Method, processor and apparatus for oil and gas pipeline operating condition identification

Abstract

This application relates to the field of computer technology, specifically to a method, processor, device, and storage medium for identifying the operating conditions of oil and gas pipelines. The method includes: acquiring first operating data of the pipeline in real time within a first preset time period according to a preset acquisition frequency; detecting the first operating data using a change point detection algorithm to determine the state change times in the first operating data; segmenting the first preset time period according to the state change times to obtain multiple operating condition time periods, and determining multiple feature values ​​corresponding to the first operating data within each operating condition time period; analyzing the first operating data through correlation analysis to determine a first correlation weight matrix between various operating features in the first operating data; embedding the first correlation weight matrix into an operating condition identification model, and inputting the multiple feature values ​​corresponding to each operating condition time period into the operating condition identification model to output the operating condition corresponding to each operating condition time period.

Description

Technical Field

[0001] This application relates to the field of computer technology, and more specifically to a method, processor, device, and storage medium for identifying the operating conditions of oil and gas pipelines. Background Technology

[0002] Oil and gas pipelines are a vital means of ensuring energy supply. Oil and natural gas are among the world's most important energy resources, and pipeline systems transport these energy sources from production sites to consumption sites, ensuring a stable energy supply for various regions. With the rapid development of oil and gas pipelines and the ever-expanding network size, frequent changes in operating conditions and complex hydraulic variations have made manual monitoring increasingly difficult. It is impossible to dynamically track the pipeline's operational status and address the multiple challenges currently facing refined oil pipelines in terms of operation, safety, and supply security. Summary of the Invention

[0003] The purpose of this application is to provide a method, processor, device, and storage medium for real-time monitoring of pipeline operation status and identification of oil and gas pipeline operating conditions.

[0004] To achieve the above objectives, embodiments of this application provide a method for identifying the operating conditions of oil and gas pipelines, the method comprising:

[0005] The pipeline's first operating data within a first preset time period is acquired in real time according to a preset acquisition frequency. The operating data includes operating flow rate, operating pressure, and operating temperature.

[0006] A change point detection algorithm is used to detect the first running data to determine the state change points in the first running data, and the time corresponding to the state change point in the first running data is determined as the state change time of the pipeline within a first preset time period.

[0007] The first preset duration is segmented according to the time of state change to obtain multiple working condition time periods, and multiple feature values ​​corresponding to the first running data in each working condition time period are determined.

[0008] The first running data is analyzed by correlation analysis to determine the first correlation weight matrix between various running features in the first running data;

[0009] The first correlation weight matrix is ​​embedded into the working condition identification model, and multiple feature values ​​corresponding to each working condition time period are input into the working condition identification model to output the working condition corresponding to each working condition time period.

[0010] In this embodiment, the operating condition identification model includes a GCN model, an LSTM model, and a fully connected layer. The first correlation weight matrix is ​​embedded into the operating condition identification model, and multiple feature values ​​corresponding to each operating condition time period are input into the operating condition identification model to output the operating condition corresponding to each operating condition time period. This includes: embedding the first correlation weight matrix into the GCN model and inputting multiple feature values ​​corresponding to each operating condition time period into a modified GCN model to output spatial features corresponding to each operating condition time period; inputting the spatial features corresponding to each operating condition time period into the LSTM model to output spatiotemporal features corresponding to each operating condition time period; and inputting the spatiotemporal features corresponding to each operating condition time period into the fully connected layer to determine the operating condition corresponding to each operating condition time period.

[0011] In this embodiment of the application, the method further includes: after determining the operating condition corresponding to each operating condition time period, determining the operating condition time period as the operating time of the operating condition corresponding to the operating condition.

[0012] In this embodiment of the application, the state change time is segmented into a first preset duration to obtain multiple working condition time periods, and the multiple feature values ​​corresponding to the first operating data in each working condition time period are determined as follows: for the first operating data in each working condition time period, the mean, variance, maximum value, minimum value, root mean square value, kurtosis and skewness of the first operating data are determined as the feature values ​​corresponding to the first operating data in the working condition time period.

