An intelligent monitoring and analysis method for carbon emissions from overhead transmission lines

Abstract

The present invention discloses an intelligent monitoring and analysis method for carbon emissions of overhead transmission lines, comprising constructing a cloud platform for collecting carbon emission data of overhead transmission lines, collecting carbon emission data of overhead transmission lines in real time during operation at different specified time periods; using a deep isolation forest algorithm to detect abnormal state sampling data and discarding it; constructing a carbon emission prediction model, outputting predicted sequence data of each future time point through monitoring data in a key node detection list; constructing a key node processing model, feeding back processed data based on a non-safe key node list, and updating the model to a three-dimensional digital twin carbon emission monitoring model of overhead transmission lines; the present invention can realize real-time monitoring and early warning of carbon emissions of overhead transmission lines, realize information-based and scientific management of carbon emission monitoring of overhead transmission lines, reduce the updating frequency of the digital twin carbon emission monitoring model, and reduce the carbon emissions of transmission lines, thereby ensuring that the operation of overhead transmission lines is more stable and energy-efficient.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence monitoring and early warning technology, and in particular to a method for intelligent monitoring and analysis of carbon emissions from overhead transmission lines. Background Art

[0002] With growing global attention to environmental protection and sustainable development, carbon emissions monitoring has become increasingly important across various industries. As a key component of power transmission, monitoring and analyzing carbon emissions from overhead transmission lines is crucial for assessing and managing the carbon footprint of power systems. Accurately monitoring and analyzing carbon emissions from overhead transmission lines can provide a scientific basis for energy conservation and emission reduction within power systems, effectively addressing climate change.

[0003] Due to the complexity of carbon emissions monitoring for overhead transmission lines, traditional two-dimensional design often encounters issues such as delayed information transfer between upstream and downstream disciplines and difficulty managing data. With the rise of digital technology, the use of three-dimensional digital twins (3D digital twins) has effectively enabled interdisciplinary design collaboration and addressed the integration and updating of massive amounts of data. Therefore, the use of 3D digital twins for digital processing of carbon emissions monitoring for overhead transmission lines has become mainstream in the industry. However, numerous challenges remain. First, there is a lack of data forecasting for carbon emissions monitoring of overhead transmission lines, making it difficult to accurately and timely detect abnormal carbon emissions through quantitative methods, resulting in high carbon emissions. Second, due to the large number of monitoring nodes, varying monitoring cycles, and different carbon emission safety thresholds for each node, each monitoring node has its own maintenance deadline, lacking a unified plan and management. This results in cumbersome updates to the 3D digital twin platform for carbon emissions monitoring of overhead transmission lines, data redundancy, and a lack of intelligent processing. Summary of the Invention

[0004] The purpose of the present invention is to provide an intelligent monitoring and analysis method for carbon emissions of overhead transmission lines to solve the above-mentioned problems existing in the prior art.

[0005] The specific application is as follows:

[0006] An intelligent monitoring and analysis method for carbon emissions from overhead transmission lines, comprising the following steps:

[0007] S1. Build a cloud platform for collecting carbon emission data for overhead transmission lines. This platform collects and summarizes carbon emission data generated by resistance loss, corona loss, and insulator leakage loss during the operation of overhead transmission lines in real time, based on specified time periods.

[0008] S2. Create several key node detection lists, monitor the carbon emission data generated by resistance loss, corona loss, and insulator leakage loss respectively, record each corresponding monitoring data in the key node detection list, and record the time period of each monitoring key node;

[0009] S3. Use the deep isolation forest algorithm to detect abnormal states of carbon emission data in all key node detection lists, and discard abnormal sampling data for preprocessing;

[0010] S4. Build a carbon emission prediction model, output the predicted sequence data for each future time point based on the monitoring data in the key node detection list, and mark the time period where the safety threshold is located based on the safety threshold of each monitoring data;

[0011] S5. Build a key node processing model, compare the sequence data predicted by the key node detection list with the safety threshold, determine whether the key node is safe, output a uniformly maintained list of non-safe key nodes based on the marked time period, and feed back the processed data based on the non-safe key node list to update the three-dimensional digital twin carbon emission monitoring model of the overhead transmission line.

[0012] Furthermore, the carbon emissions data generated by resistance loss in S1 include:

[0013] Install current transformers and voltage transformers at key nodes of the transmission line to monitor the current and voltage information in the line in real time. By calculating the product of current and voltage, the power loss of the line is obtained, and then the resistance loss is calculated;

[0014] Carbon emissions from corona losses include:

[0015] Install corona monitoring sensors and corona spectrum sensors at key nodes of transmission lines to monitor the corona discharge around the conductors in real time and obtain data on corona loss;

[0016] The carbon emission data generated by insulator leakage loss includes: installing leakage current sensors on insulators at key nodes of transmission lines, monitoring the leakage current on the surface of the insulators in real time, and obtaining data on insulator leakage loss.

