Early warning event processing system of oil and gas pipeline optical fiber early warning system

Abstract

The invention relates to an early warning event processing system of an oil and gas pipeline optical fiber early warning system, and the system comprises a preprocessing module which is used for receiving and preprocessing the real-time early warning data of a plurality of early warning events, and obtaining the preprocessed real-time early warning data; the initial merging module is used for determining a first similar early warning event in the plurality of early warning events according to the preprocessed real-time early warning data, and performing initial merging on the real-time early warning data of the first similar early warning event to obtain a merged early warning event; the dynamic upgrading and downgrading processing module is used for acquiring a multi-dimensional feature vector of the combined early warning event; using a deep learning model to adjust the initial early warning level of the combined early warning event according to the multi-dimensional feature vector to obtain an adjusted early warning level; and the output module is used for outputting the combined early warning event and the adjusted early warning level thereof. According to the system, repeated and redundant information can be reduced, and the early warning event processing efficiency, response speed and dynamic flexibility are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of oil and gas pipeline safety monitoring and informatization, and in particular to a warning event processing system of an oil and gas pipeline optical fiber warning system. Background Art

[0002] In the oil and gas pipeline industry, ensuring the safe operation of the pipeline is of vital importance. The fiber optic early warning system has become an important technical means for oil and gas pipeline safety monitoring due to its advantages such as high sensitivity, long-distance monitoring capability and anti-electromagnetic interference. However, with the increasing complexity of actual application scenarios, higher requirements are placed on the data processing capabilities and intelligence level of the early warning system. There are currently a variety of fiber optic early warning systems on the market. Most of them focus on signal recognition and extraction, classification of early warning events, etc., and pay less attention to the secondary processing of early warning information. In the face of various complex situations, they only rely on template configuration and cannot be flexibly and dynamically adjusted, resulting in a large number of repeated alarms and invalid alarms.

[0003] Disadvantages of existing technology:

[0004] Insufficient dynamics in event handling: The existing system lacks the ability to dynamically adjust event priorities based on event severity and real-time environmental factors, resulting in critical events not being handled in a timely manner.

[0005] Inefficient event merging: When processing a large number of warning events, existing systems may fail to effectively merge events with similar characteristics or time associations, resulting in a large amount of duplicate and redundant information.

[0006] Limited accuracy of deduplication algorithms: The deduplication algorithms of existing systems may not be accurate enough, resulting in duplicate items in the warning event data, affecting the accuracy and consistency of the data.

[0007] Lack of automated response: Some systems may rely on manual intervention to handle warning events and lack the ability to automatically execute pre-set response measures, which may lead to delayed responses. Summary of the invention

[0008] 1. Technical issues to be resolved

[0009] In view of the above-mentioned shortcomings and deficiencies of the prior art, the present invention provides a warning event processing system for an oil and gas pipeline optical fiber warning system, which solves the technical problem that the prior art only relies on template configuration in the face of various complex situations and cannot be flexibly and dynamically adjusted, resulting in a large number of repeated alarms and invalid alarms.

[0010] (II) Technical solution

[0011] In order to achieve the above object, the main technical solutions adopted by the present invention include:

[0012] An embodiment of the present invention provides a warning event processing system for an oil and gas pipeline optical fiber warning system, comprising:

[0013] A preprocessing module, used for receiving and preprocessing the real-time warning data of multiple warning events collected by the optical fiber warning system of the oil and gas pipeline, to obtain the preprocessed real-time warning data;

[0014] A primary merging module, used to calculate the similarity of multiple warning events according to the preprocessed real-time warning data, determine a first similar warning event among the multiple warning events, and perform a primary merging of the real-time warning data of the first similar warning event to obtain one or more merged warning events;

[0015] A dynamic upgrade and downgrade processing module is used to obtain a multi-dimensional feature vector of the merged warning event based on the real-time warning data of the merged warning event; and use a deep learning model to adjust the initial warning level of the merged warning event according to the multi-dimensional feature vector to obtain an adjusted warning level;

[0016] The output module is used to output the merged warning events and their adjusted warning levels.