[0013] In this embodiment of the application, the method further includes: acquiring historical operating data of the pipeline within a first preset historical time period before acquiring the first operating data of the pipeline within a first preset time period in real time according to a preset acquisition frequency; segmenting the first preset historical time period according to the first preset time period to obtain multiple second historical time periods and historical training datasets corresponding to each second historical time period; performing dimensionality reduction on each historical training dataset using a dimensionality reduction algorithm to obtain second operating data; and using a change point detection algorithm to detect the second operating data for each second operating data to determine the state change points in the second operating data, and determining the time corresponding to the state change points in the second operating data as the time of the pipeline in the first preset historical time period. The historical state change times within the second historical time period corresponding to the second operating data are analyzed. For each second historical time period, the second historical time period is segmented according to the historical state change times to obtain multiple historical operating condition time periods, and multiple historical feature values ​​corresponding to the second operating data within each historical operating condition time period are determined. For each second historical time period, the second operating data of the second historical time period is analyzed according to correlation analysis to determine the second correlation weight matrix between various operating features in the second operating data. The second correlation weight matrix is ​​embedded into the operating condition recognition model to be trained, and the multiple historical feature values ​​corresponding to each historical operating condition time period are input into the operating condition recognition model to be trained to output the historical recognized operating condition corresponding to each historical operating condition time period.

[0014] In this embodiment, the work condition recognition model to be trained includes a GCN model, an LSTM model, and a fully connected layer. A second correlation weight matrix is ​​embedded into the work condition recognition model, and multiple historical feature values ​​corresponding to each historical work condition time period are input into the work condition recognition model to output the historical recognized work condition corresponding to each historical work condition time period. This includes: for each second historical time period, embedding the second correlation weight matrix into the GCN model, and inputting multiple historical feature values ​​corresponding to each historical work condition time period within the second historical time period into the modified GCN model to output the spatial features corresponding to the historical work condition time period; inputting the spatial features corresponding to each historical work condition time period into the LSTM model to output the spatiotemporal features corresponding to each historical work condition time period; and inputting the spatiotemporal features corresponding to each historical work condition time period into the fully connected layer to output the historical recognized work condition corresponding to each historical work condition time period.

[0015] In this embodiment of the application, the method further includes: for each historical operating condition period in each second historical period, obtaining the actual operating condition corresponding to the historical operating condition period; comparing the actual operating condition with the historical identified operating condition; and determining that the training of the operating condition identification model is completed when the overlap rate between the historical identified operating condition and the actual operating condition reaches a preset ratio.

[0016] A second aspect of this application provides a processor configured to perform the method for identifying operating conditions of oil and gas pipelines according to any of the above embodiments.

[0017] A third aspect of this application provides an apparatus for identifying the operating conditions of oil and gas pipelines, comprising:

[0018] The memory is configured to store instructions; and the processor mentioned above.

[0019] The fourth aspect of this application provides a machine-readable storage medium storing instructions that, when executed by a processor, cause the processor to perform any of the above-described methods for identifying the operating conditions of an oil and gas pipeline.

[0020] The above technical solution allows for real-time acquisition of pipeline operation data, determination of the correlation weight matrix between various operation features in the pipeline operation data, and embedding of the correlation weight matrix into the operating condition identification model. This enables the operating condition identification model to identify the operating condition in real time by combining the degree of influence between features when the feature values ​​of the operation data are obtained.

[0021] Other features and advantages of the embodiments of this application will be described in detail in the following detailed description section. Attached Figure Description

[0022] The accompanying drawings are provided to further illustrate the embodiments of this application and form part of the specification. They are used together with the following detailed description to explain the embodiments of the invention, but do not constitute a limitation on the embodiments of this application. In the drawings:

[0023] Figure 1 The schematic diagram illustrates a process flow diagram of a method for identifying operating conditions of oil and gas pipelines according to an embodiment of this application;

[0024] Figure 2 This illustration schematically shows a working condition identification model operating according to an embodiment of the present application;

[0025] Figure 3 The diagram illustrates the internal structure of a computer device according to an embodiment of this application. Detailed Implementation

[0026] The specific embodiments of this application will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit this application.