[0017] Furthermore, the key nodes in S3 include: nodes at both ends of the transmission line, nodes in the middle of the transmission line, nodes where terrain changes occur in the transmission line, tower nodes, insulator nodes, and conductor nodes.

[0018] Furthermore, key nodes in S3 also include:

[0019] Frequent fault nodes: Based on the historical operation data of the transmission line, the frequency and location of faults are analyzed using a big data clustering analysis algorithm, and fault-prone areas are selected as key nodes;

[0020] Nodes with large losses: Based on the historical operation data of the transmission line, a big data clustering analysis algorithm is used to identify nodes with large losses in the transmission line, including areas where the resistance loss, corona loss, and insulator leakage loss meet the set thresholds. These areas are selected as key nodes.

[0021] Furthermore, S3 also includes:

[0022] When constructing several binary trees for each key node detection list, the criterion for stopping the construction of the binary tree of the key node is: whether the growth length of the current binary tree is within the termination interval determined in advance based on the maximum growth length. If so, stop the construction; if not, continue the construction. The growth length represents the ratio of the number of training samples selected from the real-time monitoring data of the key node in the current binary tree node expansion on one side of the binary tree node to the number on the other side of the binary tree node.

[0023] Furthermore, the growth length X t Indicates the number of training samples on the left side of the binary tree node selected from the real-time data of the key node detection list during the current binary tree node expansion, Y t When it represents the number of training samples selected from the real-time data of the key node detection list in the current binary tree node expansion on the right side of the binary tree node, the termination interval is [0.5, 1.2].

[0024] Furthermore, the basis for the deep isolation forest algorithm in S3 to determine whether the carbon emission data in the key node detection list is abnormal data is:

[0025] The abnormal score G(m,n) of the key node detection list m satisfies G(m,n)∈[0.9,1]. If so, the key node is an abnormal point and the sampling data of the key node is discarded; L(n) represents the average path length of the n binary trees of the constructed key node detection list point m; g(m) represents the maximum path length of the current measurement point m, and D(h(m)) represents the expectation of g(m).

[0026] Furthermore, the construction of the carbon emission prediction model includes:

[0027] Based on the historical data in the overhead transmission line carbon emission data collection cloud platform, the monitoring time period and monitoring value of each time period of any key monitoring node are obtained;

[0028] The construction of grey improved prediction model based on monitoring values ​​includes:

[0029] Generating new accumulated second original sequence data after performing first-order accumulation based on the first original sequence data;

[0030] Calculating and obtaining the grey derivative corresponding to the second original sequence data;

[0031] Establishing a grey differential equation, and obtaining the grey improved prediction model based on the grey differential equation;

[0032] The specific implementation steps of the grey improved prediction model include:

[0033] The first original sequence data is generated in chronological order using the monitoring values ​​of each time period in the historical data, and is denoted as S0 = {Y0(1), Y0(2), Y0(3), ..., Y0(n)}, where S0 represents the first original sequence data, and Y0(1), Y0(2), Y0(3), ..., Y0(n) represent the monitoring values ​​of 1, 2, 3, ..., n time periods, respectively;

[0034] Perform first-order accumulation generation processing on the basis of the first original sequence data S0:

[0035]

[0036] Wherein, k=1, 2, ..., n; Y1(k) represents the element of the first original sequence data corresponding to the second original sequence data;

[0037] On the basis of the second original sequence data, the grey derivative d(k) of Y1(k) is obtained, and the GM(1,1) grey differential equation is established according to the derivative d(k):

[0038] d(k)+a(Y1(k)+(1-b)Y1(k-1))=c,

[0039] Among them, a is the development grayscale; c is the endogenous control grayscale;

[0040] To solve a and c in the GM(1,1) gray differential equation, let is the vector to be estimated, and a and c are obtained by fitting the first-order linear differential equation using the discrete first original sequence data;

[0041] The output of the gray improved prediction model for the n+1th period is:

[0042]

[0043] Among them, Y0(n+1) is the output of the gray improved prediction model in the n+1th period, and the subsequent predicted monitoring values ​​are output based on the gray improved prediction model;

[0044] For any key node monitoring, predicted sequence data is generated, the safety threshold of each key node monitoring is obtained, and the corresponding time period in the predicted sequence data where the safety threshold is located is marked.