[0017] Optionally, the real-time warning data includes: the event type, initial warning level, occurrence time, coordinates and optical fiber number of the warning event.

[0018] Optionally, the initial merging module includes:

[0019] A first vector construction unit is used to construct a first feature vector for each warning event based on the associated pre-processed real-time warning data; the initial feature vector includes: the relative time of the warning event, the event type, and the coordinates of the warning event;

[0020] a first cluster analysis unit, configured to calculate the similarity between any two warning events based on the first feature vector, and to divide the plurality of warning events into a plurality of clusters according to the similarity of the warning events;

[0021] The first merging unit determines the warning events under the same cluster as the first similar warning events, and performs an initial merging on the real-time warning data of the similar warning events to obtain a merged warning event.

[0022] Optionally, in the first vector construction unit, the construction of the first feature vector based on the associated pre-processed real-time warning data includes: for the i-th warning event, the constructed first feature vector is expressed as: V i =[t i ,e i ,x i ,y i ]; where ti represents the relative time of the ith warning event, e i represents the event type of the ith warning event, represents the longitude of the ith warning event, and represents the latitude of the ith warning event;

[0023] In the first cluster analysis unit, the calculating the similarity between any two warning events based on the first feature vector includes: for the i-th warning event and the j-th warning event, calculating the similarity between the two based on formula (1);

[0024]

[0025] Among them, d ij Indicates the similarity between the i-th warning event and the j-th warning event.

[0026] Optionally, in the first merging unit, the initial merging of the real-time warning data of the first similar warning event includes:

[0027] For all first similar warning events in a cluster, the data under each item of the associated real-time warning data is averaged or median, and the data is used as the real-time warning data associated with the merged warning event.

[0028] Optionally, the multidimensional feature vector includes: time feature data, location feature data, event type feature data, environment feature data and historical feature data; wherein,

[0029] The time characteristic data includes: whether the occurrence time of the warning event is within a preset sensitive time period;

[0030] The location feature data includes: terrain data, geological data or hydrological data obtained based on the coordinates of the warning event based on GIS;

[0031] The event type data features include: the frequency and / or duration of the event type to which the warning event belongs;

[0032] The environmental characteristic data includes: weather data, traffic conditions or equipment status near the coordinates of the warning event; the equipment status includes: pipeline flow and pipeline status of the oil and gas pipeline;

[0033] The historical characteristic data includes: the occurrence pattern or trend of the warning event at the warning event coordinates.

[0034] Optionally, in the dynamic upgrading and downgrading processing module, the using of a deep learning model includes:

[0035] The deep learning model is: an AELDR model with adapted model parameters obtained based on a pre-training process;

[0036] The AELDR model includes:

[0037] A first LSTM model is used to predict a first result about the trend of the warning event based on the multi-dimensional feature vector;

[0038] A second LSTM model is used to predict a second result about the environmental pattern at the coordinates of the warning event based on the real-time multi-dimensional feature vector;

[0039] A CNN model is used to predict a third result about the spatial pattern and local features of the warning event based on the multi-dimensional feature vector;

[0040] A random forest model, for predicting a fourth result about a nonlinear relationship and a complex pattern of the warning event based on the multidimensional feature vector;

[0041] An SVM model is used to predict a fifth result about a decision boundary of the warning event based on the high-dimensional features of the multi-dimensional feature vector;

[0042] A fusion model is used to fuse the first result, the second result, the third result, the fourth result and the fifth result to predict the adjusted warning level and recommended decision of the warning event.

[0043] Optionally, the fusion model is a stacked meta-model.

[0044] Optionally, the pre-training process includes: acquiring historical data of a plurality of historical warning events, extracting historical multi-dimensional feature vectors of the historical data, and warning level adjustment records associated with the historical warning events as data sets for training;

[0045] The historical data includes: historical time feature data, historical location feature data, historical event type feature data, historical environment feature data, historical characteristic data and historical decision feature data.

