Power Data Big Model Construction System and Method Applied to Low-Carbon Power Grid

Abstract

The present invention relates to the technical field of power data management, specifically a power data large model construction system and method applied to a low-carbon power grid. The present invention identifies and extracts by sorting out the power control structure actually generated by the corresponding permission management account in the power data management platform; calculates and evaluates the characteristic operation index for each power control structure to measure the impact of the characteristic warning information in the corresponding power control structure on the target power grid, and by sorting out the change situation of the characteristic operation index, screens out the power control structure with a lag in processing for at least one historical warning data reception record, and at the same time identifies and extracts the characteristic warning information that will cause the situation to become complicated and the corresponding processing situation to become complicated due to the lag in control adjustment, so as to realize intelligent warning prompts for the management terminal to control the target power grid.

Description

Technical Field

[0001] The present invention relates to the technical field of power data management, and specifically to a power data large model construction system and method applied to a low-carbon power grid. Background Art

[0002] At present, the power data management platform constructed for the low-carbon power grid often has the functions of remotely analyzing the alarm data of substation equipment collected from the power grid and remotely controlling the relevant power hidden danger investigation of the power grid, making it more convenient to remotely monitor the state of the power grid and implement the corresponding power management control process at the same time. In this process, the process data of power management can also be directly processed and analyzed.

[0003] As a system that can achieve remote control and management, the necessity of timely processing some warning data that will complicate the situation and the corresponding processing process lies in ensuring the effectiveness and practicability of most of this warning data, minimizing the losses brought to the power grid and the management burden brought to the corresponding authorized users. Summary of the Invention

[0004] The purpose of the present invention is to provide a power data large model construction system and method applied to a low-carbon power grid to solve the problems raised in the above background art.

[0005] To solve the above technical problems, the present invention provides the following technical solution: A power data large model construction method applied to a low-carbon power grid, the method comprising:

[0006] Step S1: Construct a power data management platform containing a number of power control models, establish a remote communication connection between the power data management platform and the target power grid, and remotely receive the warning data transmitted by the target power grid in real time; the permission management account remotely controls the target power grid through the power data management platform based on the warning data;

[0007] Step S2: Regularly extract the historical warning data reception records from the historical operation logs of the power data management platform and all historical operation records generated by the corresponding permission management account in the power data management platform;

[0008] Step S3: Sort out the time feature distribution between the historical operation process data generated by the permission management account to remotely control the target power grid through the power data management platform and the historical warning data received by the power data management platform, judge and identify the historical warning data reception records and historical operation records with corresponding connection relationships, and construct and extract a power control structure;

[0009] Step S4: Sort out the characteristic distribution presented by the corresponding historical operation records in each power control structure, and calculate the evaluation characteristic operation index for each power control structure; classify each power control structure according to the quantity distribution presented by the corresponding historical warning data reception records in each power control structure;

[0010] Step S5: Evaluate and calculate the theoretical characteristic operation index for the corresponding power control structure, and screen out abnormal control structures according to the deviation between the characteristic operation index and the theoretical characteristic operation index of the corresponding power control structure;

[0011] Step S6: From the abnormal control structures, identify and extract the warning data that will cause control complexity due to the lag of control adjustment operations, construct and generate a power data processing warning library, and judge whether to send a warning prompt for timely processing to the management terminal based on the power data processing warning library for the newly added warning data reception records.

[0012] Preferably, Step S3 includes:

[0013] Step S3-1: Sort all the historical warning data reception records in the order of the record generation time to obtain a reception record sequence; in the reception record sequence, for each adjacent pair of historical warning data reception records in turn, extract the record generation time T 1 、T 2 ,wherein, T 1 <T 2 ,to obtain a number of characteristic periods [T 1 ,T 2 ;

[0014] Step S3-2: Traverse the record generation time of each historical operation record, and respectively collect the historical operation records within each characteristic period to obtain an operation record set corresponding to each characteristic period; traverse the total number of historical operation records included in the operation record sets in each characteristic period in turn, and set the operation record sets with the total number of included historical operation records greater than or equal to 1 as the target operation record sets; number each target operation record set in order; here, the order refers to the position order of the historical warning data reception records for extracting the corresponding target operation record sets in the above reception record sequence;