[0045] Furthermore, the construction of the key node processing model in S5 includes:

[0046] S51, constructing a time axis, and entering the generated predicted sequence data into the time axis;

[0047] S52: Obtain all time periods marked by monitoring, and perform safety threshold comparison and calculation on all key nodes within the time periods marked by monitoring;

[0048] S53. After the safety threshold comparison and calculation of all key nodes are completed, the maintenance time of the nodes determined to be non-safety critical nodes is used as the unified maintenance node. At the same time, the unified maintenance key nodes are sorted according to the maintenance time priority level, and the ones with the highest ranking are processed first. Based on the processed data fed back by the unified maintenance key nodes, the 3D digital twin carbon emission monitoring model of the overhead transmission line is updated;

[0049] The process of comparing and calculating the security thresholds of the key nodes in S52 is as follows:

[0050] S521: A method for obtaining a safety threshold function is as follows: historical information of carbon emission data corresponding to risks at key nodes is obtained, and safe data and risky data from the historical carbon emission data are selected based on the physical characteristics of the carbon emission data; values ​​are assigned to the selected data according to their size; if the data is continuous, a continuous value is assigned; if the data is discontinuous, a categorical value is assigned; and the safety threshold function is obtained for the continuous data and the discontinuous data using the Newton interpolation algorithm and the random forest algorithm, respectively.

[0051] S522: If the data is continuous, the safety threshold is determined by calculating the intersection of the safety data and the risky data using the safety threshold function. If the data is discontinuous, the safety threshold is determined by calculating the safety threshold function, i.e., the input and output of the random forest neural network model. Determining the safety threshold using the input and output of the random forest neural network model involves: inputting data corresponding to values ​​of adjacent time periods into the random forest neural network model multiple times, determining the type of data using the output, and obtaining the safety threshold based on the input data corresponding to the values ​​of adjacent time periods and the output.

[0052] S523: Compare the predicted sequence data of the key node with the safety threshold to determine whether the key node is a safety node.

[0053] Furthermore, the updating of the three-dimensional digital twin carbon emission monitoring model of the overhead transmission line in S53 includes:

[0054] Use QGIS open source software to simulate and analyze the overhead transmission line structure model and unify the overhead transmission line structure template;

[0055] Conduct parametric modeling and develop a parametric family library for overhead transmission lines. Based on the parametric family library for each professional parametric overhead transmission line, initialize data for all key nodes in actual deployment, and generate a three-dimensional digital twin carbon emission monitoring model for overhead transmission lines through parametric methods.

[0056] If there is any unsafe key node list feedback processed data, it will be updated to the overhead transmission line three-dimensional digital twin carbon emission monitoring model.

[0057] Compared with the prior art, the embodiments of the present invention achieve the following beneficial effects:

[0058] The embodiment of the present invention provides a cloud platform for collecting carbon emission data of overhead transmission lines. Based on different time periods specified, the cloud platform collects and summarizes the carbon emission data generated by resistance loss, corona loss and insulator leakage loss during the operation of overhead transmission lines in real time; creates several key node detection lists, records each monitoring record of the carbon emission data generated by resistance loss, corona loss and insulator leakage loss in the key node detection list, and records the time period of each monitoring key node; uses a deep isolation forest algorithm to detect abnormal states of carbon emission data in all key node detection lists, and discards abnormal sampling data for preprocessing; builds a carbon emission prediction model, and outputs the carbon emission data of each future time through the monitoring data in the key node detection list. The predicted sequence data of the points are collected, and the time period of the safety threshold is marked based on the safety threshold of each monitoring; a key node processing model is constructed, and the sequence data predicted by the key node detection list is compared with the safety threshold to determine whether the key node is safe, and a unified and maintained non-safe key node list is output based on the marked time period, and the processed data is fed back based on the non-safe key node list to update the three-dimensional digital twin carbon emission monitoring model of the overhead transmission line; the present invention can realize real-time monitoring and early warning of carbon emissions of overhead transmission lines, realize informatization and scientific management of carbon emission monitoring of overhead transmission lines, reduce the update frequency of the digital twin carbon emission monitoring model, and reduce carbon emissions, thereby ensuring that the operation of overhead transmission lines is more stable and energy-saving. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 This is a flow chart of a method for intelligent monitoring and analysis of carbon emissions from overhead transmission lines provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0060] The present invention will be described in detail below with reference to the accompanying drawings.

[0061] Example 1

[0062] The embodiment of the present invention provides a method for intelligent monitoring and analysis of carbon emissions from overhead transmission lines. Figure 1 , including the following steps:

[0063] S1. Build a cloud platform for collecting carbon emission data for overhead transmission lines. This platform collects and summarizes carbon emission data generated by resistance loss, corona loss, and insulator leakage loss during the operation of overhead transmission lines in real time, based on specified time periods.

[0064] S2. Create several key node detection lists, monitor the carbon emission data generated by resistance loss, corona loss, and insulator leakage loss respectively, record each corresponding monitoring data in the key node detection list, and record the time period of each monitoring key node;

[0065] S3. Use the deep isolation forest algorithm to detect abnormal states of carbon emission data in all key node detection lists, and discard abnormal sampling data for preprocessing;

[0066] S4. Build a carbon emission prediction model, output the predicted sequence data for each future time point based on the monitoring data in the key node detection list, and mark the time period where the safety threshold is located based on the safety threshold of each monitoring data;

[0067] S5. Build a key node processing model, compare the sequence data predicted by the key node detection list with the safety threshold, determine whether the key node is safe, output a uniformly maintained list of non-safe key nodes based on the marked time period, and feed back the processed data based on the non-safe key node list to update the three-dimensional digital twin carbon emission monitoring model of the overhead transmission line.

