A power big data analysis method for distributed management of smart grids

Abstract

The present invention discloses a method for analyzing electric power big data for distributed management of smart grids, which relates to the technical field of power grid data management, including: recording all electric power facilities connected to the smart grid distributed management system as target electric power facilities; selecting at least one master management electric power facility, and recording the target electric power facilities other than the master management electric power facility as slave management electric power facilities; dividing all electric power data of the smart grid distributed management system into low-frequency call electric power data and high-frequency call electric power data; obtaining electric power data request instructions of the target electric power facilities in real time, and classifying the instructions, and executing and responding to all electric power data request instructions through the master management electric power facility and the slave management electric power facility respectively. The advantages of the present invention are: realizing the read-write separation cluster management of smart grid data, while ensuring that the read-write separation cluster during the operation of the power grid is consistent with the actual operation status of the power grid, thereby achieving high reliability of the system.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid data management, and in particular to a method for analyzing power big data for distributed management of smart grids. Background Art

[0002] With the rapid development of smart grids, massive amounts of data are being generated within power systems, including real-time data, historical data, and equipment status information from various aspects of power generation, transmission, distribution, and consumption. This data holds immense value and is crucial for improving grid efficiency, optimizing resource allocation, and enhancing system stability and reliability.

[0003] In the existing distributed management process of smart grids, due to the lack of big data analysis planning for data management, the management solution is difficult to meet the real-time changes in the distributed management process of smart grids, and it is difficult to achieve efficient data management for smart grids. Summary of the Invention

[0004] In order to solve the above technical problems, a method for analyzing power big data for distributed management of smart grids is provided. This technical solution solves the problem that in the existing distributed management process of smart grids, due to the lack of big data analysis planning for data management, the management solution is difficult to meet the real-time changes in the distributed management process of smart grids, and it is difficult to achieve efficient data management for smart grids.

[0005] In order to achieve the above objects, the technical solution adopted by the present invention is:

[0006] A power big data analysis method for distributed management of smart grids, comprising:

[0007] All power facilities connected to the smart grid distributed management system are recorded as target power facilities;

[0008] Obtain historical power operation logs of all target power facilities;

[0009] Selecting at least one primary management power facility from all target power facilities, and recording the target power facilities other than the primary management power facility as secondary management power facilities;

[0010] Based on the historical power operation logs of all target power facilities, all power data of the smart grid distributed management system are divided into low-frequency call power data and high-frequency call power data. The low-frequency call power data is stored in the main management power facility, and the high-frequency call power data is stored in all target power facilities.

[0011] Obtain power data request instructions of target power facilities in real time;

[0012] If the power data request instruction of the target power facility is a read instruction, then identify whether the object of the power data request instruction is high-frequency call power data. If so, directly retrieve it from the target power facility. If not, mark the power data request instruction as a main request instruction and send it to the main management power facility;

[0013] If the power data request instruction of the target power facility is a write instruction, the power data request instruction is marked as a master request instruction and sent to the master management power facility;

[0014] At least one master management power facility responds to all master request instructions, and updates the low-frequency call power data and / or high-frequency call power data based on the master request instructions, and synchronizes the updated low-frequency call power data to all master management power facilities, and synchronizes the updated high-frequency call power data to all target power facilities.

[0015] Preferably, selecting at least one master management power facility from all target power facilities specifically includes:

[0016] Obtaining the command response performance of all target power facilities;

[0017] Determining a read instruction response requirement of the target power facility based on a historical power operation log of the target power facility;

[0018] Subtracting the read instruction response requirement of the target power facility from the instruction response performance of the target power facility to obtain the redundant instruction response performance of the target power facility;

[0019] Determine the total demand for write command responses of the smart grid distributed management system based on the historical power operation logs of all target power facilities;

[0020] Selecting a number of target power facilities whose sum of redundant instruction response performance is greater than the total write instruction response requirement from the target power facilities to form a preliminary screening target power facility group, and obtaining all preliminary screening target power facility groups that meet the conditions;

[0021] Determine the sum of target power facility redundant instruction response performance, the number of target power facilities, the inter-group communication index, and the center communication index for each preliminarily screened target power facility group;

[0022] Based on the best-inferior solution algorithm, the optimal target power facility group is screened out from the initially screened target power facility group, and all target power facilities in the optimal target power facility group are used as the main management power facilities.