[0027] It should be noted that if the embodiments of this application involve directional indicators (such as up, down, left, right, front, back, etc.), the directional indicators are only used to explain the relative positional relationship and movement of the components in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indicators will also change accordingly.

[0028] Furthermore, if the embodiments of this application involve descriptions such as "first" or "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, features defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the technical solutions of various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. If the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed in this application.

[0029] like Figure 1 The diagram illustrates a schematic flow chart of a method for identifying the operating conditions of an oil and gas pipeline according to an embodiment of this application. Figure 1 As shown, a method for identifying the operating conditions of oil and gas pipelines is provided, including the following steps:

[0030] Step 101: Acquire the first operating data of the pipeline within a first preset time period in real time according to the preset acquisition frequency, wherein the operating data includes operating flow rate, operating pressure and operating temperature;

[0031] Step 102: Use the change point detection algorithm to detect the first running data to determine the state change points in the first running data, and determine the time corresponding to the state change points in the first running data as the state change time of the pipeline within the first preset time period.

[0032] Step 103: Divide the first preset duration into segments according to the state change time to obtain multiple working condition time periods, and determine multiple feature values ​​corresponding to the first running data in each working condition time period.

[0033] Step 104: Analyze the first running data through correlation analysis to determine the first correlation weight matrix between various running features in the first running data;

[0034] Step 105: Embed the first correlation weight matrix into the working condition identification model, and input multiple feature values ​​corresponding to each working condition time period into the working condition identification model to output the working condition corresponding to each working condition time period.

[0035] The processor can collect the first operating data of the pipeline within a first preset time period in real time according to a preset sampling frequency. For example, assuming the processor sets the first preset time period to 1 minute based on user input data and the preset sampling frequency to once every 5 seconds, the processor collects the first operating data of the pipeline within one minute in real time at a sampling frequency of once every 5 seconds. This operating data includes operating flow rate, operating pressure, and operating temperature. The processor can use a change point detection algorithm, such as the CUSUM algorithm, to detect the first operating data within the first preset time period to determine the state change points in the first operating data, and determine the time corresponding to the state change point of the first operating data as the state change time of the pipeline within the first preset time period corresponding to the first operating data. After determining the state change time within the first preset duration, the processor can segment the first preset duration according to the state change time to obtain multiple operating condition time periods. For example, assuming there is one state change time within the first preset duration, the first preset duration can be divided into two operating condition time periods according to this state change time. After segmenting the first preset duration, for each operating condition time period obtained by the segmentation, the processor can determine multiple feature values ​​corresponding to the first running data within each operating condition time period. For example, assuming the first preset duration is 1 minute and the state change time is the 30th second, the processor can determine multiple feature values ​​corresponding to the first running data collected in the first 30 seconds and multiple feature values ​​corresponding to the first running data collected in the last 30 seconds.

[0036] In one embodiment, the first preset duration is segmented according to the state change time to obtain multiple working condition time periods, and multiple feature values ​​corresponding to the first operating data in each working condition time period are determined, including: for the first operating data in each working condition time period, the mean, variance, maximum value, minimum value, root mean square value, kurtosis and skewness of the first operating data are determined as the feature values ​​corresponding to the first operating data in the working condition time period.

[0037] After determining multiple operating time periods within a first preset duration, the processor can determine the first running data within each operating time period. For each operating time period, the processor can determine the mean, variance, maximum, minimum, root mean square value, kurtosis, and skewness of the first running data within that operating time period, and determine the above data as the feature values ​​corresponding to the first running data within that operating time period.

[0038] The processor can analyze the pipeline's first operating data within a first preset time period using correlation analysis to determine the first correlation weight matrix between various operating features in the first operating data. For example, the processor can determine the first correlation weight matrix between various features in the operating data, namely operating flow rate, operating pressure, and operating temperature. The processor can embed the first correlation weight matrix into the operating condition identification model and input multiple feature values ​​corresponding to each operating condition time period into the operating condition identification model embedded with the first correlation weight matrix, so that the operating condition corresponding to each operating condition time period can be output through the operating condition identification model.