[0068] Specifically, a cloud platform for collecting carbon emission data of overhead transmission lines is constructed. Based on different time periods specified, the carbon emission data generated by resistance loss, corona loss and insulator leakage loss during the operation of overhead transmission lines are collected and summarized in real time; several key node detection lists are created, and each monitoring record of the carbon emission data generated by resistance loss, corona loss and insulator leakage loss is recorded in the key node detection list, and the time period of each monitoring key node is recorded; the deep isolation forest algorithm is used to detect the abnormal state of the carbon emission data in all key node detection lists, and the abnormal sampling data is discarded for preprocessing; a carbon emission prediction model is constructed, and the predicted state at each future time point is output through the monitoring data in the key node detection list. The measured sequence data is used to mark the time period of the safety threshold based on the safety threshold of each monitoring; a key node processing model is constructed to compare the sequence data predicted by the key node detection list with the safety threshold to determine whether the key node is safe, and a unified maintained non-safe key node list is output based on the marked time period, and the processed data is fed back based on the non-safe key node list to update the three-dimensional digital twin carbon emission monitoring model of the overhead transmission line; this embodiment can realize real-time monitoring and early warning of carbon emissions of overhead transmission lines, realize informatization and scientific management of carbon emission monitoring of overhead transmission lines, reduce the update frequency of the digital twin carbon emission monitoring model, and reduce carbon emissions, thereby ensuring that the operation of overhead transmission lines is more stable and energy-saving.

[0069] In the above embodiment, specifically, the carbon emissions of overhead transmission lines mainly include carbon emissions in the construction and operation stages. This embodiment mainly studies the monitoring of carbon emissions in the operation stage, and reduces the intensive monitoring of all points in the entire overhead transmission line by collecting data from a number of key points. The key points are selected by a clustering algorithm, and the clustering algorithm adopts one or more combinations including but not limited to the k-means clustering algorithm, the K-prototype clustering algorithm, the DBSCAN algorithm, and the OPTICS algorithm. This embodiment uses the k-means clustering algorithm to select key nodes, and reduces the load of the digital twin carbon emission monitoring model through filtering of the selected key nodes.

[0070] Specifically, the K-prototype clustering algorithm is used to classify the key points and divide the data into two categories: main classification data and other classification data; the main classification data is retained and used as the key point, and the other classification data is removed; after the data is classified by the clustering algorithm, the classification with the most dense distribution and the highest score after classification is used as the main classification data, and all other classifications are used as other classification data;

[0071] The K-prototype clustering algorithm performs clustering on the sample set D = {y1, y2, ...yn The classification method is: by continuously updating the center object, the cluster division C = {C1, C2, ...C k}, C1, C2, C k Each represents a single cluster, where each cluster includes multiple single samples in the same cluster; the clustering rule is to minimize the mean square error, and the mean square error formula is: Among them, y represents being classified into cluster C i The sample data in , δ represents the Hamming distance, w i It is cluster C i The mean vector, P d The smaller the value of , the higher the similarity of samples y within the cluster.

[0072] In the above embodiment, specifically, the carbon emission data generated by the resistance loss in S1 includes:

[0073] Install current transformers and voltage transformers at key nodes of the transmission line to monitor the current and voltage information in the line in real time. By calculating the product of current and voltage, the power loss of the line is obtained, and then the resistance loss is calculated;

[0074] Carbon emissions from corona losses include:

[0075] Install corona monitoring sensors and corona spectrum sensors at key nodes of transmission lines to monitor the corona discharge around the conductors in real time and obtain data on corona loss;

[0076] The carbon emission data generated by insulator leakage loss includes: installing leakage current sensors on insulators at key nodes of transmission lines, monitoring the leakage current on the surface of the insulators in real time, and obtaining data on insulator leakage loss.

[0077] Specifically, carbon emissions at key nodes are calculated using the following formula:

[0078]

[0079] Among them, C Ri represents the key quantity of carbon emission, which corresponds to the key quantity of corona loss, the key quantity of resistance loss, and the key quantity of insulator leakage loss, respectively. Yi represents the additional impact coefficient of carbon emissions, which corresponds to the additional impact coefficient of corona loss, the additional impact coefficient of resistance loss, and the additional impact coefficient of insulator leakage loss;

[0080] The carbon emissions generated by corona loss are calculated according to the following formula:

[0081]

[0082] C Y1=0.02C R ,

[0083] Where: n is the number of loops; I is the phase current; N is the number of split phase conductors; R is the AC resistance of the conductor; τ is the maximum load loss hours; L is the line length, F d is the carbon emission factor for electricity;

[0084] The carbon emissions generated by resistance loss are calculated using the following formula:

[0085]

[0086] C Y2 =0.03C R2 ,

[0087] Where: ρ represents the resistivity of the conductor material, A represents the cross-sectional area of ​​the conductor;

[0088] The carbon emissions generated by insulator leakage loss are calculated using the following formula:

[0089] C R3 =D 2 CτLF d *0.003,

[0090] C Y3 =0.04C R3 ,

[0091] Where: D represents the insulator leakage current, and C represents the insulator leakage resistance.