[0046] Optionally, the dynamic upgrade and downgrade processing module and the output module are connected via a secondary merging module;

[0047] The secondary merging module is used to update the real-time warning data of the associated merged warning events according to the adjusted warning level output by the dynamic upgrading and downgrading processing module; and recalculate the similarity of multiple merged warning events based on the updated real-time warning data to determine whether there is a second similar warning event among the multiple merged warning events. If so, the real-time warning data associated with the second similar warning event is merged again to obtain one or more re-merged warning events.

[0048] (III) Beneficial effects

[0049] The warning event processing system proposed in the embodiment of the present invention includes a preprocessing module, a primary merging module, a dynamic upgrade and downgrade processing module and an output module. The preprocessing module is used to receive the real-time warning data of multiple warning events collected by the oil and gas pipeline optical fiber warning system and perform preprocessing to obtain preprocessed real-time warning data; the primary merging module is used to calculate the similarity of multiple warning events based on the preprocessed real-time warning data, determine the first similar warning event among the multiple warning events, and perform the primary merging of the real-time warning data of the similar warning events to obtain one or more merged warning events; the dynamic upgrade and downgrade processing module is used to obtain the multi-dimensional feature vector of the merged warning event based on the real-time warning data of the merged warning event; and use the deep learning model to adjust the initial warning level of the merged warning event according to the multi-dimensional feature vector to obtain the adjusted warning level; the output module is used to output the merged warning event and its adjusted warning level.

[0050] The warning event processing system provided in this embodiment is based on the above-mentioned preprocessing module, initial merging module, dynamic upgrading and downgrading processing module and output module. The initial merging module can merge warning events with high similarity or correlation to reduce repeated and redundant information, ensure the accuracy and consistency of event data, and improve the efficiency of warning event processing and the response speed of the system; the dynamic upgrading and downgrading module adjusts the priority of the event according to the multi-dimensional feature vector reflecting the severity of the warning event, historical data and real-time environmental factors, generates merged warning events and their adjusted warning levels, and ensures that key events can receive timely and appropriate attention to better protect the safety of oil and gas pipelines and reduce potential safety risks and losses. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 A schematic diagram of the architecture of a warning event processing system for an oil and gas pipeline optical fiber warning system provided in an embodiment;

[0052] Figure 2 A schematic diagram of the architecture of another warning event processing system of an oil and gas pipeline optical fiber warning system provided in an embodiment. DETAILED DESCRIPTION

[0053] In order to better explain the present invention and facilitate understanding, the present invention is described in detail below through specific implementation modes in conjunction with the accompanying drawings.

[0054] In order to better understand the following specific implementations, some of the terms or terms used are explained:

[0055] Fiber optic early warning: A system that uses fiber optic sensors to monitor changes in the environment around oil and gas pipelines. When an abnormal event is detected, the system will issue an early warning signal

[0056] Warning event processing: a series of processing steps to receive, analyze, optimize and respond to warning events detected by the fiber optic warning system.

[0057] Dynamic upgrade and downgrade processing: Dynamically adjust the priority of events based on the severity of warning events, historical data and real-time environmental factors to ensure that critical events are handled in a timely manner.

[0058] Event merging: Merge warning events with similar characteristics or time associations into one event to reduce duplication and redundant information and improve processing efficiency.

[0059] Priority adjustment: Adjust the priority of event handling based on the urgency and importance of the event.

[0060] Reduction of redundant information: Through merging and deduplication processing, unnecessary information in warning event data is reduced, and the efficiency and accuracy of data processing are improved.

[0061] Embodiment 1

[0062] like Figure 1 As shown, this embodiment provides a warning event processing system for an oil and gas pipeline optical fiber warning system, including:

[0063] The preprocessing module is used to receive and preprocess the real-time warning data of multiple warning events collected by the optical fiber warning system of the oil and gas pipeline to obtain the preprocessed real-time warning data.

[0064] Specifically, the real-time warning data includes: the event type, initial warning level, occurrence time, coordinates and optical fiber number of the warning event.