[0023] Preferably, the method for determining the inter-group communication index is:

[0024] If the number of target power facilities in the initial screening target power facilities is 1, the inter-group communication index is 0;

[0025] If the number of target power facilities in the preliminary screening target power facilities is greater than 1, the average value of the communication delay between a single target power facility in the preliminary screening target power facilities and the remaining target power facilities in the preliminary screening target power facilities is used as the single-group communication index of the corresponding target power facilities, and the average value of the single-group communication indexes of all target power facilities in the preliminary screening target power facilities is used as the inter-group communication index;

[0026] The method for determining the centering communication index is:

[0027] The average value of the communication delay of a single target power facility in the preliminary screening target power facilities, excluding the target power facilities in the preliminary screening target power facilities group, is used as the single centered communication index of the corresponding target power facility;

[0028] The average value of the single centered communication indices of all target power facilities in the initial screening target power facilities is taken as the centered communication index.

[0029] Preferably, the method of screening out the optimal target power facility group from the pre-screened target power facility group based on the superiority-inferiority solution algorithm and taking all target power facilities in the optimal target power facility group as primary management power facilities specifically includes:

[0030] Based on the sum of the target power facility redundant command response performance of all initially screened target power facility groups, the number of target power facilities, the inter-group communication index, and the central communication index, an initial evaluation matrix A is constructed. ,in, is the sum of the redundant instruction response performance of the target power facilities of the i-th group of initially screened target power facilities, is the number of target power facilities in the i-th group of initially screened target power facilities, is the inter-group communication index of the i-th group of primary screening target power facilities, is the median communication index of the i-th group of primary screening target power facilities, The total number of power facility groups targeted for initial screening;

[0031] Based on the standardization processing algorithm, the initial evaluation matrix A is converted into the standard evaluation matrix B. ,in, is the normalized value of the sum of the redundant instruction response performance of the target power facilities of the i-th group of pre-screened target power facilities, is the standardized value of the number of target power facilities in the i-th group of initially screened target power facilities, is the standardized value of the inter-group communication index of the i-th group of primary screening target power facilities, is the standardized value of the central communication index of the i-th group of primary screening target power facilities;

[0032] in, ;

[0033] Based on the standard evaluation matrix B, the optimal solution is determined and the worst solution ,in, ,in:

[0034] for The maximum value in for The minimum value in ;

[0035] for The minimum value in for The maximum value in ;

[0036] for The minimum value in for The maximum value in ;

[0037] for The minimum value in for The maximum value in ;

[0038] Calculate the optimization index of each initially screened target power facility group based on the best-in-bad solution formula;

[0039] Screen out the initial target power facility group corresponding to the maximum value of the optimization index as the optimal target power facility group;

[0040] The formula for the superiority and inferiority solution is specifically: ,

[0041] in, is the optimization index of the i-th primary screening target power facility group.

[0042] Preferably, the method of dividing all power data of the smart grid distributed management system into low-frequency call power data and high-frequency call power data based on the power historical operation logs of all target power facilities specifically includes:

[0043] Set an analysis period;

[0044] Based on the historical power operation logs of all target power facilities, count the number of times the power data is called in at least two analysis cycles closest to the current moment, and record it as the number of sample calls of the power data;

[0045] Based on the number of times the sample power data are called, the number of times the power data will be called in the next analysis period is predicted, which is recorded as the predicted number of times the power data will be called;

[0046] The power data with a predicted call frequency greater than a preset value is recorded as high-frequency call power data, and the power data with a predicted call frequency less than a preset value is recorded as low-frequency call power data.

[0047] Preferably, the at least one master management power facility responding to all master request instructions specifically includes:

[0048] Divide all main request instructions into several instruction sets according to the objects;

[0049] Based on the load balancing algorithm, several instruction sets are allocated to the corresponding main management power facilities;

[0050] The main management power facility responds to the main request instruction in sequence according to the instruction request time sequence in all instruction sets.