[0015] Step S3-3: When the operation record set corresponding to the characteristic period between the j-th historical warning data reception record and the (j + 1)-th historical warning data reception record is a target operation record set, judge that there is a division node between the j-th historical warning data reception record and the (j + 1)-th historical warning data reception record in the reception record sequence;

[0016] Step S3-4: Identify all partitioning nodes existing in the received record sequence, divide the received record sequence into several target subsequences based on each partitioning node; number the several target subsequences in sequence; here, the sequence refers to the position sequence of each received record subsequence in the original above-mentioned received record sequence, and establish a corresponding connection relationship between the target subsequences with the same corresponding number and the target operation record set, and construct and generate several power control structures {target subsequence, target operation record set}.

[0017] Preferably, step S4 includes:

[0018] Step S4-1: Extract the target operation record sets in each corresponding power control structure respectively, obtain the total number M of historical operation records included in each target operation record set, capture the total number K of privilege management accounts that capture M historical operation records, and capture the highest privilege level C in the K privilege management accounts max and the lowest privilege level C min , calculate the level span P = |C max -C min | for the target operation record sets in each corresponding power control structure, and calculate the characteristic operation index β = (M × K) for the target operation record sets in each corresponding power control structure P , where M ≥ 1, K ≥ 1, P ≥ 0;

[0019] Step S4-2: Traverse the total number N of historical warning data reception records included in the corresponding target subsequences in each power control structure; set the power control structure with the corresponding total number N = 1 as the first characteristic power control structure, and set the power control structure with the corresponding total number N ≥ 2 as the second characteristic power control structure.

[0020] Preferably, step S5 includes:

[0021] Step S5-1: Extract the warning data received by the power data management platform from each historical warning data reception record, perform feature extraction on the warning data to obtain a set of characteristic warning information corresponding to each historical warning data reception record; traverse the characteristic operation index of the corresponding target operation record set in each power control structure and the set of characteristic warning information corresponding to each historical warning data reception record in the corresponding target subsequence; among them, the number of sets of characteristic warning information extracted for each first characteristic power control structure is 1, and the number of sets of characteristic warning information extracted for each second characteristic power control structure is greater than 1;

[0022] Step S5-2: Assume that in a second characteristic power control structure, the characteristic operation index of the corresponding target operation record set is F, and the corresponding target subsequence is D; collect the set of characteristic warning information extracted from all first characteristic power control structures to obtain set W; sequentially in set W, capture the set of characteristic warning information that has the highest similarity with the sets of characteristic warning information corresponding to each historical warning data reception record in the target subsequence D, and at the same time identify the first characteristic power control structure corresponding to the set of characteristic warning information with the highest similarity.

[0023] Step S5-3: Assume that in set W, the set of characteristic warning information g that has the highest similarity with the set of characteristic warning information d corresponding to the i-th historical warning data reception record in the target subsequence D is captured. i The similarity between the set of characteristic warning information g and the set of characteristic warning information d is β, and the characteristic operation index of the first characteristic power control structure corresponding to the set of characteristic warning information g is Y. i where 0 ≤ β ≤ 1. g

[0024] Step S5-4: Evaluate the characteristic operation value f contributed by the i-th historical warning data reception record in the target subsequence D to a second characteristic power control structure. i = β × Y g Accumulate the characteristic operation values contributed by all historical warning data reception records in the target subsequence D to a second characteristic power control structure to obtain the theoretical characteristic operation index F'. When F' - F > η, determine that a second characteristic power control structure is an abnormal control structure, where η is a threshold and η > 0.

[0025] Generally, whenever the target power grid transmits a warning data, it is necessary to timely implement corresponding control adjustments to the target power grid through the power data management platform. For some warning data that will complicate the situation and the process of corresponding handling situations, it is necessary; and in each second characteristic power control structure, the number of historical warning data reception records is greater than or equal to 2, which means that there is at least a lag in control adjustment for one historical warning data reception record. When a second characteristic power control structure satisfies F' - F > η, it means that there is characteristic warning information in the set of characteristic warning information corresponding to this second characteristic power control structure that will complicate the situation and the process of corresponding handling situations due to the lag in control adjustment.