[0065] The initial merging module is used to calculate the similarity of multiple warning events based on the preprocessed real-time warning data, determine the first similar warning event among the multiple warning events, and perform initial merging of the real-time warning data of the first similar warning event to obtain one or more merged warning events.

[0066] The initial merge module can eliminate duplicate warning events and ensure the accuracy and consistency of real-time warning data.

[0067] The dynamic upgrade and downgrade processing module is used to obtain the multi-dimensional feature vector of the merged warning event based on the real-time warning data of the merged warning event; and use the deep learning model to adjust the initial warning level of the merged warning event according to the multi-dimensional feature vector to obtain the adjusted warning level.

[0068] The dynamic upgrade and downgrade processing module can dynamically adjust the priority of warning events, thereby ensuring the efficiency and accuracy of processing warning events.

[0069] The output module is used to output the merged warning events and their adjusted warning levels.

[0070] The warning event processing system provided by this embodiment is based on the above-mentioned preprocessing module, initial merging module, dynamic upgrade and downgrade processing module and output module. The initial merging module can merge warning events with high similarity or correlation to reduce repeated and redundant information, ensure the accuracy and consistency of event data, and improve the efficiency of warning event processing and the response speed of the system; the dynamic upgrade and downgrade module adjusts the priority of the event according to the multi-dimensional feature vector reflecting the severity of the warning event, historical data and real-time environmental factors, generates merged warning events and their adjusted warning levels, and ensures that key events can receive timely and appropriate attention to better protect the safety of oil and gas pipelines and reduce potential safety risks and losses. The warning event processing system provided by this embodiment can realize real-time analysis, optimization and response to oil and gas pipeline warning events to improve the intelligent management level of the warning system.

[0071] Embodiment 2

[0072] In order to better understand the technical solution provided in Example 1, this example describes in detail the functional modules involved in Example 1.

[0073] In a first specific implementation, in the preprocessing module, the real-time warning data includes: the event type, initial warning level, occurrence time, coordinates and optical fiber number of the warning event.

[0074] Specifically, the real-time warning data can be structured into the following form:

[0075] {

[0076] "alarmLevel":2, / / initial warning level

[0077] "endOffset":551.5, / / The offset end point of the coordinates of the warning event

[0078] "endPileUUID":"4385ffaf-c577-4464-af43-26397598beff", / / The optical fiber UUID (optical fiber number) to which the coordinate offset end point of the warning event belongs

[0079] "eventUUID":"ac7e6a4d-3d4e-4adc-9e3d-0903539e310e", / / Warning event UUID

[0080] "firstReportTime":1724206165110, / / First reporting time of the warning event

[0081] "latestReportTime":1724206405110, / / The latest reporting time of the warning event

[0082] "latitude":35.681786, / / coordinate latitude of the warning event

[0083] "longitude":119.73313, / / The coordinate longitude of the warning event

[0084] "offset":148.5, / / The starting point of the coordinate offset of the warning event

[0085] "riskPointDistance":13463.5, / / Risk point distance

[0086] "startPileUUID":"6589e96f-aba7-4f79-a03a-01d0660dac08" / / The offset of the coordinates of the warning event, the starting point of the fiber UUID (fiber number)

[0087] }

[0088] The preprocessing module provided in this embodiment collects real-time warning data of warning events from different optical fiber sensors and performs preprocessing. The preprocessing method mainly includes: data cleaning and standardizing input data.

[0089] Data cleaning mainly involves removing missing values, outliers and normalizing. Assuming that the initial real-time warning data set is D, and the cleaned real-time warning data set is D′, then

[0090]

[0091] in, represents the i-th type of data in the initial real-time warning data set, Indicates the timestamp of the i-th data, Q 0.01 and Q 0.99 They respectively represent the 1st percentile and 99th percentile of the data value of the i-th type of data in the initial real-time warning data set.

[0092] Data standardization mainly unifies the time format of real-time warning data, the accuracy of coordinate data, the standards for warning event types, the standards for warning levels, and the numerical range of data.