[0051] Preferably, the load balancing algorithm is specifically:

[0052] generating at least one instruction set allocation scheme under the condition that the total performance requirement of all instruction sets allocated to the primary management power facility is less than the redundant instruction response performance of the primary management power facility;

[0053] calculating, under the instruction set allocation scheme, a ratio of the total performance requirements of all instruction sets allocated to each primary management power facility to the redundant instruction response performance of the primary management power facility as the load of the primary management power facility;

[0054] Based on the loads of all main managed power facilities under the instruction set allocation scheme, the balance index of the instruction set allocation scheme is calculated using a balance evaluation formula;

[0055] Filter out the instruction set allocation scheme with the smallest balance index as the optimal instruction set allocation scheme;

[0056] Allocate the instruction set to the corresponding main management power facility according to the optimal instruction set allocation plan;

[0057] The balance evaluation formula is specifically: ;

[0058] Where, is the equilibrium indicator, The load of the k-th master management power facility, The total number of main managed power facilities.

[0059] Compared with the prior art, the present invention has the following beneficial effects:

[0060] This paper proposes a power big data analysis solution for distributed management of smart grids. It adopts a data read-write separation cluster model to manage data in the distributed management system of smart grids. It can effectively ensure the request response efficiency and data consistency under the condition of concurrent access by multiple facilities during the operation of smart grids.

[0061] This paper proposes a power big data analysis solution for distributed management of smart grids. Based on the historical operation logs of the smart grid distributed management system, it uses big data technology to analyze the data call instruction status of distributed power facilities and dynamically adjust the read-write separation cluster of the smart grid distributed management system. This effectively ensures that the read-write separation cluster during grid operation is consistent with the actual operating status of the grid, achieving high system reliability.

[0062] The present invention proposes a power big data analysis solution for distributed management of smart grids. By adopting data classification, main request instructions for the same data are assigned to the same main request instructions for response execution, effectively ensuring data consistency during data instruction execution. At the same time, when allocating main request instructions, the load balancing principle is followed, effectively ensuring the reliability and availability of the smart grid distributed management system when responding to instructions from power facilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 Flowchart of the power big data analysis method for distributed management of smart grid proposed in this solution;

[0064] Figure 2 A flow chart of a method for selecting at least one master management power facility from all target power facilities in this solution;

[0065] Figure 3 This is a flow chart of a method for selecting the optimal target power facility group from the pre-screened target power facility group in this solution;

[0066] Figure 4 This is a flow chart of a method for dividing all power data of a smart grid distributed management system into low-frequency call power data and high-frequency call power data in this solution;

[0067] Figure 5 A flow chart of a method for responding to all master request instructions for at least one master management power facility in this solution;

[0068] Figure 6 This is a flow chart of the load balancing algorithm in this solution;

[0069] Figure 7 Schematic diagram of the electronic device structure in the present invention;

[0070] Figure 8This is a schematic diagram of the computer-readable storage medium structure in the present invention. DETAILED DESCRIPTION

[0071] The following description is intended to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are merely examples, and those skilled in the art may conceive of other obvious variations.

[0072] Reference Figure 1 As shown, a power big data analysis method for distributed management of smart grids includes:

[0073] All power facilities connected to the smart grid distributed management system are recorded as target power facilities;

[0074] Obtain historical power operation logs of all target power facilities;

[0075] Selecting at least one primary management power facility from all target power facilities, and recording the target power facilities other than the primary management power facility as secondary management power facilities;

[0076] Based on the historical power operation logs of all target power facilities, all power data of the smart grid distributed management system are divided into low-frequency call power data and high-frequency call power data. The low-frequency call power data is stored in the main management power facility, and the high-frequency call power data is stored in all target power facilities.

[0077] Obtain power data request instructions of target power facilities in real time;

[0078] If the power data request instruction of the target power facility is a read instruction, then identify whether the object of the power data request instruction is high-frequency call power data. If so, directly retrieve it from the target power facility. If not, mark the power data request instruction as a main request instruction and send it to the main management power facility;

[0079] If the power data request instruction of the target power facility is a write instruction, the power data request instruction is marked as a master request instruction and sent to the master management power facility;

[0080] At least one master management power facility responds to all master request instructions, and updates the low-frequency call power data and / or high-frequency call power data based on the master request instructions, and synchronizes the updated low-frequency call power data to all master management power facilities, and synchronizes the updated high-frequency call power data to all target power facilities.

[0081] This solution adopts the data read-write separation cluster model to manage the data of the smart grid distributed management system, which can effectively ensure the request response efficiency and data consistency under the condition of concurrent access by multiple facilities during the operation of the smart grid.