[0026] Preferably, step S6 includes:

[0027] ​Step S6-1: For each abnormal control structure, collect the warning data received by the power data management platform in each historical warning data reception record included in the corresponding target subsequence to obtain a warning data set S; perform feature extraction on the warning data set S to obtain a set of characteristic warning information R corresponding to each abnormal control structure;

[0028] Step S6-2: Extract the set of characteristic warning information corresponding to each historical warning data reception record in the target subsequence. When the set of characteristic warning information S corresponding to a certain historical warning data reception record and the set of characteristic warning information R satisfy: S - S ∩ R = U ≠ ∅, extract the characteristic warning information included in the set U as the target characteristic warning information;

[0029] Step S6-3: Collect all the target characteristic warning information extracted from each abnormal control structure to construct and generate a power data processing warning library; whenever a new warning data reception record is detected in the power data management platform, and after performing feature extraction on the newly received warning data in the current power data management platform, if it is captured that there is characteristic warning information belonging to the power data processing warning library, send a warning prompt for timely processing to the management terminal.

[0030] To better implement the above method, a power data large model construction system is also proposed. The system includes a power data management platform construction module, a historical data management module, a power control structure construction and management module, a power control structure data management module, an abnormal control structure screening module, and a warning prompt management module;

[0031] The power data management platform construction module is used to construct a power data management platform containing several power control models, establish a remote communication connection between the power data management platform and the target power grid, remotely receive the warning data transmitted by the target power grid in real time, and the permission management account remotely controls the target power grid through the power data management platform based on the warning data;

[0032] The historical data management module is used to regularly extract historical warning data reception records and all historical operation records generated by the corresponding permission management account in the power data management platform from the historical operation logs of the power data management platform;

[0033] The power control structure construction and management module is used to sort out the time feature distribution between the historical operation process data generated by the permission management account to remotely control the target power grid through the power data management platform and the historical warning data received by the power data management platform, judge and identify the historical warning data reception records and historical operation records with corresponding connection relationships, and construct and extract the power control structure;

[0034] The power control structure data management module is used to sort out the characteristic distribution presented by the corresponding historical operation records in each power control structure, and calculate and evaluate the characteristic operation index for each power control structure; classify each power control structure according to the quantity distribution presented by the corresponding historical early warning data reception records in each power control structure.

[0035] The abnormal control structure screening module is used to evaluate and calculate the theoretical characteristic operation index of the corresponding power control structure, and screen out the abnormal control structure according to the deviation between the characteristic operation index and the theoretical characteristic operation index of the corresponding power control structure.

[0036] The early warning prompt management module is used to identify and extract the early warning data that will cause the control to become complicated due to the lag of the control adjustment operation from the abnormal control structure, construct and generate a power data processing early warning library, and judge whether to send an early warning prompt for timely processing to the management terminal based on the power data processing early warning library for the newly added early warning data reception record.

[0037] Preferably, the power control structure construction management module includes a data distribution sorting unit and a structure construction management unit.

[0038] The data distribution sorting unit is used to sort out the time characteristic distribution between the historical operation process data generated by the authority management account to control the target power grid through the power data management platform and the historical early warning data received by the power data management platform.

[0039] The structure construction management unit is used to judge and identify the historical early warning data reception records and historical operation records with corresponding connection relationships, and construct and extract the power control structure.

[0040] Preferably, the power control structure data management module includes a characteristic operation index evaluation calculation unit and a classification management unit.

[0041] The characteristic operation index evaluation calculation unit is used to sort out the characteristic distribution presented by the corresponding historical operation records in each power control structure, and calculate and evaluate the characteristic operation index for each power control structure.

[0042] The classification management unit is used to classify each power control structure according to the quantity distribution presented by the corresponding historical early warning data reception records in each power control structure.

