abnormal data group corresponding to the actual abnormal data group can be recorded as vector B. The similarity between vector A and vector B, such as cosine similarity, is calculated. If the similarity between vector A and vector B in a certain transport sub-region is greater than or equal to a preset first similarity threshold, the transport sub-region is determined to be a potential geological hazard area, i.e., the first target area. If the similarity between vector A and vector B in a certain transport sub-region is less than the first similarity threshold, the transport sub-region is determined to be a non-potential geological hazard area, i.e., a non-first target area. The elements in vector A and the elements in vector B correspond one-to-one to each preset abnormality indicator corresponding to the geological hazard unit.

[0076] This embodiment uses a primary matching operation to determine potential geological disaster areas, which can quickly identify and locate potential geological disaster areas involved in long-distance oil and gas pipelines, thereby improving the accuracy and timeliness of risk identification.

[0077] Exemplarily, in step S110, the potential risk area is determined based on the surface image information, including: determining the actual hidden danger data group of each geological hazard monomer corresponding to each transport sub-area included in the potential geological hazard area based on the surface image information; the actual hidden danger data group includes the actual data values of multiple preset hidden danger indicators corresponding to the geological hazard monomer; for each transport sub-area included in the potential geological hazard area, respectively calculating the distance between the actual data value of each preset hidden danger indicator under each corresponding geological hazard monomer and the corresponding preset standard value, and comparing the distance with the corresponding preset distance threshold to obtain a corresponding comparison result; if the comparison result shows that the number of preset hidden danger indicators under a certain geological hazard monomer corresponding to a certain transport sub-area is greater than or equal to the corresponding preset distance threshold and is greater than or equal to the preset number threshold, then the geological hazard monomer is taken as the target geological hazard monomer, and the transport sub-area is taken as the potential risk area of the target geological hazard monomer.

[0078] Specifically, the operation of determining the potential risk area based on the surface image information can be called a secondary matching operation. Through the secondary matching operation, the potential risk area in the potential geological disaster area can be further confirmed and refined based on the surface image information, that is, the "empty" collected information set or the "empty" processed data set obtained by data processing the "empty" collected information set, and the potential geological disaster area can be divided into potential risk areas and non-potential risk areas according to the multiple preset hidden danger indicators corresponding to each geological disaster monomer, as the second target area and the non-second target area respectively. Among them, the multiple preset hidden danger indicators corresponding to the geological disaster monomer can be the same as or different from the preset abnormal indicators. For example, the multiple preset hidden danger indicators corresponding to the geological disaster monomer landslide may include surface fragmentation, trailing edge cracks, vegetation changes, etc.

[0079] For example, assuming that the multiple preset hidden danger indicators corresponding to the geological hazard monomer D1 are DA11 and DA12, when performing the secondary matching operation, for each transport sub-area in the potential geological hazard area of the geological hazard monomer D1, that is, the first target area, the distance between the actual data value of the preset hidden danger indicator DA11 under the corresponding geological hazard monomer D1 and the corresponding standard data value is calculated respectively. , the distance between the actual data value of the preset hidden danger indicator DA12 and the corresponding standard data value , if a transport sub-area corresponds to and are all greater than the corresponding preset distance threshold, it indicates that the number of preset hidden danger indicators in the transport sub-area under the geological hazard monomer D1 that is greater than or equal to the corresponding preset distance threshold is 2. When the preset number threshold is 1 or 2, the transport sub-area satisfies the requirement that the number of preset hidden danger indicators in the transport sub-area that is greater than or equal to the corresponding preset distance threshold under the geological hazard monomer D1 is greater than or equal to the preset number threshold. At this time, the transport sub-area is regarded as the potential risk area of the target geological hazard monomer D1, that is, the second target area.

[0080] This embodiment determines the potential risk area based on the potential geological disaster area by using a secondary matching operation, thereby further narrowing the scope of the geological disaster risk area.

