A method and device for early warning of oil and gas based on carbon emission control

Abstract

The present invention claims protection for a petroleum and natural gas early warning method and device based on carbon emission control, which completes network element labeling of the petroleum and natural gas carbon emission terminal to be warned, outputs multiple petroleum and natural gas carbon emission early warning network elements and network element data sets; constructs a petroleum and natural gas carbon emission prediction network, completes infrared detection of multiple petroleum and natural gas carbon emission early warning network elements, and outputs the sensor detection result set of each network element; performs completion synchronization on the sensor detection result set of the target petroleum and natural gas carbon emission between two network elements in the multiple petroleum and natural gas carbon emission early warning network elements according to the synchronous interaction strategy, outputs the first propagation footprint of the target petroleum and natural gas carbon emission, merges and evaluates it, and outputs the second propagation footprint of the target petroleum and natural gas carbon emission. The present invention optimizes the panoramic footprint based on the merge processing, uses the light intensity calculation and historical footprint information to complete the prediction and evaluation of the petroleum and natural gas carbon emission, improves the robustness of the panoramic footprint and reduces the dependence on hardware.

Description

Technical Field

[0001] The present invention belongs to the technical field of carbon emission, and in particular, relates to an oil and gas early warning method and device based on carbon emission control. Background Art

[0002] Based on the ecological footprint theory and combined with the material consumption and pollution emission characteristics of oil and gas resource development, the present invention restructures the account system structure of ecological footprint, optimizes the calculation method of relevant accounts and evaluation factors, and proposes an indicator system suitable for the ecological environmental impact assessment of oil and gas mining areas, so as to measure the pressure of oil and gas resource development on the ecological environment from a new perspective. Summary of the invention

[0003] The purpose of the present invention is to provide an oil and gas early warning method and device based on carbon emission control, aiming to propose a learner to reduce the cost of data set production, improve synchronization accuracy, reduce the impact of occlusion and overlap on target tracing and reduce the errors caused by missed detection and false detection of the early warning device, and reduce the strategy's dependence on hardware based on the prediction method of the non-linear model.

[0004] According to a first aspect of the present invention, the present invention claims protection for an oil and gas early warning method based on carbon emission control, comprising:

[0005] Complete network element labeling for the oil and gas carbon emission terminals to be warned, output multiple oil and gas carbon emission warning network elements, and output oil and gas carbon emission warning network element data sets;

[0006] Constructing a petroleum and natural gas carbon emission prediction network, completing infrared detection of the plurality of petroleum and natural gas carbon emission early warning network elements, and outputting a sensor detection result set of each network element;

[0007] Performing completion synchronization on a sensor detection result set of a target oil and gas carbon emission between two network elements in the plurality of oil and gas carbon emission early warning network elements according to a synchronization interaction strategy, and outputting a first propagation footprint of the target oil and gas carbon emission;

[0008] completing merging and evaluating the first propagation footprints, and outputting a second propagation footprint of the target oil and gas carbon emissions;

[0009] The step of merging the first propagation footprints further includes:

[0010] Filter the incomplete footprints whose starting part appears in the middle of the footprint map of the non-first network element from the first propagation footprint, and output the incomplete footprint data set U fp ={fp t ,t=1,2,......};

[0011] Analyze fpt , if the footprint fp t If there is an abnormal network element, the new footprint is marked and output at the abnormal network element and added to the data set U fp ;

[0012] In fp t The footprint output data set U of the unfinished part of the network element and the footprint end is queried within the preset range of the first network element fpt For the footprints of the unfinished network elements in the dataset, compare the unfinished parts with fp t Whether the first network element is adapted and the number of unfinished network elements and fp t Whether the displacement is consistent, if they are both adapted and consistent, update Otherwise discard;

[0013] Then, for the footprints in the data set, we find the target overlapping footprints through the strategy of maximum propagation path correlation and shortest distance between the first and the last, and convert fp t Overlap the target overlapping footprint, and the output new footprint is recorded as the merged footprint fp t ;