[0093] In a specific implementation, the above data standardization includes: adjusting the numerical range of the data by the following formula.

[0094]

[0095] x′ represents the normalized data value, and x represents the initial data value.

[0096] For the time format of real-time warning data, the accuracy of coordinate data, the standard of warning event type, warning level, etc., one-hot encoding processing can also be performed to convert the initial data values ​​into binary codes.

[0097] In a second specific implementation, the initial merging module is intended to merge the warning events according to the occurrence time, event type and location of the warning events. This embodiment adopts a clustering algorithm to achieve event merging by calculating the similarity between the warning events. Specifically, the initial merging module includes: a first vector construction unit, a first clustering analysis unit and a first merging unit.

[0098] The first vector construction unit is used to construct a first feature vector for each warning event based on the associated pre-processed real-time warning data; the initial feature vector includes: the relative time of the warning event, the event type, and the coordinates of the warning event. The relative time refers to the time from the occurrence to the end of a warning event, that is, the duration of the warning event.

[0099] Specifically, for the i-th warning event, the first feature vector constructed is expressed as:

[0100] V i =[t i ,e i ,x i ,y i ];

[0101] Among them, t i represents the relative time of the ith warning event, e i represents the event type of the i-th warning event, represents the longitude of the i-th warning event, and represents the dimension of the i-th warning event.

[0102] The first cluster analysis unit is used to calculate the similarity between any two warning events based on the first feature vector, and divide the multiple warning events into multiple clusters according to the similarity of the warning events.

[0103] Specifically, the method for calculating the similarity between any two warning events based on the first feature vector is: for the i-th warning event and the j-th warning event, the similarity between the two is calculated based on formula (1);

[0104]

[0105] Among them, d ij Indicates the similarity between the i-th warning event and the j-th warning event.

[0106] The first merging unit determines the warning events under the same cluster as the first similar warning events, and performs an initial merging on the real-time warning data of the similar warning events to obtain a merged warning event.

[0107] Specifically, the specific methods of the above-mentioned initial merger are:

[0108] For all first similar warning events in a cluster, the data under each item of the associated real-time warning data is averaged or median, and the data is used as the real-time warning data associated with the merged warning event.

[0109] In a third specific implementation, for the dynamic upgrade and downgrade processing module, this embodiment uses a dynamic upgrade and downgrade algorithm based on adaptive ensemble learning (AELDR) to dynamically adjust the warning level of the warning event.

[0110] AELDR (Adaptive Ensemble Learning-based Dynamic Rating) is a dynamic upgrade and downgrade algorithm based on adaptive ensemble learning, which aims to dynamically adjust the priority of warning events based on environmental factors, historical factors, historical feedback results and expert database information. The algorithm integrates multiple machine learning models and dynamically adjusts the priority of events based on real-time data and model prediction results to improve processing efficiency and accuracy.

[0111] Specifically, the multidimensional feature vector includes: time feature data, location feature data, event type feature data, environment feature data and historical feature data; wherein,

[0112] The time characteristic data includes: whether the occurrence time of the warning event is in a preset sensitive time period. The time characteristic data converts the timestamp of the warning event into relative time and extracts the time period characteristics of the event, such as daytime, nighttime, weekdays, weekends, etc.

[0113] The location feature data includes: terrain data, geological data or hydrological data obtained based on the coordinates of the warning event based on GIS. The coordinates of the location where the warning event occurred can be converted into a relative position, for example, using the distance relative to the starting point of the pipeline, combined with GIS data, to extract environmental features around the location, such as terrain, geology, hydrology, etc.

[0114] The event type data features include: the frequency and / or duration of the event type to which the warning event belongs.

[0115] The environmental characteristic data include: weather data, traffic conditions or equipment status near the coordinates of the warning event; the equipment status includes: pipeline flow and pipeline status of the oil and gas pipeline.