[0043] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: By sorting out the historical operation process data generated by the permission management account for controlling the target power grid based on the power data management platform, and the time feature distribution presented between the power data management platform receiving historical warning data, the power control structure actually generated by the corresponding permission management account in the power data management platform is identified and extracted; by calculating and evaluating the characteristic operation index for each power control structure, the impact of the characteristic warning information in the corresponding power control structure on the target power grid is measured. By sorting out the changes in the characteristic operation index, the power control structures with lagged processing phenomena for at least one historical warning data reception record are screened out, and at the same time, the characteristic warning information that will cause the situation to become complicated and the process of the corresponding processing situation to become complicated due to the lag of the control adjustment is identified and extracted, so as to realize intelligent warning prompts for the management terminal for controlling the target power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The accompanying drawings are used to provide a further understanding of the present invention, and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention, and do not constitute a limitation to the present invention. In the accompanying drawings:

[0045] Figure 1 is a schematic flow chart of a method for constructing a large power data model applied to a low-carbon power grid of the present invention;

[0046] Figure 2 is a schematic structural diagram of a system for constructing a large power data model applied to a low-carbon power grid of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0047] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0048] Please refer to Figure 1 - Figure 2 , the present invention provides a technical solution: A method for constructing a large power data model applied to a low-carbon power grid, the method includes:

[0049] Step S1: Construct a power data management platform containing several power control models, establish a remote communication connection between the power data management platform and the target power grid, and remotely receive the warning data transmitted by the target power grid in real time; the permission management account realizes remote control of the target power grid through the power data management platform based on the warning data;

[0050] Step S2: Regularly extract historical warning data reception records from the historical operation logs of the power data management platform, as well as all historical operation records generated by the corresponding permission management accounts within the power data management platform;

[0051] Step S3: Sort out the time feature distribution between the historical operation process data generated by the permission management account to control the target power grid through the power data management platform and the historical warning data received by the power data management platform, judge and identify the historical warning data reception records and historical operation records with corresponding connection relationships, and construct and extract the power control structure;

[0052] Among them, Step S3 includes:

[0053] Step S3-1: Sort all historical warning data reception records in the order of record generation time to obtain a reception record sequence; in the reception record sequence, for each adjacent pair of historical warning data reception records in turn, extract the record generation time T 1 、T 2 ,where T 1 <T 2 ,to obtain several characteristic periods [T 1 , T 2 ;

[0054] Step S3-2: Traverse the record generation time of each historical operation record, and respectively collect the historical operation records within each characteristic period to obtain an operation record set corresponding to each characteristic period; traverse in turn the total number of historical operation records included in the operation record sets of each characteristic period, and set the operation record sets with the total number of historical operation records included greater than or equal to 1 as the target operation record sets; number the target operation record sets in order;

[0055] Step S3-3: When the operation record set corresponding to the characteristic period between the j-th historical warning data reception record and the (j + 1)-th historical warning data reception record is the target operation record set, judge that there is a division node between the j-th historical warning data reception record and the (j + 1)-th historical warning data reception record in the reception record sequence;

[0056] Step S3-4: Identify all division nodes existing in the reception record sequence, divide the reception record sequence into several target subsequences based on each division node; number the several target subsequences in order; establish a corresponding connection relationship between the target subsequences with the same corresponding numbers and the target operation record sets, and construct and generate several power control structures {target subsequence, target operation record set};

[0057] Step S4: Sort out the characteristic distribution presented by the corresponding historical operation records in each power control structure, and calculate the characteristic operation index for each power control structure; classify each power control structure according to the quantity distribution presented by the corresponding historical warning data reception records in each power control structure.