[0092] Corona loss is a function of conductor geometry, conductor electric field strength, voltage, and meteorological conditions in the area where the line is located. Corona loss does not change with changes in transmission capacity. Its magnitude is mainly related to voltage level, conductor structure, and weather conditions.

[0093] The measurement value of each node is collected three times, and the average value of the three times is taken as the measurement value.

[0094] In the above embodiment, specifically, the key nodes in S3 include: nodes at both ends of the transmission line, nodes in the middle of the transmission line, nodes where terrain changes occur in the transmission line, tower nodes, insulator nodes, and conductor nodes.

[0095] In the above embodiment, specifically, the key nodes in S3 also include:

[0096] Frequent fault nodes: Based on the historical operation data of the transmission line, the frequency and location of faults are analyzed using a big data clustering analysis algorithm, and fault-prone areas are selected as key nodes;

[0097] Nodes with large losses: Based on the historical operation data of the transmission line, a big data clustering analysis algorithm is used to identify nodes with large losses in the transmission line, including areas where the resistance loss, corona loss, and insulator leakage loss meet the set thresholds. These areas are selected as key nodes.

[0098] In the above embodiment, specifically, S3 further includes:

[0099] When constructing several binary trees for each key node detection list, the criterion for stopping the construction of the binary tree of the key node is: whether the growth length of the current binary tree is within the termination interval determined in advance based on the maximum growth length. If so, stop the construction; if not, continue the construction. The growth length represents the ratio of the number of training samples selected from the real-time monitoring data of the key node in the current binary tree node expansion on one side of the binary tree node to the number on the other side of the binary tree node.

[0100] In the above embodiment, specifically, the growth length X t Indicates the number of training samples on the left side of the binary tree node selected from the real-time data of the key node detection list during the current binary tree node expansion, Y t When it represents the number of training samples selected from the real-time data of the key node detection list in the current binary tree node expansion on the right side of the binary tree node, the termination interval is [0.5, 1.2].

[0101] In the above embodiment, specifically, the basis for the deep isolation forest algorithm in S3 to determine whether the carbon emission data in the key node detection list is abnormal data is:

[0102] The abnormal score G(m,n) of the key node detection list m satisfies G(m,n)∈[0.9,1]. If so, the key node is an abnormal point and the sampling data of the key node is discarded; L(n) represents the average path length of the n binary trees of the constructed key node detection list point m; g(m) represents the maximum path length of the current measurement point m, and D(h(m)) represents the expectation of g(m).

[0103] Specifically, in the collected data, if the data collected at the key point is determined to be abnormal data, the abnormal data will be discarded. If the number of consecutive abnormalities collected at the key point exceeds the preset threshold range, the key point will be determined to be a key monitoring point.

[0104] In the above embodiment, specifically, the step of constructing a carbon emission prediction model includes:

[0105] Based on the historical data in the overhead transmission line carbon emission data collection cloud platform, the monitoring time period and monitoring value of each time period of any key monitoring node are obtained;

[0106] The construction of grey improved prediction model based on monitoring values ​​includes:

[0107] Generating new accumulated second original sequence data after performing first-order accumulation based on the first original sequence data;

[0108] Calculating and obtaining the grey derivative corresponding to the second original sequence data;

[0109] Establishing a grey differential equation, and obtaining the grey improved prediction model based on the grey differential equation;

[0110] The specific implementation steps of the grey improved prediction model include:

[0111] The first original sequence data is generated in chronological order using the monitoring values ​​of each time period in the historical data, and is denoted as S0 = {Y0(1), Y0(2), Y0(3), ..., Y0(n)}, where S0 represents the first original sequence data, and Y0(1), Y0(2), Y0(3), ..., Y0(n) represent the monitoring values ​​of 1, 2, 3, ..., n time periods, respectively;

[0112] Perform first-order accumulation generation processing on the basis of the first original sequence data S0:

[0113]

[0114] Wherein, k=1, 2, ..., n; Y1(k) represents the element of the first original sequence data corresponding to the second original sequence data;

[0115] On the basis of the second original sequence data, the grey derivative d(k) of Y1(k) is obtained, and the GM(1,1) grey differential equation is established according to the derivative d(k):

[0116] d(k)+a(Y1(k)+(1-b)Y1(k-1))=c,

[0117] Among them, a is the development grayscale; c is the endogenous control grayscale;

[0118] To solve a and c in the GM(1,1) gray differential equation, let is the vector to be estimated, and a and c are obtained by fitting the first-order linear differential equation using the discrete first original sequence data;

[0119] The output of the gray improved prediction model for the n+1th period is:

[0120]

[0121] Among them, Y0(n+1) is the output of the gray improved prediction model in the n+1th period, and the subsequent predicted monitoring values ​​are output based on the gray improved prediction model;

[0122] For any key node monitoring, predicted sequence data is generated, the safety threshold of each key node monitoring is obtained, and the corresponding time period in the predicted sequence data where the safety threshold is located is marked.