[0081] Through the two-level matching operations mentioned above, namely the first-level matching operation and the second-level matching operation, the scope of the geological disaster risk area is gradually narrowed. Based on large-scale satellite image information, detailed surface image information and actual geological environment information, potential geological disaster areas can be quickly identified and located, ensuring the accuracy and reliability of subsequent target monitoring and identification results, providing basic support for subsequent information collection and data processing, and facilitating comprehensive risk assessment of pipeline areas.

[0082] Step S120 , based on the potential geological disaster area, the potential risk area and the actual geological environment information, the trained neural network model is used to predict the environmental impact value of the potential risk area.

[0083] Specifically, a trained neural network model is used to assist in the prediction of the environmental impact value of a potential risk area. The neural network model can employ a convolutional neural network to improve the accuracy of the model's predictions. The environmental impact value of a potential risk area represents the degree of environmental impact in that area and can reflect the environmental quality or potential risk of that area.

[0084] Exemplarily, step S120 includes: obtaining actual geological data values of various geological indicators under the target geological disaster monomer corresponding to the potential risk area from the actual geological environment information. According to the following formula, the disaster value of the target geological disaster monomer corresponding to the potential risk area is calculated: :

[0085] .

[0086] Among them, i represents the number of the geological indicator under the target geological hazard unit and the value range of i is 1 to n, and n represents the total number of geological indicators under the target geological hazard unit. It represents the preset proportional coefficient corresponding to the i-th geological indicator, which is used to adjust the ratio between the actual data and the standard data to reflect the impact of the actual data on the degree of geological hazards. It represents the actual geological data value corresponding to the i-th geological indicator, which is used to describe the characteristics and status of the geological hazard entity. It represents the standard geological data value corresponding to the i-th geological indicator, which can be determined according to the characteristics of the geological disaster and historical data, and is used as a reference value to evaluate the status of the geological disaster unit. Usually, the middle value within the normal numerical range corresponding to the geological indicator is taken. The geological indicators of the target geological disaster unit are used to indicate the geological environment characteristics of the target geological disaster unit. For example, the geological indicators of the target geological disaster unit landslide may include topographic and geomorphological indicators such as slope, elevation difference, slope direction, terrain curvature, and rock and soil property indicators such as lithology, rock structure, porosity and permeability, shear strength, or other types of indicators. Disaster value The larger the value, the more serious the disaster.

[0087] The hazard value of the target geological hazard unit corresponding to the potential geological hazard area, potential risk area and potential risk area is calculated. , input the trained neural network model and obtain the environmental impact value of the potential risk area output by the neural network model.

[0088] Specifically, this embodiment can use the results of the first-level matching operation and the second-level matching operation, as well as the disaster value of the target geological disaster unit corresponding to the potential risk area, i.e., the second target area, as input data of the trained neural network model. It can also use the time-series surface deformation value of the second target area obtained based on the synthetic aperture radar (SAR) data and the collected influencing factors and disaster-causing factors as input data of the trained neural network model to assist the model in making predictions.

[0089] This embodiment can predict and analyze environmental impact values through a trained neural network model, provide intelligent risk assessment and prediction capabilities, and help decision makers take preventive measures in advance.

[0090] Exemplarily, the neural network model is trained by the following steps: obtaining the target potential geological hazard area of the long-distance oil and gas pipeline, the target potential risk area in the target potential geological hazard area, and the environmental impact value corresponding to the target potential risk area, and constructing a training data set, a verification data set, and a test data set. The target potential geological hazard area, the target potential risk area, and the hazard value of the training target geological hazard monomer corresponding to the target potential risk area are used as the input of the neural network model, and the environmental impact value of the target potential risk area is used as the output of the neural network model. The neural network model is iteratively trained using the training data set and the verification data set until the preset model performance indicators are met or the training converges. The trained neural network model is evaluated using the test data set. If all model performance indicators of the neural network model meet the corresponding preset standards, the training of the neural network model is determined to be complete.