[0039] In one embodiment, the operating condition identification model includes a GCN model, an LSTM model, and a fully connected layer. A first correlation weight matrix is ​​embedded into the operating condition identification model, and multiple feature values ​​corresponding to each operating condition time period are input into the operating condition identification model to output the operating condition corresponding to each operating condition time period. This includes: embedding the first correlation weight matrix into the GCN model and inputting multiple feature values ​​corresponding to each operating condition time period into a modified GCN model to output spatial features corresponding to each operating condition time period; inputting the spatial features corresponding to each operating condition time period into the LSTM model to output spatiotemporal features corresponding to each operating condition time period; and inputting the spatiotemporal features corresponding to each operating condition time period into the fully connected layer to determine the operating condition corresponding to each operating condition time period.

[0040] The working condition identification model includes the GCN model, the LSTM model, and fully connected layers, for example... Figure 2 As shown, when the processor embeds the first correlation matrix into the operating condition identification model, it also embeds the first correlation weight matrix into the GCN model included in the operating condition identification model. For each operating condition time period, the processor can input multiple feature values ​​corresponding to each operating condition time period into the GCN model that has already embedded the first correlation weight matrix. This allows the GCN model to output spatial features corresponding to each operating condition time period. The spatial features output by the GCN model for each operating condition time period are then input into the LSTM model. The LSTM model extracts temporal features from the spatial features corresponding to each operating condition time period to output spatiotemporal features corresponding to each operating condition time period. Finally, the spatiotemporal features output by the LSTM model for each operating condition time period are input into a fully connected layer. This allows the fully connected layer to output the operating condition corresponding to each operating condition time period, thereby identifying the operating condition of the pipeline within each operating condition time period.

[0041] In one embodiment, the method further includes: after determining the operating condition corresponding to each operating condition time period, determining the operating condition time period as the operating condition running time corresponding to the operating condition.

[0042] After the operating condition identification model determines the operating condition corresponding to each operating condition time period, the operating condition time period can be defined as the operating time of that operating condition. For example, assuming that the operating condition identification model determines the first operating condition time period within a first preset duration as operating condition A and the second operating condition time period as operating condition B, the processor can define the first operating condition time period as the operating time of operating condition A and the second operating condition time period as the operating time of operating condition B.

[0043] In one embodiment, the method further includes: acquiring historical operating data of the pipeline within a first preset historical time period before acquiring the first operating data of the pipeline within a first preset time period in real time according to a preset acquisition frequency; segmenting the first preset historical time period according to the first preset time period to obtain multiple second historical time periods and a historical training dataset corresponding to each second historical time period; performing dimensionality reduction on each historical training dataset using a dimensionality reduction algorithm to obtain second operating data; and using a change point detection algorithm to detect the second operating data for each second operating data to determine the state change points in the second operating data, and determining the time corresponding to the state change points in the second operating data as the time of the pipeline in the first preset historical time period. The historical state change times within the second historical time period corresponding to the second operating data are analyzed. For each second historical time period, the second historical time period is segmented according to the historical state change times to obtain multiple historical operating condition time periods, and multiple historical feature values ​​corresponding to the second operating data within each historical operating condition time period are determined. For each second historical time period, the second operating data of the second historical time period is analyzed according to correlation analysis to determine the second correlation weight matrix between various operating features in the second operating data. The second correlation weight matrix is ​​embedded into the operating condition recognition model to be trained, and the multiple historical feature values ​​corresponding to each historical operating condition time period are input into the operating condition recognition model to be trained to output the historical recognized operating condition corresponding to each historical operating condition time period.