[0116] The historical feature data includes: the occurrence pattern or trend of the warning event at the warning event coordinates. It is also possible to extract the features of historical feedback results such as user feedback and expert evaluation, and extract keywords and emotional tendencies in the feedback results.

[0117] An expert database can also be set up to provide expert database information such as extracted expert knowledge and experience rules.

[0118] In order to facilitate the dynamic upgrading and downgrading processing module to process the multi-dimensional feature vector, the information gain can also be used as a division criterion to divide the multi-dimensional feature vector of the warning event into multiple data subsets.

[0119]

[0120] Where D is the data set, A is the feature, H(D) is the entropy of the data set, is a value of feature A The corresponding data subset.

[0121] In the multidimensional feature vector, each type of data has relatively uniform processing characteristics, such as time, coordinates, type, warning level, etc. By dividing the subsets to obtain data with more concentrated features, the AELDR model can process each type of data subset.

[0122] In the dynamic upgrading and downgrading processing module, the deep learning model is: an AELDR model with adapted model parameters obtained based on a pre-training process.

[0123] The AELDR model includes:

[0124] The first LSTM model is used to predict a first result about the trend of the warning event based on the multi-dimensional feature vector.

[0125] The second LSTM model is used to predict a second result about the environmental pattern at the coordinates of the warning event based on the real-time multi-dimensional feature vector.

[0126] The CNN model is used to predict a third result about the spatial pattern and local features of the warning event based on the multi-dimensional feature vector.

[0127] The random forest model is used to predict a fourth result about the nonlinear relationship and complex pattern of the warning event based on the multidimensional feature vector.

[0128] The SVM model is used to predict a fifth result about the decision boundary of the warning event based on the high-dimensional features of the multi-dimensional feature vector.

[0129] A fusion model is used to fuse the first result, the second result, the third result, the fourth result and the fifth result to predict the adjusted warning level and recommended decision of the warning event.

[0130] Specifically, the fusion model is a stacked meta-model obtained through a pre-training process and having appropriate model parameters.

[0131] Optionally, the pre-training process includes: acquiring historical data of multiple historical warning events, extracting historical multi-dimensional feature vectors about the historical data, and warning level adjustment records associated with the historical warning events as data sets for training.

[0132] The historical data includes: historical time feature data, historical location feature data, historical event type feature data, historical environment feature data, historical characteristic data and historical decision feature data.

[0133] Based on the AELDR model with adapted model parameters obtained based on the above-mentioned pre-training process, it is possible to extract the time, space, level, and type characteristics of the real-time warning data corresponding to each warning event, and make a comprehensive decision based on geographical environment, historical factors, time, space and other information. For different types of operations, the learning mechanism is improved by collecting event processing feedback. Manual operations are generally delayed and do not need to be intervened for a short time, while mechanical operations require short-term intervention. If the warning event is reported at night, because there are not many activities at night and the probability of false alarms is high, it can be downgraded. If the warning event occurs during the busy farming season, it is easy to cause interference, so it can be downgraded. If there is bad weather at the time of the warning event, factors such as heavy rain are easy to cause interference, so it can be downgraded. If the location of the warning event is on a highway, water flow, etc., it can be downgraded. If there are frequent alarms in a short period of time, but the environmental factors are normal, it can be upgraded. If the location of the event is analyzed based on the historical processing results, it is found that there are too many unnecessary alarms, and the event can be downgraded. Alarms of the same type, range and time period that frequently occur can be merged and upgraded. For example, an artificial alarm occurs at location A. After it is confirmed and eliminated, X minutes later, artificial alarms of different levels continue to appear 100 meters upstream and downstream of location A. At this time, the alarms can be merged and the alarm level is increased.