[0091] Specifically, the disaster values of the target potential geological hazard area, the target potential risk area, and the training target geological hazard monomer corresponding to the target potential risk area can be obtained by referring to the methods for obtaining the disaster values of the potential geological hazard area, the potential risk area, and the target geological hazard monomer corresponding to the potential risk area provided in the above-mentioned embodiments. During the model training process, the environmental impact value of the target potential risk area can be used as the true label, and the parameters in the neural network model can be adjusted through the back propagation algorithm to make the prediction results of the neural network model closer to the corresponding true label. When the various model performance indicators of the neural network model reach the corresponding preset standards, the neural network model training is determined to be complete. Among them, the various model performance indicators can include accuracy, precision, recall rate, etc.

[0092] This embodiment uses the disaster value of the target potential geological hazard area, the target potential risk area, and the training target geological hazard unit corresponding to the target potential risk area as the input data of the neural network model during the model training process, and uses the environmental impact value of the target potential risk area as the output data of the neural network model, hoping that the neural network model can learn the complex relationship between the input data and the output data, thereby achieving accurate prediction of the environmental impact value. Among them, the environmental impact value here can be a continuous value, representing the degree of environmental impact on the corresponding potential risk area. Since the neural network model can learn the corresponding relationship between input data and output data through training, the trained neural network model can accurately predict the corresponding output data based on new, unseen input data, thereby helping subsequent steps to better understand and evaluate the environmental conditions of the pipeline area.

[0093] Step S130: Acquire pipeline monitoring data of potential risk areas.

[0094] Specifically, step S130 can utilize various sensors installed on the long-distance oil and gas pipeline to conduct real-time monitoring of the pipeline in the potential risk area, thereby obtaining pipeline monitoring data of the potential risk area, realizing real-time monitoring of the long-distance oil and gas pipeline area and sudden geological disasters, and comprehensively covering the complex terrain and geological environment along the entire pipeline.

[0095] Step S140 determines the risk assessment value of the potential risk area based on the environmental impact value and pipeline monitoring data. A determination is made as to whether the risk assessment value is greater than a preset risk threshold. If so, a Class I risk signal is generated for the potential risk area; otherwise, a Class II risk signal is generated for the potential risk area. A Class I risk signal indicates a clear risk in the potential risk area, while a Class II risk signal indicates a potential risk hazard.

[0096] Exemplarily, in step S140, the risk assessment value of the potential risk area is determined based on the environmental impact value and the pipeline monitoring data, including: obtaining the actual monitoring value corresponding to each target monitoring indicator from the pipeline monitoring data; counting the total time value of the actual monitoring value corresponding to each target monitoring indicator falling into the corresponding standard value interval within the preset time window; calculating the ratio of the total time value corresponding to each target monitoring indicator to the time of the preset time window; and calculating the risk assessment value of the potential risk area according to the following formula: :

[0097] .

[0098] in, 、 are all preset evaluation coefficients; j represents the number of the target monitoring indicator and the value range of j is 1 to m, and m represents the total number of target monitoring indicators; Indicates the preset risk coefficient corresponding to the jth target monitoring indicator; Indicates the ratio of the total time value corresponding to the jth target monitoring indicator to the time of the preset time window; Indicates the environmental impact value of the potential risk area.

[0099] Specifically, assuming that the standard numerical range corresponding to the jth target monitoring indicator is ,in, Indicates the minimum standard value corresponding to the target monitoring indicator. Indicates the maximum standard value corresponding to the target monitoring indicator, then step S140 can count the actual monitoring value corresponding to the target monitoring indicator in the preset time window and fall into the standard value interval The total time value of each time point in the time window is calculated and marked as .

[0100] It should be noted that the preset risk threshold can be set according to actual needs, and this embodiment does not limit this.

[0101] This implementation method dynamically generates risk assessment values by comprehensively analyzing environmental impact values and pipeline monitoring data, which can achieve dynamic assessment of long-distance oil and gas pipeline areas and rapid response to sudden geological disasters, fully utilize the comprehensive information of multi-source data, and improve the accuracy of risk identification.