[0123] Specifically, the data detected at each key node are resistance loss carbon emissions, corona loss carbon emissions and insulator leakage carbon emissions, corresponding to three safety thresholds: resistance loss carbon emission safety threshold, corona loss carbon emission safety threshold and insulator leakage carbon emission safety threshold.

[0124] Furthermore, the construction of the key node processing model in S5 includes:

[0125] S51, constructing a time axis, and entering the generated predicted sequence data into the time axis;

[0126] S52: Obtain all time periods marked by monitoring, and perform safety threshold comparison and calculation on all key nodes within the time periods marked by monitoring;

[0127] S53. After the safety threshold comparison and calculation of all key nodes are completed, the maintenance time of the nodes determined to be non-safety critical nodes is used as the unified maintenance node. At the same time, the unified maintenance key nodes are sorted according to the maintenance time priority level, and the ones with the highest ranking are processed first. Based on the processed data fed back by the unified maintenance key nodes, the 3D digital twin carbon emission monitoring model of the overhead transmission line is updated;

[0128] The process of comparing and calculating the security thresholds of the key nodes in S52 is as follows:

[0129] S521: A method for obtaining a safety threshold function is as follows: historical information of carbon emission data corresponding to risks at key nodes is obtained, and safe data and risky data from the historical carbon emission data are selected based on the physical characteristics of the carbon emission data; values ​​are assigned to the selected data according to their size; if the data is continuous, a continuous value is assigned; if the data is discontinuous, a categorical value is assigned; and the safety threshold function is obtained for the continuous data and the discontinuous data using the Newton interpolation algorithm and the random forest algorithm, respectively.

[0130] S522: If the data is continuous, the safety threshold is determined by calculating the intersection of the safety data and the risky data using the safety threshold function. If the data is discontinuous, the safety threshold is determined by calculating the safety threshold function, i.e., the input and output of the random forest neural network model. Determining the safety threshold using the input and output of the random forest neural network model involves: inputting data corresponding to values ​​of adjacent time periods into the random forest neural network model multiple times, determining the type of data using the output, and obtaining the safety threshold based on the input data corresponding to the values ​​of adjacent time periods and the output.

[0131] S523: Compare the predicted sequence data of the key node with the safety threshold to determine whether the key node is a safety node.

[0132] In the above embodiment, specifically, the updating of the overhead transmission line three-dimensional digital twin carbon emission monitoring model in S53 includes:

[0133] Use QGIS open source software to simulate and analyze the overhead transmission line structure model and unify the overhead transmission line structure template;

[0134] Conduct parametric modeling and develop a parametric family library for overhead transmission lines. Based on the parametric family library for each professional parametric overhead transmission line, initialize data for all key nodes in actual deployment, and generate a three-dimensional digital twin carbon emission monitoring model for overhead transmission lines through parametric methods.

[0135] If there is any unsafe key node list feedback processed data, it will be updated to the overhead transmission line three-dimensional digital twin carbon emission monitoring model.

[0136] Specifically, in the monitoring display of the three-dimensional digital twin carbon emission monitoring model of the overhead transmission line, the key node is marked in red as an unsafe carbon emission state, and the key node is marked in green as a safe carbon emission state. If the key node is converted to a safe state through safety maintenance measures under an abnormal state, the key node will be marked green.

[0137] The algorithm and display provided herein are not inherently related to any particular computer, virtual system or other device. Various general-purpose systems can also be used together with the teachings based on this. According to the above description, it is obvious that the structure required for constructing this type of system. In addition, the present invention is not directed to any specific programming language. It should be understood that various programming languages ​​can be utilized to realize the content of the present invention described herein, and the above description of specific languages ​​is for the purpose of disclosing the best mode of the present invention.

[0138] In the description provided herein, numerous specific details are described. However, it is understood that embodiments of the present invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques are not shown in detail so as not to obscure the understanding of this description.

[0139] Similarly, it should be understood that in order to streamline the present disclosure and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, this disclosed method should not be interpreted as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as reflected in the claims below, inventive aspects lie in less than all the features of the individual embodiments disclosed above. Accordingly, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of the invention.

[0140] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and in addition may be divided into multiple submodules or subunits or subcomponents. All features disclosed in this specification (including the accompanying claims, abstracts and drawings) and all processes or units of any method or device disclosed herein may be combined in any combination, except that at least some of such features and / or processes or units are mutually exclusive. Unless expressly stated otherwise, each feature disclosed in this specification (including the accompanying claims, abstracts and drawings) may be replaced by an alternative feature providing the same, equivalent or similar purpose.