[0134] In addition, the dynamic upgrade and downgrade process will take the above factors into consideration. If there is an alarm at one place, the first alarm will be confirmed and subsequent alarms will continue. According to the historical alarm frequency and alarm confirmation results, alarms of the same type will be dynamically merged and the alarm level will be adjusted. The merge scope and merge time will be dynamically adjusted, and multiple frequent alarms will be merged and downgraded or the alarm level will be increased. If there are geographical and weather factors at the location where the alarm occurs, it will be further judged whether it is bad weather or bad terrain such as roads, construction sites, and water flows. Noise reduction will also be performed based on these factors. It will also be further judged based on the time of the alarm, day and night, spring, summer, autumn, and winter, and other factors to reduce and increase the alarm level. Finally, it will be determined whether it is an automatic decision based on the historical alarm processing results of this location and the frequency and time of occurrence of the current warning event.

[0135] In a fourth specific implementation, the dynamic up-grading and down-grading processing module and the output module are connected via a secondary merging module.

[0136] The secondary merging module is used to update the real-time warning data of the associated merged warning events according to the adjusted warning level output by the dynamic upgrading and downgrading processing module; and recalculate the similarity of multiple merged warning events based on the updated real-time warning data to determine whether there is a second similar warning event among the multiple merged warning events. If so, the real-time warning data associated with the second similar warning event is merged again to obtain one or more re-merged warning events.

[0137] The secondary merge module removes duplicate pushes and optimized duplicate alarms to avoid data confusion.

[0138] The output module outputs the optimized warning events to trigger corresponding response measures, and sends the optimized warning events to operators to provide real-time monitoring and decision support.

[0139] This embodiment proposes an efficient warning event processing system, which can identify and merge warning events with similar characteristics or time associations, reduce duplication and redundant information, and improve the efficiency of data processing. In addition, this embodiment adopts the AELDR model to dynamically upgrade and downgrade warning events, and can flexibly adjust the priority of events according to the severity of the event, historical data, and real-time environmental factors. This dynamism is relatively lacking in the prior art, and it ensures that critical events can receive timely and appropriate attention. A more efficient, accurate and automated fiber optic early warning system is provided to better protect the safety of oil and gas pipelines and reduce potential safety risks and losses.

[0140] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, EEPROM, optical storage, etc.) containing computer-usable program code.

[0141] The present invention is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present invention. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions.

[0142] It should be noted that in the claims, any reference numerals placed between brackets shall not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claims. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The invention may be implemented by means of hardware comprising several different components and by means of a suitably programmed computer. In the claims enumerating several means, several of these means may be embodied by the same hardware. The use of the words first, second, third, etc., is for convenience of expression only and does not indicate any order. These words may be understood as part of the component name.

[0143] In addition, it should be noted that, in the description of this specification, the description of the terms "one embodiment", "some embodiments", "embodiment", "example", "specific example" or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

[0144] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments after knowing the basic creative concept. Therefore, the claims should be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present invention.

[0145] Obviously, those skilled in the art can make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention should also include these modifications and variations.

Claims

1. A warning event processing system for an oil and gas pipeline optical fiber warning system, characterized in that: include: A preprocessing module, used for receiving and preprocessing the real-time warning data of multiple warning events collected by the optical fiber warning system of the oil and gas pipeline, to obtain the preprocessed real-time warning data; A primary merging module, used to calculate the similarity of multiple warning events according to the preprocessed real-time warning data, determine a first similar warning event among the multiple warning events, and perform a primary merging of the real-time warning data of the first similar warning event to obtain one or more merged warning events; A dynamic upgrade and downgrade processing module, used for obtaining a multi-dimensional feature vector of the merged warning event based on the real-time warning data of the merged warning event; and using a deep learning model to adjust the initial warning level of the merged warning event according to the multi-dimensional feature vector to obtain an adjusted warning level; The output module is used to output the merged warning events and their adjusted warning levels.

2. The warning event processing system according to claim 1, characterized in that: The real-time warning data includes: the event type, initial warning level, occurrence time, coordinates and optical fiber number of the warning event.