[0044] Before acquiring the first operational data within the pipeline in real time to identify its operating conditions, the processor can train the operating condition identification model. The processor can acquire historical operational data of the pipeline within a first preset historical time period and segment this period according to a first preset duration to obtain multiple second historical time periods and the historical operational data corresponding to each second historical time period. The processor can then define the historical operational data corresponding to each second historical time period as the historical training dataset for that second historical time period. For each historical training dataset, the processor can use a dimensionality reduction algorithm to reduce the dimensionality of the historical training dataset to obtain the second operational data corresponding to each second historical time period. For example, assuming the historical operational data includes the pipeline's historical operating pressure, historical operating flow rate, and historical operating temperature, the processor can use the PCA dimensionality reduction algorithm to reduce the original three columns of data to one dimension. The dimensionality-reduced data is then defined as the second operational data corresponding to each second historical time period. For each second historical period corresponding to the second running data, the processor can use a change point detection algorithm, such as the CUSUM algorithm, to detect the second running data, determine the state change points in the second running data, and determine the time corresponding to the state change points in the second running data as the historical state change time of the pipeline in the second historical period corresponding to the second running data.

[0045] For each second historical time period, after determining the historical state change times within that period, the processor can segment the second historical time period based on these times to obtain multiple historical operating condition time periods. After segmenting the second historical time period, the processor can determine multiple historical feature values ​​corresponding to the second operating data within each segmented historical operating condition time period. Specifically, for the second operating data within each historical operating condition time period, the processor can determine the mean, variance, maximum, minimum, root mean square value, kurtosis, and skewness of the second operating data as multiple historical feature values ​​corresponding to the second operating data within that historical operating condition time period.

[0046] For each second historical time period, the processor can analyze the second operating data of the pipeline within the second historical time period through correlation analysis to determine the second correlation weight matrix between various operating features in the second operating data. The processor can embed the second correlation weight matrix into the working condition recognition model to be trained, and input multiple historical feature values ​​corresponding to each historical working condition time period into the working condition recognition model to be trained with the second correlation weight matrix embedded, so as to output the historical recognition working condition corresponding to each historical working condition time period through the working condition recognition model to be trained.

[0047] In one embodiment, the work condition recognition model to be trained includes a GCN model, an LSTM model, and a fully connected layer. A second correlation weight matrix is ​​embedded into the work condition recognition model, and multiple historical feature values ​​corresponding to each historical work condition time period are input into the work condition recognition model to output the historical recognized work condition corresponding to each historical work condition time period. This includes: for each second historical time period, embedding the second correlation weight matrix into the GCN model, and inputting multiple historical feature values ​​corresponding to each historical work condition time period within the second historical time period into the modified GCN model to output the spatial features corresponding to the historical work condition time period; inputting the spatial features corresponding to each historical work condition time period into the LSTM model to output the spatiotemporal features corresponding to each historical work condition time period; and inputting the spatiotemporal features corresponding to each historical work condition time period into the fully connected layer to output the historical recognized work condition corresponding to each historical work condition time period.

[0048] The training condition recognition model includes a GCN model, an LSTM model, and a fully connected layer. For each second historical time period, the processor can embed the second correlation matrix into the GCN model included in the training condition recognition model. For each historical work condition time period within the second historical time period, the processor can input multiple historical feature values ​​corresponding to each historical work condition time period into the GCN model that has already embedded the second correlation weight matrix. This allows the GCN model to output spatial features corresponding to each historical work condition time period. The spatial features corresponding to each historical work condition time period output by the GCN model are then input into the LSTM model. The LSTM model extracts temporal features from the spatial features corresponding to each historical work condition time period to output spatiotemporal features corresponding to each historical work condition time period. Finally, the spatiotemporal features corresponding to each historical work condition time period output by the LSTM model are input into the fully connected layer. This allows the fully connected layer to output the operating condition corresponding to each historical work condition time period, thereby outputting the historical recognition condition of the pipeline within each historical work condition time period.

[0049] In one embodiment, the method further includes: for each historical operating condition period in each second historical period, obtaining the actual operating condition corresponding to the historical operating condition period; comparing the actual operating condition with the historical identified operating condition; and determining that the training of the operating condition identification model is complete when the overlap rate between the historical identified operating condition and the actual operating condition reaches a preset ratio.

[0050] For each historical operating condition period in each second historical period, the processor can obtain the actual operating condition corresponding to the historical operating condition period and compare the actual operating condition with the historical recognized operating condition output by the operating condition recognition model to be trained. When the overlap rate between the historical recognized operating condition and the actual operating condition reaches a preset ratio, that is, when the accuracy of the historical recognized operating condition output by the operating condition recognition model to be trained reaches a preset ratio, the processor can determine that the operating condition recognition model to be trained has been trained.