[0102] By generating a Class I or Class II risk signal for potential risk areas, it also provides a scientific basis for geological disaster warning and response for long-distance oil and gas pipelines, allowing relevant managers to implement different management methods for potential risk areas based on different risk signals, which helps to prevent and mitigate disasters in a timely and effective manner.

[0103] Step S150: Determine and issue a warning signal corresponding to the potential risk area based on the first-class risk signal and the second-class risk signal.

[0104] For example, step S150 includes: determining the total number of risk signals corresponding to a type of potential risk area within a preset warning window. and the total number of type II risk signals ; Calculate the warning value corresponding to the potential risk area according to the following formula: : . Where e represents the base of natural logarithms. 、 All are preset proportional coefficients. Determine the warning value The corresponding target warning interval, the warning level corresponding to the target warning interval is used as the warning value Corresponding to the warning level, a warning signal corresponding to the warning level is generated and issued.

[0105] Specifically, the warning interval and its corresponding warning level can be set in advance in the form of a warning interval-warning level form. When the target warning interval is reached, the warning value Compare with the preset warning interval-warning level form, if the warning value If it falls into a certain warning interval, the warning interval is used as the target warning interval, and the warning level corresponding to the target warning interval is used as the warning value. The corresponding warning level.

[0106] For example, suppose the total number of risk signals of a type corresponding to a potential risk area is =3, the total number of corresponding second-class risk signals =5, the warning value corresponding to the potential risk area obtained by calculation The value of 3.4 is 3.4. When 3.4 is compared with the warning intervals in the warning interval-warning level table, it is found that it falls into the warning interval [3, 5). When the warning level corresponding to the warning interval [3, 5) is level three, a warning signal corresponding to the level three warning level is generated, such as a medium-level warning signal.

[0107] This implementation method can remind relevant personnel to strengthen monitoring and management of potential risk areas by determining and issuing early warning signals corresponding to potential risk areas, and guide relevant departments to take corresponding risk prevention and control measures to ensure the safe operation of oil and gas pipelines.

[0108] Exemplarily, the long-distance oil and gas pipeline risk monitoring and early warning method further includes: obtaining a preset patrol frequency value PL for the potential risk area within an adjustment window. Calculating the patrol frequency value PP for the next adjustment window according to the following formula, so that patrol personnel conduct pipeline patrols in the potential risk area according to the patrol frequency value PP when the next adjustment window arrives: Among them, D represents the warning level corresponding to the potential risk area. The larger the value, the more serious the risk of the potential risk area.

[0109] Specifically, since the larger the warning level value is, the higher the warning level is, the more serious the risk of the potential risk area is, the higher the warning level is, the more serious the risk of the potential risk area is. The calculated inspection frequency along the route will also be greater.

[0110] This implementation method optimizes the allocation of inspection resources, improves the efficiency of inspection and management, reduces omissions caused by human factors, and improves the level of pipeline safety management by adjusting the inspection frequency according to the early warning level. It enables relevant departments to identify and discover potential risks in advance, take preventive and response measures in a timely manner, reduce the impact of geological disasters on long-distance oil and gas pipelines, reduce maintenance and emergency response costs, and reduce economic losses.

[0111] It should be noted that the relevant calculations involved in the above-mentioned embodiments of the present disclosure are all performed by removing the dimensions and taking the numerical values of each parameter for calculation, and each relevant preset parameter can be selected and set by those skilled in the art according to actual conditions.

[0112] Compared with the existing technology, the risk monitoring and early warning method for long-distance oil and gas pipelines provided in the embodiments of the present disclosure comprehensively collects geological information of the long-distance oil and gas pipeline area, realizes multi-dimensional data acquisition with sky-air-ground coordination, and provides more comprehensive and accurate monitoring data. Through the comprehensive application and intelligent analysis of multi-dimensional data, it can quickly identify and locate potential geological disaster areas, improve the accuracy and timeliness of risk identification, realize efficient monitoring and early warning of geological risks of long-distance oil and gas pipelines, and greatly improve the safety management level and operational reliability of the pipeline.