[0014] Analyze fp t The end network element of the footprint. If the end network element is close to the edge or the network element where the end network element is located is the last network element of the terminal, fp t Footprints from dataset U fp Otherwise, in fp t In the next network element of the network element where the last network element is located, check whether there is a footprint of the first network element at the adjacent coordinates of the last network element. If there is a footprint, find the target overlapping footprint through the strategy of maximum propagation path correlation and shortest distance between the first and the last network element. t The incomplete footprint overlaps with the target overlapping footprint, and the overlapping footprint is recorded as the merged footprint Fp t ;

[0015] Loop until U fp When all processing is completed or the amount of data in the data set does not change, discard U fp The footprints in.

[0016] Furthermore, before constructing the oil and gas carbon emission prediction network, the method further includes:

[0017] The historical record footprints are first filtered through keyword information with different priorities and then the filtered data set is further filtered through outlier information filtering;

[0018] The keyword information filtering combines two footprints with different priorities into one footprint to reduce the discreteness of the footprints and increase the number of valid historical records. The abnormal value information filtering shrinks and overlaps the historical record footprints into a new footprint.

[0019] Furthermore, the construction of the oil and gas carbon emission prediction network also includes:

[0020] A data set is constructed based on a learner, wherein the learner first manually labels the oil and gas carbon emission targets of some network elements in the historical record terminal to construct a basic model;

[0021] The basic model then outputs the defective labeled data for the early warning of the oil and gas carbon emissions of other network elements, and after manual adjustment, all the data are used as the construction set to re-build the model.

[0022] Furthermore, the oil and gas carbon emission prediction network also includes:

[0023] It is constructed based on the target early warning strategy based on collaborative filtering, including fuzzy prediction network, iterative prediction network, and regular prediction network;

[0024] The fuzzy prediction network is established based on the industry-specified carbon emission data pre-allocated to each carbon footprint area, and the total carbon emission of the subsequent time series period is compared with the fuzzy prediction network to output the predicted carbon emission;

[0025] The iterative prediction network completes statistics on the total carbon emissions of each known carbon footprint area, and adjusts the fuzzy prediction network in combination with constraint association, outputs an iterative prediction network, compares the total carbon emissions of subsequent time series periods with the iterative prediction network, and outputs predicted carbon emissions;

[0026] The regular prediction network completes the statistics of the total carbon emissions of each known carbon footprint area to form a standardized carbon emission footprint library, establishes a regular prediction network based on the standardized carbon emission footprint library and association rule analysis, compares the total carbon emissions of subsequent time series periods with the regular prediction network, and outputs predicted carbon emissions.

[0027] Further, the performing completion synchronization on the sensor detection result set of the target oil and gas carbon emission between two network elements in the multiple oil and gas carbon emission early warning network elements according to the synchronization interaction strategy and outputting the first propagation footprint of the target oil and gas carbon emission also includes:

[0028] The synchronization interaction strategy is based on a multi-strategy synchronization method, including a life cycle correlation maximum strategy and a propagation path effectiveness strategy;

[0029] The maximum life cycle correlation strategy is that the oil and gas carbon emission changes from the t-1th network element to the tth network element through the oil and gas carbon emission identification model output with the highest feature completion;

[0030] The propagation path effectiveness strategy refers to the highest displacement and propagation path data volume and propagation trajectory completion of oil and gas carbon emissions from the t-1th network element to the tth network element;

[0031] The priorities of the maximum lifecycle correlation strategy and the propagation path effectiveness strategy are different, wherein the priority of the maximum lifecycle correlation strategy is greater than the priority of the propagation path effectiveness strategy.