[0051] In one embodiment, a processor is provided, configured to perform any of the above-described methods for identifying operating conditions of oil and gas pipelines.

[0052] The above technical solution acquires pipeline operation data in real time, determines the correlation weight matrix between various operation features in the pipeline operation data, and then embeds the correlation weight matrix into the operating condition identification model. This allows the operating condition identification model to identify the operating condition by combining the degree of influence between features when it acquires the feature values ​​of the operation data. The operating condition identification model also includes an LSTM model, which can extract time features. Thus, by combining the degree of influence between various features of the operation data and the degree of time dependence, the operating condition of the pipeline can be identified more accurately in real time.

[0053] In one embodiment, an apparatus for identifying the operating conditions of an oil and gas pipeline is provided, comprising: a memory configured to store instructions; and a processor as described above.

[0054] In one embodiment, a machine-readable storage medium is provided, on which instructions are stored, which, when executed by a processor, cause the processor to perform any of the methods for identifying the operating conditions of oil and gas pipelines described in the above embodiments.

[0055] The memory may include non-permanent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0056] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 3As shown. The computer device includes a processor A01, a network interface A02, memory (not shown), and a database (not shown) connected via a system bus. The processor A01 provides computing and control capabilities. The memory includes internal memory A03 and a non-volatile storage medium A04. The non-volatile storage medium A04 stores an operating system B01, a computer program B02, and a database (not shown). The internal memory A03 provides an environment for the operation of the operating system B01 and the computer program B02 stored in the non-volatile storage medium A04. The database stores relevant data generated during pipeline operation. The network interface A02 communicates with external terminals via a network connection. When executed by the processor A01, the computer program B02 implements a method for identifying the operating conditions of oil and gas pipelines.

[0057] Figure 1 This is a flowchart illustrating a method for identifying the operating conditions of oil and gas pipelines in one embodiment. It should be understood that, although... Figure 1 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 1 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0058] This application provides a device including a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it performs the following steps: acquiring first operating data of a pipeline within a first preset time period in real time according to a preset acquisition frequency, wherein the operating data includes operating flow rate, operating pressure, and operating temperature; detecting the first operating data using a change point detection algorithm to determine state change points in the first operating data, and determining the time corresponding to the state change points in the first operating data as the state change time of the pipeline within the first preset time period; segmenting the first preset time period according to the state change time to obtain multiple operating condition time periods, and determining multiple feature values ​​corresponding to the first operating data within each operating condition time period; analyzing the first operating data through correlation analysis to determine a first correlation weight matrix between various operating features in the first operating data; embedding the first correlation weight matrix into an operating condition identification model, and inputting the multiple feature values ​​corresponding to each operating condition time period into the operating condition identification model to output the operating condition corresponding to each operating condition time period.

[0059] In one embodiment, the operating condition identification model includes a GCN model, an LSTM model, and a fully connected layer. A first correlation weight matrix is ​​embedded into the operating condition identification model, and multiple feature values ​​corresponding to each operating condition time period are input into the operating condition identification model to output the operating condition corresponding to each operating condition time period. This includes: embedding the first correlation weight matrix into the GCN model and inputting multiple feature values ​​corresponding to each operating condition time period into a modified GCN model to output spatial features corresponding to each operating condition time period; inputting the spatial features corresponding to each operating condition time period into the LSTM model to output spatiotemporal features corresponding to each operating condition time period; and inputting the spatiotemporal features corresponding to each operating condition time period into the fully connected layer to determine the operating condition corresponding to each operating condition time period.

[0060] In one embodiment, the method further includes: after determining the operating condition corresponding to each operating condition time period, determining the operating condition time period as the operating condition running time corresponding to the operating condition.

[0061] In one embodiment, the state change time is segmented into a first preset duration to obtain multiple operating condition time periods, and multiple feature values ​​corresponding to the first operating data in each operating condition time period are determined, including: for the first operating data in each operating condition time period, the mean, variance, maximum value, minimum value, root mean square value, kurtosis and skewness of the first operating data are determined as the feature values ​​corresponding to the first operating data in the operating condition time period.