[0141] Furthermore, those skilled in the art will appreciate that although some embodiments herein include certain features included in other embodiments but not other features, combinations of features from different embodiments are intended to be within the scope of the present invention and to form different embodiments. For example, in the claims below, any of the claimed embodiments may be used in any combination.

[0142] The various component embodiments of the present invention can be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. It should be understood by those skilled in the art that a microprocessor or digital signal processor (DSP) can be used in practice to implement some or all of the functions of some or all of the components in the apparatus according to an embodiment of the present invention. The present invention can also be implemented as a device or device program (e.g., a computer program and a computer program product) for executing a part or all of the methods described herein. Such a program implementing the present invention can be stored on a computer-readable medium, or can have the form of one or more signals. Such a signal can be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

Claims

1. A method for intelligent monitoring and analysis of carbon emissions from overhead transmission lines, characterized in that: The following steps are involved: S1. Build a cloud platform for collecting carbon emission data for overhead transmission lines. This platform collects and summarizes carbon emission data generated by resistance loss, corona loss, and insulator leakage loss during the operation of overhead transmission lines in real time, based on specified time periods. S2. Create several key node detection lists, monitor the carbon emission data generated by resistance loss, corona loss, and insulator leakage loss respectively, record each corresponding monitoring data in the key node detection list, and record the time period of each monitoring key node; S3. Use the deep isolation forest algorithm to detect abnormal states of carbon emission data in all key node detection lists, and discard abnormal sampling data for preprocessing; S4. Build a carbon emission prediction model, output the predicted sequence data for each future time point based on the monitoring data in the key node detection list, and mark the time period where the safety threshold is located based on the safety threshold of each monitoring data; S5. Build a key node processing model, compare the sequence data predicted by the key node detection list with the safety threshold, determine whether the key node is safe, output a uniformly maintained list of non-safe key nodes based on the marked time period, and feed back the processed data based on the non-safe key node list to update the 3D digital twin carbon emission monitoring model of the overhead transmission line; The basis for the deep isolation forest algorithm in S3 to determine whether the carbon emission data in the key node detection list is abnormal data is: The abnormal score G(m,n) of the key node detection list m satisfies G(m,n)∈[0.9,1]. If so, the key node is an abnormal point and the sampling data of the key node is discarded; , L(n) represents the average path length of the n binary trees of the constructed key node detection list point m; g(m) represents the maximum path length of the current measurement point m, and D(h(m)) represents the expectation of g(m).

2. The method for intelligent monitoring and analysis of carbon emissions from overhead transmission lines according to claim 1, characterized in that: The carbon emissions generated by resistance losses in S1 include: Install current transformers and voltage transformers at key nodes of the transmission line to monitor the current and voltage information in the line in real time. By calculating the product of current and voltage, the power loss of the line is obtained, and then the resistance loss is calculated; Carbon emissions from corona losses include: Install corona monitoring sensors and corona spectrum sensors at key nodes of transmission lines to monitor the corona discharge around the conductors in real time and obtain data on corona loss; The carbon emission data generated by insulator leakage loss includes: installing leakage current sensors on insulators at key nodes of transmission lines, monitoring the leakage current on the surface of the insulators in real time, and obtaining data on insulator leakage loss.

3. The method for intelligent monitoring and analysis of carbon emissions from overhead transmission lines according to claim 1, characterized in that: The key nodes in S3 include: nodes at both ends of the transmission line, nodes in the middle of the transmission line, nodes where terrain changes occur in the transmission line, tower nodes, insulator nodes, and conductor nodes.

4. The method for intelligent monitoring and analysis of carbon emissions from overhead transmission lines according to claim 1, characterized in that: Key nodes in S3 also include: Frequent fault nodes: Based on the historical operation data of the transmission line, the frequency and location of faults are analyzed using a big data clustering analysis algorithm, and the fault-prone areas are selected as key nodes; Nodes with large losses: Based on the historical operation data of the transmission line, a big data clustering analysis algorithm is used to identify nodes with large losses in the transmission line, including areas where the resistance loss, corona loss, and insulator leakage loss meet the set thresholds. These areas are selected as key nodes.

5. The method for intelligent monitoring and analysis of carbon emissions from overhead transmission lines according to claim 1, characterized in that: S3 also includes: When constructing several binary trees for each key node detection list, the criterion for stopping the construction of the binary tree of the key node is: whether the growth length of the current binary tree is within the termination interval determined in advance based on the maximum growth length. If so, stop the construction; if not, continue the construction. The growth length represents the ratio of the number of training samples selected from the real-time monitoring data of the key node in the current binary tree node expansion on one side of the binary tree node to the number on the other side of the binary tree node.