[0058] Among them, step S4 includes:

[0059] Step S4-1: Extract the target operation record sets within the corresponding power control structures respectively, obtain the total number of historical operation records M included in each target operation record set, capture the total number of privilege management accounts K for operating M historical operation records, and capture the highest privilege level C among the K privilege management accounts max and the lowest privilege level C min , calculate the level span P of the corresponding privilege management accounts within each target operation record set as P = |C max -C min |, and calculate the characteristic operation index β = (M × K) for the target operation record sets within the corresponding power control structures P , where M ≥ 1, K ≥ 1, P ≥ 0;

[0060] For example, in a certain target operation record set, the total number of historical operation records M = 10, and the total number of privilege management accounts K for operating the above 10 historical operation records is captured as 4;

[0061] Among them, the privilege levels of the 4 privilege management accounts are: privilege management account 1 is privilege level 3, privilege management account 2 is privilege level 2, privilege management account 3 is privilege level 3, and privilege management account 4 is privilege level 1; if the higher the privilege level value, the greater the specific operation privilege of the corresponding privilege management account, then among the above 4 privilege management accounts, the highest privilege level C max = 3 and the lowest privilege level C min = 1, then the level span P of the privilege management accounts is 3 - 1 = 2; if the higher the privilege level value, the smaller the specific operation privilege of the corresponding privilege management account, then among the above 4 privilege management accounts, the highest privilege level C max = 1 and the lowest privilege level C min = 3, then the level span P of the privilege management accounts is 3 - 1 = 2.

[0062] Step S4-2: Traverse the total number of historical warning data reception records N included in the corresponding target subsequences within each power control structure; set the power control structure with the corresponding total number N = 1 as the first characteristic power control structure, and set the power control structure with the corresponding total number N ≥ 2 as the second characteristic power control structure;

[0063] Step S5: Evaluate and calculate the theoretical characteristic operation index for the corresponding power control structure, and screen out abnormal control structures according to the deviation between the characteristic operation index of the corresponding power control structure and the theoretical characteristic operation index;

[0064] Among them, step S5 includes:

[0065] Step S5-1: Extract the warning data received by the power data management platform in each historical warning data reception record, perform feature extraction on the warning data, and obtain a set of characteristic warning information corresponding to each historical warning data reception record; traverse the characteristic operation index of the corresponding target operation record set in each power control structure and the set of characteristic warning information corresponding to each historical warning data reception record within the corresponding target subsequence;

[0066] Step S5-2: Assume that in a certain second characteristic power control structure, the characteristic operation index of the corresponding target operation record set is F, and the corresponding target subsequence is D; collect the sets of characteristic warning information extracted for all first characteristic power control structures to obtain set W; sequentially in set W, capture the set of characteristic warning information that has the highest similarity with the set of characteristic warning information corresponding to each historical warning data reception record within the target subsequence D, and at the same time identify the first characteristic power control structure corresponding to the set of characteristic warning information with the highest similarity;

[0067] Step S5-3: Assume that in set W, the set of characteristic warning information g that has the highest similarity with the set of characteristic warning information d corresponding to the i-th historical warning data reception record within the target subsequence D i between them, the set of characteristic warning information g with the highest similarity is g, and the similarity between the set of characteristic warning information g and the set of characteristic warning information d i is β, and the characteristic operation index of the first characteristic power control structure corresponding to the set of characteristic warning information g is Y g , where 0 ≤ β ≤ 1;

[0068] Step S5-4: Evaluate the characteristic operation value f contributed by the i-th historical warning data reception record within the target subsequence D to a certain second characteristic power control structure i = β × Y g , accumulate the characteristic operation values contributed by all historical warning data reception records within the target subsequence D to a certain second characteristic power control structure to obtain the theoretical characteristic operation index F', when F' - F > η, determine that a certain second characteristic power control structure is an abnormal control structure, where η is a threshold, when η > 0;

[0069] Step S6: From the anomaly control structure, identify and extract the warning data that will cause the control to become complicated due to the lag of the control adjustment operation, construct and generate a warning database for power data processing, and based on the warning database for power data processing, determine whether to send a warning prompt for timely processing to the management terminal for the newly added warning data reception record;

[0070] Among them, step S6 includes:

[0071] Step S6-1: For each anomaly control structure, collect the warning data received by the power data management platform in each historical warning data reception record included in the corresponding target subsequence, to obtain a warning data set S; perform feature extraction on the warning data set S to obtain a set of characteristic warning information R corresponding to each anomaly control structure;

[0072] Step S6-2: Extract the set of characteristic warning information corresponding to each historical warning data reception record in the target subsequence. When the set of characteristic warning information S corresponding to a certain historical warning data reception record and the set of characteristic warning information R satisfy: S - S ∩ R = U ≠ ∅, extract the characteristic warning information included in the set U as the target characteristic warning information;

[0073] Step S6-3: Collect all the target characteristic warning information extracted from each anomaly control structure, construct and generate a warning database for power data processing; whenever it is monitored that a new warning data reception record is added to the power data management platform, and after performing feature extraction on the newly received warning data in the current power data management platform, it is captured that there is characteristic warning information belonging to the warning database for power data processing, send a warning prompt for timely processing to the management terminal.