[0062] In one embodiment, the method further includes: acquiring historical operating data of the pipeline within a first preset historical time period before acquiring the first operating data of the pipeline within a first preset time period in real time according to a preset acquisition frequency; segmenting the first preset historical time period according to the first preset time period to obtain multiple second historical time periods and a historical training dataset corresponding to each second historical time period; performing dimensionality reduction on each historical training dataset using a dimensionality reduction algorithm to obtain second operating data; and using a change point detection algorithm to detect the second operating data for each second operating data to determine the state change points in the second operating data, and determining the time corresponding to the state change points in the second operating data as the time of the pipeline in the first preset historical time period. The historical state change times within the second historical time period corresponding to the second operating data are analyzed. For each second historical time period, the second historical time period is segmented according to the historical state change times to obtain multiple historical operating condition time periods, and multiple historical feature values ​​corresponding to the second operating data within each historical operating condition time period are determined. For each second historical time period, the second operating data of the second historical time period is analyzed according to correlation analysis to determine the second correlation weight matrix between various operating features in the second operating data. The second correlation weight matrix is ​​embedded into the operating condition recognition model to be trained, and the multiple historical feature values ​​corresponding to each historical operating condition time period are input into the operating condition recognition model to be trained to output the historical recognized operating condition corresponding to each historical operating condition time period.

[0063] In one embodiment, the work condition recognition model to be trained includes a GCN model, an LSTM model, and a fully connected layer. A second correlation weight matrix is ​​embedded into the work condition recognition model, and multiple historical feature values ​​corresponding to each historical work condition time period are input into the work condition recognition model to output the historical recognized work condition corresponding to each historical work condition time period. This includes: for each second historical time period, embedding the second correlation weight matrix into the GCN model, and inputting multiple historical feature values ​​corresponding to each historical work condition time period within the second historical time period into the modified GCN model to output the spatial features corresponding to the historical work condition time period; inputting the spatial features corresponding to each historical work condition time period into the LSTM model to output the spatiotemporal features corresponding to each historical work condition time period; and inputting the spatiotemporal features corresponding to each historical work condition time period into the fully connected layer to output the historical recognized work condition corresponding to each historical work condition time period.

[0064] In one embodiment, the method further includes: for each historical operating condition period in each second historical period, obtaining the actual operating condition corresponding to the historical operating condition period; comparing the actual operating condition with the historical identified operating condition; and determining that the training of the operating condition identification model is complete when the overlap rate between the historical identified operating condition and the actual operating condition reaches a preset ratio.

[0065] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0066] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0067] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0068] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0069] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0070] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0071] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0072] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0073] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for identifying the operating conditions of oil and gas pipelines, characterized in that, The method includes: The pipeline's first operating data within a first preset time period is acquired in real time according to a preset acquisition frequency. The operating data includes operating flow rate, operating pressure, and operating temperature. A change point detection algorithm is used to detect the first running data to determine the state change points in the first running data, and the time corresponding to the state change point in the first running data is determined as the state change time of the pipeline within a first preset time period. The first preset duration is segmented according to the state change time to obtain multiple working condition time periods, and multiple feature values ​​corresponding to the first running data in each working condition time period are determined. The first running data is analyzed by correlation analysis to determine the first correlation weight matrix between various running features in the first running data; The first correlation weight matrix is ​​embedded into the working condition identification model, and multiple feature values ​​corresponding to each working condition time period are input into the working condition identification model to output the working condition corresponding to each working condition time period. The operating condition identification model includes a GCN model, an LSTM model, and a fully connected layer. The step of embedding the first correlation weight matrix into the operating condition identification model and inputting multiple feature values ​​corresponding to each operating condition time period into the operating condition identification model to output the operating condition corresponding to each operating condition time period includes: embedding the first correlation weight matrix into the GCN model and inputting multiple feature values ​​corresponding to each operating condition time period into the modified GCN model to output spatial features corresponding to each operating condition time period; inputting the spatial features corresponding to each operating condition time period into the LSTM model to output spatiotemporal features corresponding to each operating condition time period; and inputting the spatiotemporal features corresponding to each operating condition time period into the fully connected layer to determine the operating condition corresponding to each operating condition time period.