[0032] Furthermore, it also includes:

[0033] Calculate the oil and gas carbon emissions of the t-1th network element With the tth network element The characteristic completion degree s of oil and natural gas carbon emissions corresponds to the synchronization degree m output s , according to the path validity strategy, the corresponding synchronization degree m is output v , output the final synchronization degree M by adding the preset priorities;

[0034] Combined with the synchronization state of the lost network element when there is an update traceability, the M output synchronization matrix of all oil and gas carbon emissions in the t-1th network element and the oil and gas carbon emissions in the tth network element is calculated;

[0035] According to the synchronous interaction strategy, synchronous interaction is completed for the carbon emissions of oil and natural gas in the t-1th network element and the tth network element;

[0036] The synchronization interaction is cyclically performed until all adjacent network elements complete synchronization, and the first propagation footprint of the target oil and gas carbon emissions is output.

[0037] Furthermore, the characteristic completion of calculating oil and gas carbon emissions also includes:

[0038] The distance d is calculated as follows:

[0039]

[0040] in, are the coordinates of the oil and gas carbon emission m in the tth network element, represents the distance between the oil and gas carbon emission m in the tth network element and the oil and gas carbon emission n in the t-1th network element, are the coordinates of the oil and gas carbon emission n in the t-1th network element respectively;

[0041] The feature completion degree uses the dot multiplication operation based on the formula:

[0042]

[0043] in is the carbon footprint intensity vector expression of the oil and gas carbon emission s feature in the t-1th network element, is the carbon footprint intensity vector expression of the k-th feature of oil and gas carbon emissions in the t-th network element;

[0044] Propagation path validity strategy in inferring oil and gas carbon emissions from t-1 network element When the tth network element appears at the point Distance Consider at the same time Directions There is no large angle change in the next network element, according to the formula:

[0045]

[0046] where τ s ,τ v is a hyperparameter, represents the path data volume of oil and gas carbon emission m of the t-1th network element, represents the angle between the carbon emission of oil and gas of the tth network element m and the carbon emission of oil and gas of the t-1th network element n, represents the propagation path of oil and gas carbon emission m of the t-1th network element;

[0047] By priority w s ,w v Add the total synchronization of the output M = w s m s +w v m v .

[0048] Further, completing the evaluation process for the first propagation footprint further includes:

[0049] Obtain the unfinished footprints in the middle of the merged footprints. If the tth network element is not completed, calculate the carbon emissions of oil and gas The light intensity of the output And the amount of light intensity data and the transmission path of oil and gas carbon emissions and path data volume

[0050] The direction of the carbon footprint intensity vector is predicted by adding the light intensity and path output according to the preset priority: The amount of predicted carbon footprint intensity vector data is

[0051] The predicted carbon footprint intensity vector is added as a footprint to the merged footprint, and the process is repeated until the entire panoramic footprint is complete, and a second propagation footprint of the target oil and gas carbon emissions is output.

[0052] According to a second aspect of the present invention, the present invention claims protection for an oil and gas early warning device based on carbon emission control, comprising:

[0053] The oil and gas carbon emission network element labeling unit completes the network element labeling of the oil and gas carbon emission terminal to be warned, outputs multiple oil and gas carbon emission warning network elements, and outputs the oil and gas carbon emission warning network element data set;

[0054] The oil and gas carbon emission characteristic early warning unit builds an oil and gas carbon emission prediction network, completes infrared detection of the plurality of oil and gas carbon emission early warning network elements, and outputs a sensor detection result set of each network element;

[0055] A feature similarity synchronization unit performs completion synchronization on a sensor detection result set of a target oil and gas carbon emission between two network elements in the plurality of oil and gas carbon emission early warning network elements according to a synchronization interaction strategy, and outputs a first propagation footprint of the target oil and gas carbon emission;

[0056] A panoramic footprint optimization unit performs merging and evaluation processing on the first propagation footprint, and outputs a second propagation footprint of the target oil and gas carbon emissions;

[0057] The oil and gas early warning device based on carbon emission control is used to execute the oil and gas early warning method based on carbon emission control.