[0074] To better implement the above method, a power data large model construction system is also proposed. The system includes a power data management platform construction module, a historical data management module, a power control structure construction management module, a power control structure data management module, an anomaly control structure screening module, and a warning prompt management module;

[0075] The power data management platform construction module is used to construct a power data management platform containing several power control models, establish a remote communication connection between the power data management platform and the target power grid, remotely receive the warning data transmitted by the target power grid in real time, and the permission management account remotely controls the target power grid through the power data management platform based on the warning data;

[0076] The historical data management module is used to regularly extract historical warning data reception records and all historical operation records generated by the corresponding permission management account in the power data management platform from the historical operation logs of the power data management platform;

[0077] The power control structure construction management module is used to sort out the time feature distribution between the historical operation process data generated by the permission management account to control the target power grid through the power data management platform and the historical warning data received by the power data management platform, judge and identify the historical warning data reception records and historical operation records with corresponding connection relationships, and construct and extract the power control structure;

[0078] Among them, the power control structure construction management module includes a data distribution sorting unit and a structure construction management unit;

[0079] The data distribution sorting unit is used to sort out the time feature distribution between the historical operation process data generated by the permission management account to control the target power grid through the power data management platform and the historical warning data received by the power data management platform;

[0080] The structure construction management unit is used to judge and identify the historical warning data reception records and historical operation records with corresponding connection relationships, and construct and extract the power control structure;

[0081] The power control structure data management module is used to sort out the feature distribution presented by the corresponding historical operation records in each power control structure, calculate and evaluate the feature operation index for each power control structure; classify each power control structure according to the quantity distribution presented by the corresponding historical warning data reception records in each power control structure;

[0082] Among them, the power control structure data management module includes a feature operation index evaluation and calculation unit and a classification management unit;

[0083] The feature operation index evaluation and calculation unit is used to sort out the feature distribution presented by the corresponding historical operation records in each power control structure, and calculate and evaluate the feature operation index for each power control structure;

[0084] The classification management unit is used to classify each power control structure according to the quantity distribution presented by the corresponding historical warning data reception records in each power control structure;

[0085] The abnormal control structure screening module is used to evaluate and calculate the theoretical feature operation index of the corresponding power control structure, and screen out the abnormal control structure according to the deviation between the feature operation index and the theoretical feature operation index of the corresponding power control structure;

[0086] The warning prompt management module is used to identify and extract the warning data that will cause the control to become complicated due to the lag of the control adjustment operation from the abnormal control structure, construct and generate a power data processing warning library, and judge whether to send a warning prompt for timely processing to the management terminal based on the power data processing warning library for the newly added warning data reception record.

[0087] It should be noted that in this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device.