3. The warning event processing system according to claim 1, characterized in that: The initial merging module includes: A first vector construction unit is used to construct a first feature vector for each warning event based on the associated pre-processed real-time warning data; the initial feature vector includes: the relative time of the warning event, the event type, and the coordinates of the warning event; a first cluster analysis unit, configured to calculate the similarity between any two warning events based on the first feature vector, and to divide the plurality of warning events into a plurality of clusters according to the similarity of the warning events; The first merging unit determines the warning events under the same cluster as the first similar warning events, and performs an initial merging on the real-time warning data of the similar warning events to obtain a merged warning event.

4. The warning event processing system according to claim 3, characterized in that: In the first vector construction unit, the first feature vector is constructed based on the associated pre-processed real-time warning data, including: for the i-th warning event, the constructed first feature vector is expressed as: V i =[t i ,e i ,x i ,y i ]; where t i represents the relative time of the ith warning event, e i represents the event type of the ith warning event, represents the longitude of the ith warning event, and represents the latitude of the ith warning event; In the first cluster analysis unit, the calculating the similarity between any two warning events based on the first feature vector includes: for the i-th warning event and the j-th warning event, calculating the similarity between the two based on formula (1); Among them, d ij Indicates the similarity between the i-th warning event and the j-th warning event.

5. The warning event processing system according to claim 3, characterized in that: In the first merging unit, the real-time warning data of the first similar warning event is initially merged, including: For all first similar warning events in a cluster, the data under each item of the associated real-time warning data is averaged or median, and the data is used as the real-time warning data associated with the merged warning event.

6. The warning event processing system according to claim 1, characterized in that: The multidimensional feature vector includes: time feature data, location feature data, event type feature data, environmental feature data and historical feature data; wherein, The time characteristic data includes: whether the occurrence time of the warning event is within a preset sensitive time period; The location feature data includes: terrain data, geological data or hydrological data obtained based on the coordinates of the warning event based on GIS; The event type data features include: the frequency and / or duration of the event type to which the warning event belongs; The environmental characteristic data includes: weather data, traffic conditions or equipment status near the coordinates of the warning event; the equipment status includes: pipeline flow and pipeline status of the oil and gas pipeline; The historical characteristic data includes: the occurrence pattern or trend of the warning event at the warning event coordinates.

7. The warning event processing system according to claim 6, characterized in that: In the dynamic upgrading and downgrading processing module, the use of a deep learning model includes: The deep learning model is: an AELDR model with adapted model parameters obtained based on a pre-training process; The AELDR model includes: A first LSTM model is used to predict a first result about the trend of the warning event based on the multi-dimensional feature vector; A second LSTM model is used to predict a second result about the environmental pattern at the coordinates of the warning event based on the real-time multi-dimensional feature vector; A CNN model is used to predict a third result about the spatial pattern and local features of the warning event based on the multi-dimensional feature vector; A random forest model, for predicting a fourth result about a nonlinear relationship and a complex pattern of the warning event based on the multidimensional feature vector; An SVM model is used to predict a fifth result about a decision boundary of the warning event based on the high-dimensional features of the multi-dimensional feature vector; A fusion model is used to fuse the first result, the second result, the third result, the fourth result and the fifth result to predict the adjusted warning level and recommended decision of the warning event.

8. The warning event processing system according to claim 7, characterized in that: The fusion model is a stacked meta-model.

9. The warning event processing system according to claim 7, characterized in that: The pre-training process includes: acquiring historical data of a plurality of historical warning events, extracting historical multi-dimensional feature vectors about the historical data, and warning level adjustment records associated with the historical warning events as data sets for training; The historical data includes: historical time feature data, historical location feature data, historical event type feature data, historical environment feature data, historical characteristic data and historical decision feature data.

10. The warning event processing system according to claim 1, characterized in that: The dynamic upgrade and downgrade processing module and the output module are connected via a secondary merging module; The secondary merging module is used to update the real-time warning data of the associated merged warning events according to the adjusted warning level output by the dynamic upgrade and downgrade processing module; And based on the updated real-time warning data, the similarity of multiple merged warning events is calculated again to determine whether there is a second similar warning event among the multiple merged warning events. If so, the real-time warning data associated with the second similar warning event is merged again to obtain one or more merged warning events.

Images