[0058] The solution of the present invention constructs a petroleum and natural gas carbon emission target recognition model through a learner, and uses the model to automatically warn of the petroleum and natural gas carbon emission targets in each network element in the footprint sequence. At the same time, the characteristic expression carbon footprint intensity vector of each petroleum and natural gas carbon emission target is saved for synchronous interaction. During the synchronous interaction, multi-strategy synchronization is used to improve the matching rate. The loss rate caused by the error of the early warning device and the occlusion overlap is reduced based on the method of updating the synchronization of the lost network element of the traceability. The panoramic footprint is optimized based on the merging process, and the oil and natural gas carbon emission is predicted and evaluated using the light intensity calculation and historical footprint information, thereby improving the robustness of the panoramic footprint and reducing the dependence on hardware. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 A flowchart of an oil and gas early warning method based on carbon emission control according to an embodiment of the present invention;

[0060] Figure 2 This is a structural diagram of an oil and gas early warning device based on carbon emission control involved in an embodiment of the present invention. DETAILED DESCRIPTION

[0061] The following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of the present invention.

[0062] According to the first embodiment of the present invention, referring to the attached Figure 1 The present invention claims protection for an oil and gas early warning method based on carbon emission control, comprising:

[0063] Complete network element labeling for the oil and gas carbon emission terminals to be warned, output multiple oil and gas carbon emission warning network elements, and output oil and gas carbon emission warning network element data sets;

[0064] Build an oil and gas carbon emission prediction network, complete infrared detection of multiple oil and gas carbon emission early warning network elements, and output the sensor detection result set of each network element;

[0065] According to the synchronization interaction strategy, the sensor detection result set of the target oil and gas carbon emission between two network elements in the multiple oil and gas carbon emission early warning network elements is synchronized to output the first propagation footprint of the target oil and gas carbon emission;

[0066] The first propagation footprint is merged and evaluated, and the second propagation footprint of the target oil and gas carbon emissions is output.

[0067] Furthermore, before building the oil and gas carbon emission prediction network, it also includes:

[0068] The historical record footprints are first filtered through keyword information with different priorities and then the filtered data set is further filtered through outlier information filtering;

[0069] Keyword information filtering combines two footprints with different priorities into one footprint to reduce the discreteness of footprints and increase the number of valid historical records. Outlier information filtering shrinks and overlaps historical record footprints into a new footprint.

[0070] Furthermore, the construction of oil and gas carbon emission prediction network also includes:

[0071] The learner constructs a data set based on the learner. The learner first manually labels the oil and gas carbon emission targets of some network elements in the historical record terminal to build a basic model.

[0072] The basic model then completes the early warning output of the oil and gas carbon emissions of other network elements with defective labeled data, and after manual adjustment, all the data is used as the construction set to rebuild the model again.

[0073] Furthermore, the oil and gas carbon emission prediction network also includes:

[0074] It is constructed based on the target early warning strategy based on collaborative filtering, including fuzzy prediction network, iterative prediction network, and regular prediction network;

[0075] The fuzzy prediction network is established based on the industry-specified carbon emission data pre-allocated to each carbon footprint area, and the total carbon emission of the subsequent time series period is compared with the fuzzy prediction network to output the predicted carbon emission;

[0076] The iterative prediction network completes statistics on the total carbon emissions of each known carbon footprint area, and adjusts the fuzzy prediction network in combination with constraint association, outputs an iterative prediction network, compares the total carbon emissions of subsequent time series periods with the iterative prediction network, and outputs predicted carbon emissions;

[0077] The regular prediction network completes the statistics of the total carbon emissions of each known carbon footprint area to form a standardized carbon emission footprint library, establishes a regular prediction network based on the standardized carbon emission footprint library and association rule analysis, compares the total carbon emissions of subsequent time series periods with the regular prediction network, and outputs predicted carbon emissions.