[0082] Reference Figure 2 As shown, selecting at least one master management power facility from all target power facilities specifically includes:

[0083] Obtaining the command response performance of all target power facilities;

[0084] Determining a read instruction response requirement of the target power facility based on a historical power operation log of the target power facility;

[0085] Subtracting the read instruction response requirement of the target power facility from the instruction response performance of the target power facility to obtain the redundant instruction response performance of the target power facility;

[0086] Determine the total demand for write command responses of the smart grid distributed management system based on the historical power operation logs of all target power facilities;

[0087] Selecting a number of target power facilities whose sum of redundant instruction response performance is greater than the total write instruction response requirement from the target power facilities to form a preliminary screening target power facility group, and obtaining all preliminary screening target power facility groups that meet the conditions;

[0088] Determine the sum of target power facility redundant instruction response performance, the number of target power facilities, the inter-group communication index, and the center communication index for each preliminarily screened target power facility group;

[0089] Based on the best-inferior solution algorithm, the optimal target power facility group is screened out from the initially screened target power facility group, and all target power facilities in the optimal target power facility group are used as the main management power facilities.

[0090] Based on the historical operation logs of the smart grid distributed management system, big data technology is used to analyze the data call instruction status of distributed power facilities, analyze the total demand for write instructions of the smart grid, and dynamically adjust the read-write separation cluster of the smart grid distributed management system based on the total demand for write instructions. This effectively ensures that the read-write separation cluster during the operation of the power grid is consistent with the actual operation status of the power grid, achieving high reliability of the system.

[0091] The method for determining the inter-group communication index is:

[0092] If the number of target power facilities in the initial screening target power facilities is 1, the inter-group communication index is 0;

[0093] If the number of target power facilities in the preliminary screening target power facilities is greater than 1, the average value of the communication delay between a single target power facility in the preliminary screening target power facilities and the remaining target power facilities in the preliminary screening target power facilities is used as the single-group communication index of the corresponding target power facilities, and the average value of the single-group communication indexes of all target power facilities in the preliminary screening target power facilities is used as the inter-group communication index;

[0094] The method for determining the centering communication index is:

[0095] The average value of the communication delay of a single target power facility in the preliminary screening target power facilities, excluding the target power facilities in the preliminary screening target power facilities group, is used as the single centered communication index of the corresponding target power facility;

[0096] The average value of the single centered communication indices of all target power facilities in the initial screening target power facilities is taken as the centered communication index.

[0097] As the main management power facility, it needs to receive read and write request instructions for data from the target power facility, and at the same time needs to synchronize the updated data to the remaining target power facilities. Therefore, when evaluating the main management power facility, it is necessary to comprehensively consider the communication status between multiple main management power facilities and the communication status between the main management power facility and the remaining target power facilities.

[0098] Reference Figure 3 As shown, based on the superiority and inferiority solution algorithm, the optimal target power facility group is screened out from the pre-screened target power facility group, and all target power facilities in the optimal target power facility group are used as the main management power facilities, specifically including:

[0099] Based on the sum of the target power facility redundant command response performance of all initially screened target power facility groups, the number of target power facilities, the inter-group communication index, and the central communication index, an initial evaluation matrix A is constructed. ,in, is the sum of the redundant instruction response performance of the target power facilities of the i-th group of initially screened target power facilities, is the number of target power facilities in the i-th group of initially screened target power facilities, is the inter-group communication index of the i-th group of primary screening target power facilities, is the median communication index of the i-th group of primary screening target power facilities, The total number of power facility groups targeted for initial screening;

[0100] Based on the standardization processing algorithm, the initial evaluation matrix A is converted into the standard evaluation matrix B. ,in, is the normalized value of the sum of the redundant instruction response performance of the target power facilities of the i-th group of pre-screened target power facilities, is the standardized value of the number of target power facilities in the i-th group of initially screened target power facilities, is the standardized value of the inter-group communication index of the i-th group of primary screening target power facilities, is the standardized value of the central communication index of the i-th group of primary screening target power facilities;

[0101] in, ;

[0102] Based on the standard evaluation matrix B, the optimal solution is determined and the worst solution ,in, ,in:

[0103] for The maximum value in for The minimum value in ;

[0104] for The minimum value in for The maximum value in ;

[0105] for The minimum value in for The maximum value in ;

[0106] for The minimum value in for The maximum value in ;

[0107] Calculate the optimization index of each initially screened target power facility group based on the best-in-bad solution formula;

[0108] Screen out the initial target power facility group corresponding to the maximum value of the optimization index as the optimal target power facility group;

[0109] The formula for the superiority and inferiority solution is specifically: ,

[0110] in, is the optimization index of the i-th primary screening target power facility group.