6. The method for intelligent monitoring and analysis of carbon emissions from overhead transmission lines according to claim 5, characterized in that: The growth length a= , Indicates the number of training samples on the left side of the binary tree node selected from the real-time data of the key node detection list during the current binary tree node expansion. When it represents the number of training samples selected from the real-time data of the key node detection list in the current binary tree node expansion on the right side of the binary tree node, the termination interval is [0.5, 1.2].

7. The method for intelligent monitoring and analysis of carbon emissions from overhead transmission lines according to claim 1, characterized in that: The carbon emission prediction model is constructed as follows: Based on the historical data in the overhead transmission line carbon emission data collection cloud platform, the monitoring time period and monitoring value of each time period of any key monitoring node are obtained; The construction of grey improved prediction model based on monitoring values ​​includes: Generating second original sequence data after performing first-order accumulation based on the first original sequence data; Calculating and obtaining the grey derivative corresponding to the second original sequence data; Establishing a grey differential equation, and obtaining the grey improved prediction model based on the grey differential equation; The specific implementation steps of the grey improved prediction model include: The first original sequence data is generated in chronological order based on the monitoring values ​​of each time period in the historical data, and is recorded as ={ (1), (2), (3), ..., (n)}, where represents the first original sequence data, (1), (2), (3),…, (n) represents the monitoring values ​​of 1, 2, 3, ..., n time periods respectively; In the first raw sequence data Based on this, we perform first-order accumulation generation processing: ， Where, k = 1, 2, ..., n; (k) represents an element of the first original sequence data corresponding to the second original sequence data; Based on the second original sequence data, obtain (k), and establish the GM(1,1) grey differential equation based on the derivative d(k): ， Among them, a is the development grayscale; c is the endogenous control grayscale; Solve a and c in the gray differential equation GM(1,1), let is the vector to be estimated, and a and c are obtained by fitting the first-order linear differential equation using the discrete first original sequence data; The output of the gray improved prediction model for the n+1th period is: ， in, (n+1) is the output of the grey improved prediction model in the n+1th period, and the subsequent prediction monitoring values ​​are output based on the grey improved prediction model; For any key node monitoring, predicted sequence data is generated, the safety threshold of each key node monitoring is obtained, and the corresponding time period in the predicted sequence data where the safety threshold is located is marked.

8. The method for intelligent monitoring and analysis of carbon emissions from overhead transmission lines according to claim 1, characterized in that: The construction of the key node processing model described in S5 includes: S51, constructing a time axis, and entering the generated predicted sequence data into the time axis; S52: Obtain all time periods marked by monitoring, and perform safety threshold comparison and calculation on all key nodes within the time periods marked by monitoring; S53. After the safety threshold comparison and calculation of all key nodes are completed, the maintenance time of the nodes determined to be non-safety critical nodes is used as the unified maintenance node. At the same time, the unified maintenance key nodes are sorted according to the maintenance time priority level, and the ones with the highest ranking are processed first. Based on the processed data fed back by the unified maintenance key nodes, the 3D digital twin carbon emission monitoring model of the overhead transmission line is updated; The process of comparing and calculating the security thresholds of the key nodes in S52 is as follows: S521: A method for obtaining a safety threshold function is as follows: historical information of carbon emission data corresponding to risks at key nodes is obtained, and safe data and risky data from the historical carbon emission data are selected based on the physical characteristics of the carbon emission data; values ​​are assigned to the selected data according to their size; if the data is continuous, a continuous value is assigned; if the data is discontinuous, a categorical value is assigned; and the safety threshold function is obtained for the continuous data and the discontinuous data using the Newton interpolation algorithm and the random forest algorithm, respectively. S522: If the data is continuous, the safety threshold is determined by calculating the intersection of the safety data and the risky data using the safety threshold function. If the data is discontinuous, the safety threshold is determined by calculating the safety threshold function, i.e., the input and output of the random forest neural network model. Determining the safety threshold using the input and output of the random forest neural network model involves: inputting data corresponding to values ​​of adjacent time periods into the random forest neural network model multiple times, determining the type of data using the output, and obtaining the safety threshold based on the input data corresponding to the values ​​of adjacent time periods and the output. S523: Compare the predicted sequence data of the key node with the safety threshold to determine whether the key node is a safety node.

9. The method for intelligent monitoring and analysis of carbon emissions from overhead transmission lines according to claim 8, characterized in that: The update to the three-dimensional digital twin carbon emission monitoring model of overhead transmission lines described in S53 includes: Use QGIS open source software to simulate and analyze the overhead transmission line structure model and unify the overhead transmission line structure template; Conduct parametric modeling and develop a parametric family library for overhead transmission lines. Based on the parametric family library for each professional parametric overhead transmission line, initialize data for all key nodes in actual deployment, and generate a three-dimensional digital twin carbon emission monitoring model for overhead transmission lines through parametric methods. If there is any unsafe key node list feedback processed data, it will be updated to the overhead transmission line three-dimensional digital twin carbon emission monitoring model.

Images