[0078] Further, according to the synchronization interaction strategy, the sensor detection result set of the target oil and gas carbon emission between two network elements in the multiple oil and gas carbon emission early warning network elements is synchronized to output the first propagation footprint of the target oil and gas carbon emission, and further includes:

[0079] The synchronous interaction strategy is based on a multi-strategy synchronization method, including the maximum life cycle correlation strategy and the propagation path effectiveness strategy;

[0080] The strategy with the maximum life cycle correlation is that the oil and gas carbon emissions change from the t-1th network element to the tth network element, and the characteristic completion degree output by the oil and gas carbon emissions identification model is the highest;

[0081] The propagation path effectiveness strategy refers to the displacement of oil and gas carbon emissions from the t-1th network element to the tth network element, and the data volume and propagation trajectory completion of the propagation path are the highest;

[0082] The priorities of the maximum lifecycle correlation strategy and the propagation path effectiveness strategy are different, among which the priority of the maximum lifecycle correlation strategy is greater than the priority of the propagation path effectiveness strategy.

[0083] Furthermore, it also includes:

[0084] Calculate the oil and gas carbon emissions of the t-1th network element With the tth network element The characteristic completion degree s of oil and natural gas carbon emissions corresponds to the synchronization degree m outputs , according to the path validity strategy, the corresponding synchronization degree m is output v , output the final synchronization degree M by adding the preset priorities;

[0085] Combined with the synchronization state of the lost network element when there is an update traceability, the M output synchronization matrix of all oil and gas carbon emissions in the t-1th network element and the oil and gas carbon emissions in the tth network element is calculated;

[0086] According to the synchronous interaction strategy, the synchronous interaction of the carbon emissions of oil and gas in the t-1th network element and the tth network element is completed;

[0087] The synchronization interaction is cyclically performed until all adjacent network elements complete synchronization, and the first propagation footprint of the target oil and gas carbon emissions is output.

[0088] Furthermore, the characteristic completion of calculating oil and gas carbon emissions also includes:

[0089] The distance d is calculated as follows:

[0090]

[0091] in, are the coordinates of the oil and gas carbon emission m in the tth network element, represents the distance between the oil and gas carbon emission m in the tth network element and the oil and gas carbon emission n in the t-1th network element, are the coordinates of the oil and gas carbon emission n in the t-1th network element respectively;

[0092] The feature completion is calculated using the dot multiplication formula:

[0093]

[0094] in is the carbon footprint intensity vector expression of the oil and gas carbon emission s feature in the t-1th network element, is the carbon footprint intensity vector expression of the k-th feature of oil and gas carbon emissions in the t-th network element;

[0095] Propagation path validity strategy in inferring oil and gas carbon emissions from t-1 network element When the tth network element appears at the point Distance Consider at the same time Directions There is no large angle change in the next network element, according to the formula:

[0096]

[0097] where τ s ,τv is a hyperparameter, represents the path data volume of oil and gas carbon emission m of the t-1th network element, represents the angle between the carbon emission of oil and gas of the tth network element m and the carbon emission of oil and gas of the t-1th network element n, represents the propagation path of oil and gas carbon emission m of the t-1th network element;

[0098] By priority w s ,w v Add the total synchronization of the output M = w s m s +w v m v .

[0099] Furthermore, completing the merging process for the first propagation footprints also includes:

[0100] Filter the incomplete footprints whose starting part appears in the middle of the footprint map of the non-first network element from the first propagation footprint, and output the incomplete footprint dataset U fp ={fp t ,t=1,2,......};

[0101] Analyze fp t , if the footprint fp t If there is an abnormal network element, the new footprint is marked and output at the abnormal network element and added to the data set U fp ;

[0102] In fp t Query the footprints of the unfinished network elements and the footprint output data set at the end of the footprint within the preset range of the first network element For the footprints of the unfinished network elements in the dataset, compare the unfinished parts with fp t Whether the first network element is adapted and the number of unfinished network elements and fp t Whether the displacement is consistent, if they are both adapted and consistent, update Otherwise discard;

[0103] Then, for the footprints in the data set, we find the target overlapping footprints through the strategy of maximum propagation path correlation and shortest distance between the first and the last, and convert fp t Overlap the target overlapping footprint, and the output new footprint is recorded as the merged footprint fp t ;