[0113] Another embodiment of the present disclosure relates to a risk monitoring and early warning system for long-distance oil and gas pipelines, such as Figure 2 As shown, it includes an acquisition module 210, a matching module 220, a neural network module 230, a pipeline monitoring module 240, an evaluation and analysis module 250, and an early warning module 260.

[0114] The acquisition module 210 is used to collect satellite image information of each transportation sub-area of the long-distance oil and gas pipeline; collect surface image information of potential geological disaster areas fed back by the matching module; and collect actual geological environment information of potential risk areas fed back by the matching module.

[0115] The matching module 220 is used to determine potential geological disaster areas based on satellite image information and feed the potential geological disaster areas back to the acquisition module; and to determine potential risk areas based on surface image information and feed the potential risk areas back to the acquisition module.

[0116] The neural network module 230 is used to predict the environmental impact value of the potential risk area using a trained neural network model based on satellite image information, surface image information, and actual geological environment information.

[0117] The pipeline monitoring module 240 is used to obtain pipeline monitoring data of potential risk areas.

[0118] The assessment and analysis module 250 is used to determine the risk assessment value of the potential risk area based on the environmental impact value and pipeline monitoring data; determine whether the risk assessment value is greater than the preset risk threshold; if so, generate a Class I risk signal for the potential risk area; otherwise, generate a Class II risk signal for the potential risk area; wherein, the Class I risk signal is used to indicate that there is a clear risk in the potential risk area, and the Class II risk signal is used to indicate that there may be hidden risks in the potential risk area.

[0119] The early warning module 260 is used to determine and issue early warning signals corresponding to potential risk areas based on the first-class risk signals and the second-class risk signals.

[0120] Exemplarily, the long-distance oil and gas pipeline risk monitoring and early warning system also includes a regulation module.

[0121] The adjustment module is used to obtain the preset patrol frequency value PL of the potential risk area in the adjustment window; according to the following formula, the patrol frequency value PP of the next adjustment window is calculated, so that the inspection personnel can conduct pipeline patrols in the potential risk area according to the patrol frequency value PP when the next adjustment window arrives:

[0122] .

[0123] Among them, D represents the warning level corresponding to the potential risk area. The larger the value, the more serious the risk of the potential risk area.

[0124] The specific implementation method of the oil and gas long-distance pipeline risk monitoring and early warning system provided by the embodiment of the present disclosure can be found in the description of the oil and gas long-distance pipeline risk monitoring and early warning method provided by the embodiment of the present disclosure, and will not be repeated here.

[0125] Compared with the existing technology, the long-distance oil and gas pipeline risk monitoring and early warning system provided by the embodiment of the present disclosure comprehensively collects geological information of the long-distance oil and gas pipeline area, realizes multi-dimensional data acquisition with sky-air-ground coordination, and provides more comprehensive and accurate monitoring data. Through the comprehensive application and intelligent analysis of multi-dimensional data, it can quickly identify and locate potential geological disaster areas, improve the accuracy and timeliness of risk identification, realize efficient monitoring and early warning of geological risks of long-distance oil and gas pipelines, and greatly improve the safety management level and operational reliability of the pipeline.

[0126] Those skilled in the art will appreciate that the above-mentioned embodiments are specific embodiments for implementing the present disclosure, and that in actual applications, various changes may be made thereto in form and detail without departing from the spirit and scope of the present disclosure.

Claims

1. A risk monitoring and early warning method for long-distance oil and gas pipelines, characterized in that: The method comprises: Collect satellite image information of each transport sub-area of the long-distance oil and gas pipeline, and determine potential geological disaster areas based on the satellite image information; collect surface image information of the potential geological disaster areas, and determine potential risk areas based on the surface image information; collect actual geological environment information of the potential risk areas; Based on the potential geological disaster area, the potential risk area and the actual geological environment information, the trained neural network model is used to predict the environmental impact value of the potential risk area; the environmental impact value is used to indicate the degree of environmental impact on the potential risk area; Obtaining pipeline monitoring data in the potential risk area; Determine a risk assessment value for the potential risk area based on the environmental impact value and the pipeline monitoring data; determine whether the risk assessment value is greater than a preset risk threshold; if so, generate a Class I risk signal for the potential risk area; otherwise, generate a Class II risk signal for the potential risk area; wherein the Class I risk signal is used to indicate that a risk is clearly present in the potential risk area, and the Class II risk signal is used to indicate that a potential risk hazard may exist in the potential risk area; Based on the first-class risk signal and the second-class risk signal, a warning signal corresponding to the potential risk area is determined and issued.

