Power distribution network typical scene division method based on fuzzy C-means clustering theory

Abstract

The invention discloses a power distribution network typical scene division method based on a fuzzy C-means clustering theory, and aims to explore a novel local power system development mode adapted to rural energy revolution and dual-carbon objectives, comprising the following steps: 1, systematically constructing a multi-dimensional division system of a rural power distribution network typical scene, a comprehensive index system is constructed from four aspects of a new energy development level, a power grid construction level, a load level and an energy storage construction level; 2, establishing a combined weighting method model of an entropy weight method and a fuzzy analytic hierarchy process; 3, a fuzzy C-means clustering algorithm is put forward as a power grid scene division method; and 4, defining an evaluation standard for a division result through the clustering algorithm. According to the method, the power distribution network scene is divided, so that an important reference is provided for planning, construction and operation of the power distribution network, and the method has important significance in promoting energy revolution and realizing a dual-carbon target.

Description

Technical Field

[0001] The present invention belongs to the field of typical scene division, and in particular relates to a division method for typical scenes of rural power distribution networks based on fuzzy C-means clustering theory. Background Art

[0002] In the study of division indicators, how to scientifically divide distribution network scenarios has always been a research hotspot in academia; reasonable and scientific division schemes can provide development directions for long-term grid planning, meet the development of scenario grids and improve the economic efficiency of grid operation; a variety of evaluation index systems for new power systems have been constructed from the aspects of intelligent construction, coordinated development, operation risks, and investment benefits, covering all major links of new power systems. However, the emphasis in the process of studying the evaluation index system is different, resulting in a single index system that does not have the ability to comprehensively evaluate the investment, construction, development, production, operation, risks, and benefits of new power systems; among them, there are serious subjective colors in the empowerment method, insufficient division focus, too few scenarios required for the survey, and lack of representativeness; in the division of rural scenarios, rural data is diverse and complex, and the current division method cannot meet the rationality and effectiveness of the division results; Summary of the Invention

[0003] In order to address the shortcomings of the above-mentioned existing technologies, the present invention proposes a division method for typical scenarios of rural distribution networks based on fuzzy C-means clustering theory, in order to accurately divide the distribution network scenarios, thereby providing an important reference for the planning, construction and operation of the distribution network, which is of great significance to promoting the energy revolution and achieving the dual carbon goals.

[0004] In order to achieve the above-mentioned object, the present invention adopts the following technical solutions:

[0005] The characteristic of the method for dividing typical scenarios of distribution network based on fuzzy C-means clustering theory of the present invention is that it is carried out according to the following steps:

[0006] Step 1: Construct a comprehensive indicator system for multi-dimensional classification of typical distribution network scenarios;

[0007] Step 1.1: Construct an indicator system for the new energy development level;

[0008] Step 1.2: Construct an indicator system for the horizontal dimension of power grid construction;

[0009] Step 1.3: Construct an indicator system for the load level dimension;

[0010] Step 1.4: Construct an indicator system for energy storage construction level dimensions;

[0011] Step 2: Obtain the original data of the sample scenarios under each indicator system, and use the entropy weight method and fuzzy hierarchical analysis method to obtain the combined weight vector ;

[0012] Step 3: Use the fuzzy C-means clustering algorithm and combine it with the combined weight vector , all sample scenarios are divided, and the division results of typical distribution network scenarios are obtained.

[0013] The characteristic of the method for dividing typical scenarios of distribution network based on fuzzy C-means clustering theory described in the present invention is that step 1.1 is performed as follows:

[0014] Step 1.1.1 Use formula (1) to construct the total photovoltaic installed capacity index for the jth sample scenario: :

[0015] (1)

[0016] In formula (1), represents the centralized photovoltaic installed capacity of the jth sample scenario, represents the distributed photovoltaic installed capacity of the jth sample scenario, and m represents the total number of sample scenarios;

[0017] Step 1.1.2 Use formula (2) to construct the photovoltaic installed capacity growth rate index for the jth sample scenario: :

[0018] (2)

[0019] In formula (2), represents the total installed capacity of photovoltaic power generation in the jth sample scenario, Indicates the jth sample scene in y time period The total installed capacity of photovoltaic power generation equipment before; y represents the time period the number of

[0020] Step 1.1.3 Use formula (3) to construct the new energy penetration index of the jth sample scenario :

[0021]

[0022] In formula (3), represents the wind power installed capacity of the jth sample scenario, represents the installed capacity of biomass energy in the jth sample scenario;

[0023] Step 1.1.4 Use formula (4) to construct the distributed photovoltaic carrying capacity index of the jth sample scenario :

[0024] (4)

[0025] In formula (4), represents the distributed photovoltaic output coefficient of the jth sample scenario at time t, represents the load of the jth sample scenario at time t, Indicates that the substation in the jth sample scenario provides the maximum power.