[0104] Analyze fp t The end network element of the footprint. If the end network element is close to the edge or the network element where the end network element is located is the last network element of the terminal, fp t Footprints from dataset U fp Otherwise, in fp tIn the next network element of the network element where the last network element is located, check whether there is a footprint of the first network element at the adjacent coordinates of the last network element. If there is a footprint, find the target overlapping footprint through the strategy of maximum propagation path correlation and shortest distance between the first and the last network element. t The incomplete footprint overlaps with the target overlapping footprint, and the overlapping footprint is recorded as the merged footprint Fp t ;

[0105] Loop until U fp When all processing is completed or the amount of data in the data set does not change, discard U fp The footprints in.

[0106] Further, completing the evaluation process for the first propagation footprint further includes:

[0107] There are unfinished footprints in the middle of the merged footprints. If the tth network element is not completed, the carbon emissions of oil and gas are calculated. The light intensity of the output And the amount of light intensity data and the transmission path of oil and gas carbon emissions and path data volume

[0108] The direction of the carbon footprint intensity vector is predicted by adding the light intensity and path output according to the preset priority: The amount of predicted carbon footprint intensity vector data is

[0109] The predicted carbon footprint intensity vector is added as a footprint to the merged footprint, and the cycle is repeated until the entire panoramic footprint is complete, and the second propagation footprint of the target oil and gas carbon emissions is output.

[0110] According to the second embodiment of the present invention, referring to Figure 2 The present invention claims protection for an oil and gas early warning device based on carbon emission control, comprising:

[0111] The oil and gas carbon emission network element labeling unit completes the network element labeling of the oil and gas carbon emission terminal to be warned, outputs multiple oil and gas carbon emission warning network elements, and outputs the oil and gas carbon emission warning network element data set;

[0112] The oil and gas carbon emission characteristic early warning unit builds an oil and gas carbon emission prediction network, completes infrared detection of multiple oil and gas carbon emission early warning network elements, and outputs the sensor detection result set of each network element;

[0113] The feature similarity synchronization unit performs completion synchronization on the sensor detection result set of the target oil and gas carbon emission between two network elements in the multiple oil and gas carbon emission early warning network elements according to the synchronization interaction strategy, and outputs the first propagation footprint of the target oil and gas carbon emission;

[0114] The panoramic footprint optimization unit completes the merging and evaluation processing of the first propagation footprint and outputs the second propagation footprint of the target oil and gas carbon emissions;

[0115] A petroleum and natural gas early warning device based on carbon emission control is used to execute a petroleum and natural gas early warning method based on carbon emission control.