[0111] The best-in-competence algorithm is a commonly used multi-criteria decision analysis method used to evaluate and compare a series of alternative solutions to find the optimal solution. In this solution, multiple read-write splitting cluster solutions are generated while meeting the total demand for write instructions under the current power grid state. The best-in-competence algorithm is used to comprehensively evaluate the distance between each read-write splitting cluster solution and the ideal optimal solution. By making full use of big data information, the gap between the evaluated solutions is accurately reflected, and the read-write splitting cluster solution that best matches the power grid state is selected to ensure the efficient operation of the smart grid distributed management system.

[0112] Reference Figure 4 As shown, based on the historical power operation logs of all target power facilities, all power data of the smart grid distributed management system are divided into low-frequency call power data and high-frequency call power data, specifically including:

[0113] Set an analysis period;

[0114] Based on the historical power operation logs of all target power facilities, count the number of times the power data is called in at least two analysis cycles closest to the current moment, and record it as the number of sample calls of the power data;

[0115] Based on the number of times the sample power data are called, the number of times the power data will be called in the next analysis period is predicted, which is recorded as the predicted number of times the power data will be called;

[0116] The power data with a predicted call frequency greater than a preset value is recorded as high-frequency call power data, and the power data with a predicted call frequency less than a preset value is recorded as low-frequency call power data.

[0117] The prediction methods include but are not limited to averaging and linear regression prediction;

[0118] In some preferred embodiments, the specific steps of linear regression prediction are:

[0119] Number the number of times all power data samples are called in order from farthest to closest to the current moment;

[0120] Based on the principle of linear regression, a prediction formula is constructed;

[0121] Substitute the sample call times of all power data into the prediction formula to obtain the predicted call times of power data;

[0122] The specific prediction formula is: ;

[0123] Where, The number of predicted calls for power data, The number of times the sample of power data numbered v is called, The total number of times the power data sample is called.

[0124] Reference Figure 5 As shown, at least one master management power facility responds to all master request instructions specifically including:

[0125] Divide all main request instructions into several instruction sets according to the objects;

[0126] Based on the load balancing algorithm, several instruction sets are allocated to the corresponding main management power facilities;

[0127] The main management power facility responds to the main request instruction in sequence according to the instruction request time sequence in all instruction sets.

[0128] By adopting data classification, the main request instructions for the same data are assigned to the same main request instructions for response execution, which effectively ensures the data consistency when the data instructions are executed. At the same time, when allocating the main request instructions, the load balancing principle is followed, which effectively ensures the reliability and availability of the smart grid distributed management system in responding to the instructions of the power facilities.

[0129] Reference Figure 6 As shown in the figure, the load balancing algorithm is as follows:

[0130] generating at least one instruction set allocation scheme under the condition that the total performance requirement of all instruction sets allocated to the primary management power facility is less than the redundant instruction response performance of the primary management power facility;

[0131] calculating, under the instruction set allocation scheme, a ratio of the total performance requirements of all instruction sets allocated to each primary management power facility to the redundant instruction response performance of the primary management power facility as the load of the primary management power facility;

[0132] Based on the loads of all main managed power facilities under the instruction set allocation scheme, the balance index of the instruction set allocation scheme is calculated using a balance evaluation formula;

[0133] Filter out the instruction set allocation scheme with the smallest balance index as the optimal instruction set allocation scheme;

[0134] Allocate the instruction set to the corresponding main management power facility according to the optimal instruction set allocation plan;

[0135] The balance evaluation formula is specifically: ;

[0136] Where, is the equilibrium indicator, The load of the k-th master management power facility, The total number of main managed power facilities.

[0137] In this scheme, based on the standard deviation evaluation principle, the standard deviation of the load of all main management power facilities under the instruction set distribution scheme is calculated, and the standard deviation is used as the balance indicator. The smaller the balance indicator, the more concentrated the load distribution of each main management power facility and the more balanced the load of each main management power facility.