2. The method according to claim 1, characterized in that Determining a potential geological disaster area based on the satellite image information includes: Determine, based on the satellite image information, an actual abnormal data group for each geological disaster unit corresponding to each of the transport sub-areas; the actual abnormal data group includes actual data values of a plurality of preset abnormal indicators corresponding to the geological disaster unit; For each of the transport sub-areas, respectively calculating the similarity between the actual abnormal data group and the corresponding standard abnormal data group of each geological hazard monomer; the standard abnormal data group includes preset standard data values of a plurality of preset abnormal indicators corresponding to the geological hazard monomer; If the similarity is greater than or equal to a preset first similarity threshold, the corresponding transport sub-area is used as the geological disaster area corresponding to the geological disaster monomer.

3. The method according to claim 1, characterized in that Determining the potential risk area according to the surface image information includes: Determine, based on the surface image information, an actual hidden danger data group for each geological disaster unit corresponding to each of the transport sub-areas included in the potential geological disaster area; the actual hidden danger data group includes actual data values of multiple preset hidden danger indicators corresponding to the geological disaster units; For each of the transport sub-areas included in the potential geological hazard area, respectively calculating the distance between the actual data value of each of the preset hidden danger indicators under each corresponding geological hazard monomer and the corresponding preset standard value, and comparing the distance with the corresponding preset distance threshold to obtain a corresponding comparison result; If the comparison result shows that the distance under a certain geological hazard monomer corresponding to a certain transport sub-area is greater than or equal to the corresponding preset distance threshold and the number of preset hidden danger indicators is greater than or equal to the preset number threshold, then the geological hazard monomer is taken as the target geological hazard monomer, and the transport sub-area is taken as the potential risk area of the target geological hazard monomer.

4. The method according to claim 3, characterized in that The method of predicting the environmental impact value of the potential risk area by using a trained neural network model based on the potential geological disaster area, the potential risk area and the actual geological environment information includes: Acquire actual geological data values of various geological indicators under the target geological hazard monomer corresponding to the potential risk area from the actual geological environment information; According to the following formula, the disaster value of the target geological disaster unit corresponding to the potential risk area is calculated: : ； Wherein, i represents the number of the geological indicator under the target geological hazard unit and the value range of i is 1 to n, and n represents the total number of geological indicators under the target geological hazard unit; Indicates the preset proportional coefficient corresponding to the i-th geological indicator; represents the actual geological data value corresponding to the i-th geological indicator; represents the standard geological data value corresponding to the i-th geological indicator; The potential geological disaster area, the potential risk area and the disaster value of the target geological disaster unit corresponding to the potential risk area are , input the trained neural network model to obtain the environmental impact value of the potential risk area output by the neural network model.

5. The method according to claim 1, wherein Determining the risk assessment value of the potential risk area based on the environmental impact value and the pipeline monitoring data includes: Obtaining an actual monitoring value corresponding to each target monitoring indicator from the pipeline monitoring data; Count the total time that the actual monitoring value corresponding to each target monitoring indicator falls into the corresponding standard value interval within the preset time window; Calculate the time ratio of the total time value corresponding to each target monitoring indicator to the preset time window respectively; Calculate the risk assessment value of the potential risk area according to the following formula: : ； in, 、 are all preset evaluation coefficients; j represents the number of the target monitoring indicator and the value range of j is 1 to m, and m represents the total number of target monitoring indicators; Indicates the preset risk coefficient corresponding to the jth target monitoring indicator; Represents the time ratio of the total time value corresponding to the j-th target monitoring indicator to the preset time window; The environmental impact value representing the potential risk area.