[0116] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A petroleum and natural gas early warning method based on carbon emission control, characterized in that: include: Complete network element labeling for the oil and gas carbon emission terminals to be warned, output multiple oil and gas carbon emission warning network elements, and output oil and gas carbon emission warning network element data sets; Constructing a petroleum and natural gas carbon emission prediction network, completing infrared detection of the plurality of petroleum and natural gas carbon emission early warning network elements, and outputting a sensor detection result set of each network element; Performing completion synchronization on a sensor detection result set of a target oil and gas carbon emission between two network elements in the plurality of oil and gas carbon emission early warning network elements according to a synchronization interaction strategy, and outputting a first propagation footprint of the target oil and gas carbon emission; completing merging and evaluating the first propagation footprints, and outputting a second propagation footprint of the target oil and gas carbon emissions; The step of merging the first propagation footprints further includes: Filter the incomplete footprints whose starting part appears in the middle of the footprint map of the non-first network element from the first propagation footprint, and output the incomplete footprint data set U fp ={fp t ,t=1,2,......}; Analysis of footprint fp t If there is an abnormal network element, the new footprint is marked and output at the abnormal network element and added to the data set U fp ; In fp t The footprint output data set U of the unfinished part of the network element and the footprint end is queried within the preset range of the first network element fpt For the footprints of the unfinished network elements in the dataset, compare the unfinished parts with fp t Whether the first network element is adapted and the number of unfinished network elements and fp t Whether the displacement is consistent, if they are both adapted and consistent, update the U fpt Otherwise discard; Then, for the footprints in the data set, we find the target overlapping footprints through the strategy of maximum propagation path correlation and shortest distance between the first and the last, and convert fp t Overlap the target overlapping footprint, and the output new footprint is recorded as the merged footprint fp t ; Analyze fp t The end network element of the footprint. If the end network element is close to the edge or the network element where the end network element is located is the last network element of the terminal, fp t Footprints from dataset U fp Otherwise, in fp t In the next network element of the network element where the last network element is located, check whether there is a footprint of the first network element at the adjacent coordinates of the last network element. If there is a footprint, find the target overlapping footprint through the strategy of maximum propagation path correlation and shortest distance between the first and the last network element. t The incomplete footprint overlaps with the target overlapping footprint, and the overlapping footprint is recorded as the merged footprint fp t ; Loop until U fp When all processing is completed or the amount of data in the data set does not change, discard U fp Footprints in The step of performing completion synchronization on a sensor detection result set of a target oil and gas carbon emission between two network elements in the plurality of oil and gas carbon emission early warning network elements according to a synchronization interaction strategy and outputting a first propagation footprint of the target oil and gas carbon emission also includes: The synchronization interaction strategy is based on a multi-strategy synchronization method, including a life cycle correlation maximum strategy and a propagation path effectiveness strategy; The maximum life cycle correlation strategy is that the oil and gas carbon emission changes from the t-1th network element to the tth network element through the oil and gas carbon emission identification model output with the highest feature completion; The propagation path effectiveness strategy refers to the highest displacement and propagation path data volume and propagation trajectory completion of oil and gas carbon emissions from the t-1th network element to the tth network element; The priorities of the maximum lifecycle correlation strategy and the propagation path effectiveness strategy are different, wherein the priority of the maximum lifecycle correlation strategy is greater than the priority of the propagation path effectiveness strategy.

2. The oil and gas early warning method based on carbon emission control according to claim 1 is characterized in that: Before constructing the oil and gas carbon emission prediction network, the method further includes: The historical record footprints are first filtered through keyword information with different priorities and then the filtered data set is further filtered through outlier information filtering; The keyword information filtering combines two footprints with different priorities into one footprint to reduce the discreteness of the footprints and increase the number of valid historical records. The abnormal value information filtering shrinks and overlaps the historical record footprints into a new footprint.

3. The oil and gas early warning method based on carbon emission control according to claim 2 is characterized in that: The construction of the oil and gas carbon emission prediction network also includes: A data set is constructed based on a learner, wherein the learner first manually labels the oil and gas carbon emission targets of some network elements in the historical record terminal to construct a basic model; The basic model then outputs the defective labeled data for the early warning of the oil and gas carbon emissions of other network elements, and after manual adjustment, all the data are used as the construction set to re-build the model.

4. The oil and gas early warning method based on carbon emission control according to claim 2 is characterized in that: The oil and gas carbon emission prediction network also includes: It is constructed based on the target early warning strategy based on collaborative filtering, including fuzzy prediction network, iterative prediction network, and regular prediction network; The fuzzy prediction network is established based on the industry-specified carbon emission data pre-allocated to each carbon footprint area, and the total carbon emission of the subsequent time series period is compared with the fuzzy prediction network to output the predicted carbon emission; The iterative prediction network completes statistics on the total carbon emissions of each known carbon footprint area, and adjusts the fuzzy prediction network in combination with constraint association, outputs an iterative prediction network, compares the total carbon emissions of subsequent time series periods with the iterative prediction network, and outputs predicted carbon emissions; The regular prediction network completes the statistics of the total carbon emissions of each known carbon footprint area to form a standardized carbon emission footprint library, establishes a regular prediction network based on the standardized carbon emission footprint library and association rule analysis, compares the total carbon emissions of subsequent time series periods with the regular prediction network, and outputs predicted carbon emissions.