[0138] Furthermore, the method according to the embodiment of the present application can also be used with the aid of Figure 7 The electronic device architecture shown in FIG. Figure 7 As shown, the electronic device 500 may include a bus 501, one or more CPUs 502, a read-only memory (ROM) 503, a random access memory (RAM) 504, a communication port 505 connected to a network, an input / output component 506, a hard disk 507, etc. The storage device in the electronic device 500, such as the ROM 503 or the hard disk 507, may store a power big data analysis method for distributed management of a smart grid provided in this application. The electronic device 500 may also include a user interface 508. Of course, Figure 7 The architecture shown is only exemplary and can be omitted according to actual needs when implementing different devices. Figure 7 One or more components of an electronic device are shown.

[0139] Figure 8 This is a schematic diagram of the computer-readable storage medium structure provided by an embodiment of the present application. Figure 8 As shown, a computer-readable storage medium 600 according to one embodiment of the present application is shown. A computer-readable program is stored on the computer-readable storage medium 600. When the computer-readable program is executed by a processor, a method for analyzing power big data for distributed management of a smart grid according to an embodiment of the present application described with reference to the above figures can be executed. The storage medium 600 includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory. Non-volatile memory may include, for example, read-only memory (ROM), a hard disk, a flash memory, etc.

[0140] In summary, the advantages of the present invention are: effectively ensuring that the read-write separation cluster during the operation of the power grid is consistent with the actual operation state of the power grid, thereby achieving high reliability of the system.

[0141] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions merely illustrate the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for analyzing power big data for distributed management of smart grids, characterized in that: include: All power facilities connected to the smart grid distributed management system are recorded as target power facilities; Obtain historical power operation logs of all target power facilities; Selecting at least one primary management power facility from all target power facilities, and recording the target power facilities other than the primary management power facility as secondary management power facilities; Based on the historical power operation logs of all target power facilities, all power data of the smart grid distributed management system are divided into low-frequency call power data and high-frequency call power data. The low-frequency call power data is stored in the main management power facility, and the high-frequency call power data is stored in all target power facilities. Obtain power data request instructions of target power facilities in real time; If the power data request instruction of the target power facility is a read instruction, then identify whether the object of the power data request instruction is high-frequency call power data. If so, directly retrieve it from the target power facility. If not, mark the power data request instruction as a main request instruction and send it to the main management power facility; If the power data request instruction of the target power facility is a write instruction, the power data request instruction is marked as a master request instruction and sent to the master management power facility; At least one master management power facility responds to all master request instructions, and updates the low-frequency call power data and / or high-frequency call power data based on the master request instructions, and synchronizes the updated low-frequency call power data to all master management power facilities, and synchronizes the updated high-frequency call power data to all target power facilities.

2. The method for analyzing power big data for distributed management of smart grid according to claim 1, characterized in that: The step of selecting at least one primary management power facility from all target power facilities specifically includes: Obtaining the command response performance of all target power facilities; Determining a read instruction response requirement of the target power facility based on a historical power operation log of the target power facility; Subtracting the read instruction response requirement of the target power facility from the instruction response performance of the target power facility to obtain the redundant instruction response performance of the target power facility; Determine the total demand for write command responses of the smart grid distributed management system based on the historical power operation logs of all target power facilities; Selecting a number of target power facilities whose sum of redundant instruction response performance is greater than the total write instruction response requirement from the target power facilities to form a preliminary screening target power facility group, and obtaining all preliminary screening target power facility groups that meet the conditions; Determine the sum of target power facility redundant instruction response performance, the number of target power facilities, the inter-group communication index, and the center communication index for each preliminarily screened target power facility group; Based on the best-inferior solution algorithm, the optimal target power facility group is screened out from the initially screened target power facility group, and all target power facilities in the optimal target power facility group are used as the main management power facilities.

3. The method for analyzing power big data for distributed management of smart grid according to claim 2, characterized in that: The method for determining the inter-group communication index is: If the number of target power facilities in the initial screening target power facilities is 1, the inter-group communication index is 0; If the number of target power facilities in the preliminary screening target power facilities is greater than 1, the average value of the communication delay between a single target power facility in the preliminary screening target power facilities and the remaining target power facilities in the preliminary screening target power facilities is used as the single-group communication index of the corresponding target power facilities, and the average value of the single-group communication indexes of all target power facilities in the preliminary screening target power facilities is used as the inter-group communication index; The method for determining the centering communication index is: The average value of the communication delay of a single target power facility in the preliminary screening target power facilities, excluding the target power facilities in the preliminary screening target power facilities group, is used as the single centered communication index of the corresponding target power facility; The average value of the single centered communication indices of all target power facilities in the initial screening target power facilities is taken as the centered communication index.