[0088] Finally, it should be noted that the above are only the preferred embodiments of the present invention and are not used to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for constructing a large power data model for a low-carbon power grid, characterized in that: The method includes: Step S1: Construct a power data management platform containing several power control models, establish a remote communication connection between the power data management platform and the target power grid, and remotely receive the early warning data transmitted by the target power grid in real time; based on the early warning data, the permission management account realizes remote control of the target power grid through the power data management platform; Step S2: Regularly extract the historical early warning data reception records from the historical operation logs of the power data management platform and all historical operation records generated by the corresponding permission management account in the power data management platform; Step S3: Sort out the time characteristic distribution between the historical operation process data generated by the permission management account to control the target power grid through the power data management platform and the historical early warning data received by the power data management platform, judge and identify the historical early warning data reception records and historical operation records with corresponding connection relationships, and construct and extract the power control structure; Step S4: Sort out the characteristic distribution of the corresponding historical operation records in each power control structure, and calculate and evaluate the characteristic operation index for each power control structure; classify each power control structure according to the quantity distribution of the corresponding historical early warning data reception records in each power control structure; Step S5: Evaluate and calculate the theoretical characteristic operation index for the corresponding power control structure, and screen out abnormal control structures according to the deviation between the characteristic operation index and the theoretical characteristic operation index of the corresponding power control structure; Step S6: Identify and extract the early warning data that will cause control complexity due to the lag of control adjustment operations from the abnormal control structures, construct and generate a power data processing early warning library, and judge whether to send an early warning prompt for timely processing to the management terminal based on the power data processing early warning library for the newly added early warning data reception records; The said Step S3 includes: Step S3-1: Sort all historical early warning data reception records in the order of the record generation time, and obtain a reception record sequence; in the reception record sequence, successively extract the record generation times T1 and T2 for each adjacent pair of historical early warning data reception records, where T1 < T2, to obtain several characteristic periods [T1, T2]; Step S3-2: Traverse the record generation times of each historical operation record, and respectively collect the historical operation records within each characteristic period to obtain an operation record set corresponding to each characteristic period; successively traverse the total number of historical operation records included in the operation record sets of each characteristic period, and set the operation record sets with the total number of historical operation records included being greater than or equal to 1 as the target operation record sets; number each target operation record set in order; Step S3-3: When the operation record set corresponding to the characteristic period between the jth historical early warning data reception record and the (j + 1)th historical early warning data reception record is the target operation record set, judge that there is a division node between the jth historical early warning data reception record and the (j + 1)th historical early warning data reception record in the reception record sequence; Step S3-4: Identify all the division nodes existing in the received record sequence, divide the received record sequence into a number of target subsequences based on each division node; number the target subsequences in sequence; establish a corresponding connection relationship between the target subsequences with the same corresponding numbers and the target operation record set, and construct and generate a number of power management and control structures {target subsequence, target operation record set}; The step S4 comprises: Step S4-1: Extract the target operation record sets in the corresponding power management and control structures respectively, obtain the total number M of historical operation records contained in each target operation record set, capture the total number K of permission management accounts that operate M historical operation records, and capture the highest permission level C among the K permission management accounts. max and minimum clearance level C min , calculate the level span P=|C of the corresponding permission management account in each target operation record set max -C min |, calculate the characteristic operation index β=(M×K) for the target operation record set in each power control structure P , where M≥1, K≥1, P≥0; Step S4-2: Traverse the total number N of historical warning data reception records contained in the corresponding target subsequence in each power control structure; set the power control structure with the corresponding total number N=1 as the first characteristic power control structure, and set the power control structure with the corresponding total number N≥2 as the second characteristic power control structure; The step S5 comprises: Step S5-1: extract the warning data received by the power data management platform in each historical warning data reception record, perform feature extraction on the warning data, and obtain a feature warning information set corresponding to each historical warning data reception record; traverse the feature operation index of the corresponding target operation record set in each power management and control structure and the feature warning information set corresponding to each historical warning data reception record in the corresponding target subsequence; Step S5-2: Assume that in a second characteristic power control structure, the characteristic operation index of the corresponding target operation record set is F, and the corresponding target subsequence is D; collect the characteristic warning information sets extracted from all the first characteristic power control structures to obtain a set W; in the set W, capture the characteristic warning information sets that meet the highest similarity with the characteristic warning information sets corresponding to each historical warning data reception record in the target subsequence D, and identify the first characteristic power control structure corresponding to the characteristic warning information set with the highest similarity; Step S5-3: Assume that the characteristic warning information set d corresponding to the i-th historical warning data reception record in the target subsequence D is captured in the set W i The feature warning information set that meets the highest similarity is g, and the feature warning information set g and the feature warning information set d i The similarity between them is β, and the characteristic operation index of the first characteristic power control structure corresponding to the characteristic warning information set g is Y g , where 0≦β≦1; Step S5-4: Evaluate the characteristic operation value f contributed by the i-th historical warning data reception record in the target subsequence D to the second characteristic power control structure i =β×Y g , accumulate the characteristic operation values ​​contributed by all historical warning data reception records in the target subsequence D to the second characteristic power control structure, and obtain the theoretical characteristic operation index F'. When F'-F>η, the second characteristic power control structure is judged to be an abnormal control structure, where η is a threshold value, and η>0; The step S6 comprises: Step S6-1: For each abnormal control structure, respectively, the warning data received by the power data management platform in each historical warning data reception record contained in the corresponding target subsequence is collected to obtain a warning data set S; feature extraction is performed on the warning data set S to obtain a feature warning information set R corresponding to each abnormal control structure; Step S6-2: extract the characteristic warning information set corresponding to each historical warning data reception record in the target subsequence, when the characteristic warning information set S corresponding to a certain historical warning data reception record satisfies the following conditions with the characteristic warning information set R: , extract the feature warning information contained in the set U as the target feature warning information; Step S6-3: Gather all target feature warning information extracted from each abnormal control structure to build a power data processing warning library; whenever a new warning data reception record is monitored on the power data management platform, and after feature extraction of the newly received warning data in the current power data management platform, capture the feature warning information that exists in the power data processing warning library, and send a warning prompt for timely processing to the management terminal.