5. The oil and gas early warning method based on carbon emission control according to claim 4 is characterized in that: Also includes: Calculate the oil and gas carbon emissions of the t-1th network element With the tth network element The characteristic completion degree s of oil and natural gas carbon emissions corresponds to the synchronization degree m output s , according to the path validity strategy, the corresponding synchronization degree m is output v , output the final synchronization degree M by adding the preset priorities; Combined with the synchronization state of the lost network element when there is an update traceability, the M output synchronization matrix of all oil and gas carbon emissions in the t-1th network element and the oil and gas carbon emissions in the tth network element is calculated; According to the synchronous interaction strategy, synchronous interaction is completed for the carbon emissions of oil and natural gas in the t-1th network element and the tth network element; The synchronization interaction is cyclically performed until all adjacent network elements complete synchronization, and the first propagation footprint of the target oil and gas carbon emissions is output.

6. The oil and gas early warning method based on carbon emission control according to claim 5 is characterized in that: The feature completion for calculating carbon emissions from oil and gas also includes: The distance d is calculated as follows: in, are the coordinates of the oil and gas carbon emission m in the tth network element, represents the distance between the oil and gas carbon emission m in the tth network element and the oil and gas carbon emission n in the t-1th network element, are the coordinates of the oil and gas carbon emission n in the t-1th network element respectively; The feature completion degree uses the dot multiplication operation based on the formula: in is the carbon footprint intensity vector expression of the oil and gas carbon emission s feature in the t-1th network element, is the carbon footprint intensity vector expression of the k-th feature of oil and gas carbon emissions in the t-th network element; Propagation path validity strategy in inferring oil and gas carbon emissions from t-1 network element When the tth network element appears at the point Distance Consider at the same time Directions There is no large angle change in the next network element, according to the formula: where τ v is a hyperparameter, represents the path data volume of oil and gas carbon emission m of the t-1th network element, represents the angle between the carbon emission of oil and gas of the tth network element m and the carbon emission of oil and gas of the t-1th network element n, represents the propagation path of oil and gas carbon emission m of the t-1th network element; By priority w s ,w v Add the total synchronization of the output M = w s m s +w v m v .

7. The oil and gas early warning method based on carbon emission control according to claim 6 is characterized in that: Completing the evaluation process for the first propagation footprint also includes: Obtain the unfinished footprints in the middle of the merged footprints. If the tth network element is not completed, calculate the carbon emissions of oil and gas The light intensity of the output And the amount of light intensity data and the transmission path of oil and gas carbon emissions and path data volume The direction of the carbon footprint intensity vector is predicted by adding the light intensity and path output according to the preset priority: The amount of predicted carbon footprint intensity vector data is The predicted carbon footprint intensity vector is added as a footprint to the merged footprint, and the process is repeated until the entire panoramic footprint is complete, and a second propagation footprint of the target oil and gas carbon emissions is output.

8. An oil and gas early warning device based on carbon emission control, characterized in that: include: The oil and gas carbon emission network element labeling unit completes the network element labeling of the oil and gas carbon emission terminal to be warned, outputs multiple oil and gas carbon emission warning network elements, and outputs the oil and gas carbon emission warning network element data set; The oil and gas carbon emission characteristic early warning unit builds an oil and gas carbon emission prediction network, completes infrared detection of the plurality of oil and gas carbon emission early warning network elements, and outputs a sensor detection result set of each network element; A feature similarity synchronization unit performs completion synchronization on a sensor detection result set of a target oil and gas carbon emission between two network elements in the plurality of oil and gas carbon emission early warning network elements according to a synchronization interaction strategy, and outputs a first propagation footprint of the target oil and gas carbon emission; A panoramic footprint optimization unit performs merging and evaluation processing on the first propagation footprint, and outputs a second propagation footprint of the target oil and gas carbon emissions; The oil and gas early warning device based on carbon emission control is used to execute the oil and gas early warning method based on carbon emission control as described in any one of claims 1-7.

Images