4. The method for analyzing power big data for distributed management of smart grid according to claim 3, characterized in that: The method of selecting an optimal target power facility group from the pre-screened target power facility group based on the superiority-inferiority solution algorithm and using all target power facilities in the optimal target power facility group as primary management power facilities specifically includes: Based on the sum of the target power facility redundant command response performance of all initially screened target power facility groups, the number of target power facilities, the inter-group communication index, and the central communication index, an initial evaluation matrix A is constructed. ,in, is the sum of the redundant instruction response performance of the target power facilities of the i-th group of initially screened target power facilities, is the number of target power facilities in the i-th group of initially screened target power facilities, is the inter-group communication index of the i-th group of primary screening target power facilities, is the median communication index of the i-th group of primary screening target power facilities, The total number of power facility groups targeted for initial screening; Based on the standardization processing algorithm, the initial evaluation matrix A is converted into the standard evaluation matrix B. ,in, is the normalized value of the sum of the redundant instruction response performance of the target power facilities of the i-th group of pre-screened target power facilities, is the standardized value of the number of target power facilities in the i-th group of initially screened target power facilities, is the standardized value of the inter-group communication index of the i-th group of primary screening target power facilities, is the standardized value of the central communication index of the i-th group of primary screening target power facilities; in, ; Based on the standard evaluation matrix B, the optimal solution is determined and the worst solution ,in, ,in: for The maximum value in for The minimum value in ; for The minimum value in for The maximum value in ; for The minimum value in for The maximum value in ; for The minimum value in for The maximum value in ; Calculate the optimization index of each initially screened target power facility group based on the best-in-bad solution formula; Screen out the initial target power facility group corresponding to the maximum value of the optimization index as the optimal target power facility group; The formula for the superiority and inferiority solution is specifically: , in, is the optimization index of the i-th primary screening target power facility group.

5. The method for analyzing electric power big data for distributed management of smart grid according to claim 4, characterized in that: The method of dividing all power data of the smart grid distributed management system into low-frequency call power data and high-frequency call power data based on the power historical operation logs of all target power facilities specifically includes: Set an analysis period; Based on the historical power operation logs of all target power facilities, count the number of times the power data is called in at least two analysis cycles closest to the current moment, and record it as the number of sample calls of the power data; Based on the number of times the sample power data are called, the number of times the power data will be called in the next analysis period is predicted, which is recorded as the predicted number of times the power data will be called; The power data with a predicted call frequency greater than a preset value is recorded as high-frequency call power data, and the power data with a predicted call frequency less than a preset value is recorded as low-frequency call power data.

6. The method for analyzing electric power big data for distributed management of smart grid according to claim 5, characterized in that: The at least one master management power facility responding to all master request instructions specifically includes: Divide all main request instructions into several instruction sets according to the objects; Based on the load balancing algorithm, several instruction sets are allocated to the corresponding main management power facilities; The main management power facility responds to the main request instruction in sequence according to the instruction request time sequence in all instruction sets.

7. The method for analyzing electric power big data for distributed management of smart grid according to claim 6, characterized in that: The load balancing algorithm is specifically as follows: generating at least one instruction set allocation scheme under the condition that the total performance requirement of all instruction sets allocated to the primary management power facility is less than the redundant instruction response performance of the primary management power facility; calculating, under the instruction set allocation scheme, a ratio of the total performance requirements of all instruction sets allocated to each primary management power facility to the redundant instruction response performance of the primary management power facility as the load of the primary management power facility; Based on the loads of all main managed power facilities under the instruction set allocation scheme, the balance index of the instruction set allocation scheme is calculated using a balance evaluation formula; Filter out the instruction set allocation scheme with the smallest balance index as the optimal instruction set allocation scheme; Allocate the instruction set to the corresponding main management power facility according to the optimal instruction set allocation plan; The balance evaluation formula is specifically: ; Where, is the equilibrium indicator, The load of the k-th master management power facility, The total number of main managed power facilities.

8. An electronic device, characterized in that: include: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the power big data analysis method for distributed management of smart grids as described in any one of claims 1-7.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the power big data analysis method for distributed management of smart grids according to any one of claims 1 to 7 is implemented.

Images