2. The method for identifying the operating conditions of oil and gas pipelines according to claim 1, characterized in that, The method further includes: After determining the operating conditions corresponding to each operating condition time period, the operating condition time period is determined as the operating time corresponding to the operating condition.

3. The method for identifying the operating conditions of oil and gas pipelines according to claim 1, characterized in that, The first preset duration is segmented according to the state change time to obtain multiple working condition time periods, and multiple feature values ​​corresponding to the first operating data within each working condition time period are determined, including: For the first operating data within each operating condition time period, the mean, variance, maximum, minimum, root mean square value, kurtosis, and skewness of the first operating data are determined as the feature values ​​corresponding to the first operating data within the operating condition time period.

4. The method for identifying the operating conditions of oil and gas pipelines according to claim 1, characterized in that, The method further includes: Before acquiring the first operating data of the pipeline within the first preset time period in real time according to the preset acquisition frequency, acquire the historical operating data of the pipeline within the first preset historical time period; The first preset historical time period is segmented according to the first preset duration to obtain multiple second historical time periods and a historical training dataset corresponding to each second historical time period. For each historical training dataset, a dimensionality reduction algorithm is used to reduce the dimensionality of the historical training dataset to obtain the second running data; For each second running data, a change point detection algorithm is used to detect the second running data to determine the state change point in the second running data, and the time corresponding to the state change point in the second running data is determined as the historical state change time of the pipeline in the second historical time period corresponding to the second running data. For each second historical period, the second historical period is segmented according to the time of historical state change to obtain multiple historical operating condition periods, and multiple historical feature values ​​corresponding to the second operating data in each historical operating condition period are determined. For each second historical period, the second operational data of the second historical period is analyzed based on correlation analysis to determine the second correlation weight matrix between various operational features in the second operational data; The second correlation weight matrix is ​​embedded into the working condition recognition model to be trained, and multiple historical feature values ​​corresponding to each historical working condition period are input into the working condition recognition model to be trained, so as to output the historical recognized working condition corresponding to each historical working condition period.

5. The method for identifying the operating conditions of oil and gas pipelines according to claim 4, characterized in that, The work condition recognition model to be trained includes a GCN model, an LSTM model, and a fully connected layer. The second correlation weight matrix is ​​embedded into the work condition recognition model to be trained, and multiple historical feature values ​​corresponding to each historical work condition period are input into the work condition recognition model to be trained, so that the output historical recognized work conditions corresponding to each historical work condition period include: For each second historical period, the second correlation weight matrix is ​​embedded into the GCN model, and multiple historical feature values ​​corresponding to each historical working condition period within the second historical period are input into the modified GCN model to output spatial features corresponding to the historical working condition period. The spatial features corresponding to each historical working condition period are input into the LSTM model to output the spatiotemporal features corresponding to each historical working condition period. The spatiotemporal features corresponding to each historical working condition period are input into the fully connected layer to output the historical identified working condition corresponding to each historical working condition period.

6. The method for identifying the operating conditions of oil and gas pipelines according to claim 4 or 5, characterized in that, The method further includes: For each historical operating condition period in each second historical period, obtain the actual operating condition corresponding to the historical operating condition period; The actual operating conditions are compared with the historical identified operating conditions; When the overlap rate between the historical identified operating conditions and the actual operating conditions reaches a preset ratio, the training of the operating condition identification model to be trained is determined to be complete.

7. A processor, characterized in that, The processor is configured to perform the method for identifying operating conditions of oil and gas pipelines according to any one of claims 1 to 6.

8. A device for identifying the operating conditions of oil and gas pipelines, characterized in that, include: The memory is configured to store instructions; as well as The processor according to claim 7.

9. A machine-readable storage medium storing instructions thereon, characterized in that, When executed by a processor, this instruction causes the processor to perform the method for identifying operating conditions of oil and gas pipelines according to any one of claims 1 to 6.

Images