2. A power data big model construction system, used to execute the power data big model construction method applied to low-carbon power grid according to claim 1, characterized in that: The system includes a power data management platform construction module, a historical data management module, a power control structure construction management module, a power control structure data management module, an abnormal control structure screening module, and an early warning prompt management module; The power data management platform construction module is used to construct a power data management platform including several power management and control models, establish a remote communication connection between the power data management platform and the target power grid, remotely receive the early warning data transmitted by the target power grid in real time, and the authority management account realizes remote management and control of the target power grid through the power data management platform based on the early warning data; The historical data management module is used to periodically extract historical warning data reception records and all historical operation records generated by the corresponding authority management account in the power data management platform from the historical operation log of the power data management platform; The power control structure construction management module is used to sort out the historical operation process data generated by the authority management account to achieve control over the target power grid through the power data management platform, and the time characteristic distribution presented between the historical warning data received by the power data management platform, judge and identify the historical warning data reception records and historical operation records that have corresponding connection relationships, and construct and extract the power control structure; The power control structure data management module is used to sort out the characteristic distribution of the corresponding historical operation records in each power control structure, and calculate and evaluate the characteristic operation index of each power control structure; Classify each power control structure according to the quantity distribution presented by the corresponding historical warning data reception records in each power control structure; The abnormal control structure screening module is used to evaluate and calculate the theoretical characteristic operation index of the corresponding power control structure, and screen out the abnormal control structure according to the deviation between the characteristic operation index of the corresponding power control structure and the theoretical characteristic operation index; The early warning management module is used to identify and extract early warning data that may complicate management and control due to delayed management and control adjustment operations from the abnormal management and control structure, build and generate an electric power data processing early warning library, determine whether to receive records of newly added early warning data based on the electric power data processing early warning library, and send early warning prompts for timely processing to the management terminal.

3. A power data big model construction system according to claim 2, characterized in that: The power control structure construction management module includes a data distribution combing unit and a structure construction management unit; The data distribution combing unit is used to comb the historical operation process data generated by the authority management account to achieve control over the target power grid through the power data management platform, and the time characteristic distribution presented between the historical warning data received by the power data management platform; The structure construction management unit is used to judge and identify the historical warning data reception records and historical operation records that have corresponding connection relationships, and construct and extract the power management and control structure.

4. A power data big model construction system according to claim 2, characterized in that: The power control structure data management module includes a characteristic operation index evaluation calculation unit and a classification management unit; The characteristic operation index evaluation and calculation unit is used to sort out the characteristic distribution presented by the corresponding historical operation records in each power management and control structure, and calculate and evaluate the characteristic operation index of each power management and control structure; The classification management unit is used to classify each power management and control structure according to the quantity distribution presented by the corresponding historical warning data reception records in each power management and control structure.

Images