6. The method according to claim 5, characterized in that The determining and issuing of a warning signal corresponding to the potential risk area based on the first-class risk signal and the second-class risk signal includes: Determine the total number of risk signals of the type corresponding to the potential risk area in the preset warning window and the total number of the two types of risk signals ; According to the following formula, the warning value corresponding to the potential risk area is calculated: : ； Where, e represents the base of natural logarithm; 、 All are preset proportional coefficients; Determine the warning value The corresponding target warning interval, the warning level corresponding to the target warning interval is used as the warning value The corresponding warning level generates and issues the warning signal corresponding to the warning level.

7. The method according to claim 6, characterized in that The method further comprises: Obtaining a preset patrol frequency value PL along the potential risk area within the adjustment window; According to the following formula, the patrol frequency value PP in the next adjustment window is calculated, so that the inspection personnel can patrol the pipeline along the potential risk area according to the patrol frequency value PP when the next adjustment window arrives: ； Wherein, D represents the warning level corresponding to the potential risk area. The larger the value, the more serious the risk of the potential risk area.

8. The method according to any one of claims 1 to 7, characterized in that The neural network model is trained by the following steps: Obtaining the target potential geological hazard area of the oil and gas long-distance pipeline, the target potential risk area within the target potential geological hazard area, and the environmental impact value corresponding to the target potential risk area, and constructing a training data set, a validation data set, and a test data set; The target potential geological hazard area, the target potential risk area, and the hazard value of the training target geological hazard monomer corresponding to the target potential risk area are used as inputs of the neural network model, and the environmental impact value of the target potential risk area is used as output of the neural network model. The neural network model is iteratively trained using the training data set and the validation data set until a preset model performance index is met or the training converges; The trained neural network model is evaluated using the test data set. If all the model performance indicators of the neural network model meet the corresponding preset standards, the training of the neural network model is determined to be complete.

9. A risk monitoring and early warning system for long-distance oil and gas pipelines, characterized in that: The system includes an acquisition module, a matching module, a neural network module, a pipeline monitoring module, an evaluation and analysis module, and an early warning module; The acquisition module is used to collect satellite image information of each transportation sub-area of the oil and gas long-distance pipeline; and collect surface image information of the potential geological disaster area fed back by the matching module; and collecting actual geological environment information of the potential risk area fed back by the matching module; The matching module is used to determine potential geological disaster areas based on the satellite image information and feed the potential geological disaster areas back to the acquisition module; and, determining a potential risk area based on the surface image information, and feeding the potential risk area back to the acquisition module; The neural network module is used to predict the environmental impact value of the potential risk area based on the satellite image information, the surface image information, and the actual geological environment information using a trained neural network model; the environmental impact value is used to indicate the degree of environmental impact on the potential risk area; The pipeline monitoring module is used to obtain pipeline monitoring data of the potential risk area; The assessment and analysis module is used to determine the risk assessment value of the potential risk area based on the environmental impact value and the pipeline monitoring data; Determine whether the risk assessment value is greater than a preset risk threshold; if so, generate a Class I risk signal for the potential risk area; otherwise, generate a Class II risk signal for the potential risk area; wherein the Class I risk signal is used to indicate that the potential risk area clearly has a risk, and the Class II risk signal is used to indicate that the potential risk area may have a risk hazard; The early warning module is used to determine and issue an early warning signal corresponding to the potential risk area based on the first type of risk signal and the second type of risk signal.

10. The system according to claim 9, characterized in that The system further comprises: The adjustment module is used to obtain the preset patrol frequency value PL of the potential risk area in the adjustment window; and calculate the patrol frequency value PP of the next adjustment window according to the following formula, so that the inspection personnel can conduct pipeline patrols in the potential risk area according to the patrol frequency value PP when the next adjustment window arrives: ； Wherein, D represents the warning signal corresponding to the potential risk area.

Images