[0026] Furthermore, the step 1.2 is performed as follows:

[0027] Step 1.2.1 Use formula (5) to construct the power supply voltage qualification rate index of the jth sample scenario :

[0028] (5)

[0029] In formula (5), represents the unqualified operation time of the grid voltage in the jth sample scenario, represents the total operation time of the power grid in the jth sample scenario;

[0030] Step 1.2.2 Use formula (6) to construct the N-1 pass rate index of the jth sample scenario :

[0031] (6)

[0032] In formula (6), represents the total number of line transformers in the jth sample scenario that meet the N-1 condition, represents the total number of line transformer elements in the jth sample scenario;

[0033] Step 1.2.3 Use formula (7) to construct the grid reliability index of the jth sample scenario :

[0034] (7)

[0035] In formula (7), Indicates the user of the jth sample scenario in the time period The average power outage time under Indicates the jth sample scene in the time period The number of periods under

[0036] Step 1.2.4 Use formula (8) to construct the distribution automation coverage index of the jth sample scenario :

[0037] (8)

[0038] In formula (8), represents the number of automated devices in the jth sample scene, represents the total number of devices in the jth sample scene;

[0039] Step 1.2.5 Use formula (9) to construct the line contact rate index of the jth sample scenario :

[0040] (9)

[0041] In formula (9), represents the total length of the line with tie switches in the jth sample scenario, represents the total length of the inner line of the jth sample scene;

[0042] Step 1.2.6 Use formula (10) to construct the line overload ratio index of the jth sample scenario :

[0043] (10)

[0044] In formula (10), represents the number of overloaded lines in the jth sample scenario, Indicates the total number of lines of different calibers in the jth sample scene;

[0045] Step 1.2.7 Use equation (11) to construct the single radiation line ratio index of the jth sample scenario: :

[0046] (11)

[0047] In formula (11), Indicates the number of radial lines in the power grid that rely on a single power source for power supply in the jth sample scenario.

[0048] Furthermore, the step 1.4 is performed as follows:

[0049] Step 1.3.1 Use formula (12) to construct the load growth rate index of the jth sample scenario :

[0050] (12)

[0051] In formula (12), Indicates the jth sample scene in y time period Any time period before The highest power load under Indicates the jth sample scene in the time period The highest power load under

[0052] Step 1.3.2 Use formula (13) to construct the distribution transformer capacity load ratio index of the jth sample scenario :

[0053] (13)

[0054] In formula (13), represents the capacity of the distribution transformer in the jth sample scenario, represents the load on the distribution transformer in the jth sample scenario;

[0055] Step 1.3.3 Use formula (14) to construct the 10kV per household distribution transformer capacity index for the jth sample scenario: :

[0056] (14)

[0057] In formula (14), represents the e-th 10kV distribution transformer capacity in the j-th sample scenario, represents the number of power users in the jth sample scenario; E represents the total capacity of 10kV distribution transformer;

[0058] Step 1.3.4 Use formula (15) to construct the maximum peak-to-valley difference index of the jth sample scene :

[0059] (15)

[0060] In formula (15), Indicates the jth sample scene in the time period The maximum power load on the next day d is, Indicates the jth sample scene in the time period The minimum power load on the next day d.

[0061] Furthermore, the step 1.4 is performed as follows:

[0062] Step 1.4.1 Use formula (16) to construct the number of charging piles in the charging station of the jth sample scenario: :

[0063] (16)

[0064] In formula (16), represents the total number of charging piles in the charging station of the jth sample scenario, represents the user living area of ​​the jth sample scenario, Indicates the base area;

[0065] Step 1.4.2 Use Equation (17) to construct the electric vehicle ownership rate index for the jth sample scenario: :

[0066] (17)

[0067] In formula (17), represents the number of users who own electric vehicles in the jth sample scenario, represents the number of users in the jth sample scenario.

[0068] Furthermore, the step 2 is performed as follows:

[0069] Step 2.1: Use equations (18) to (20) to normalize the sample data and obtain the normalized sample data matrix :

[0070] (18)

[0071] (19)

[0072] (20)

[0073] In formula (18)-formula (20), Represents the sample data matrix under the multi-dimensional comprehensive indicator system, represents the original data of the jth sample scenario under the i-th indicator, Express Normalized sample data, where ;

[0074] Step 2.2: Calculate the objective weight vector obtained by the entropy weight method using equations (21) to (24): :

[0075] (twenty one)

[0076] (twenty two)

[0077] (twenty three)

[0078] (twenty four)

[0079] In formula (21) to formula (24), Indicates the normalized data ratio corresponding to the jth sample scenario under the i-th indicator, represents the information entropy value corresponding to the sample data of all sample scenarios under the i-th indicator, represents the objective weight value of the i-th indicator;

[0080] Step 2.3: Use equations (25) to (27) to construct the fuzzy judgment matrix :

[0081] (25)

[0082] (26)

[0083] (27)

[0084] In formula (25) to formula (27), express The i-th index and the The importance between the indicators; express Middle The importance between the first indicator and the i-th indicator;

[0085] Step 2.4: Use equations (28) and (29) to construct the fuzzy consistency matrix :

[0086] (28)

[0087] (29)

[0088] In formula (28)-formula (29), represents the fuzzy consistency auxiliary quantity of the i-th indicator, Indicates the relationship between the i-th index and the Fuzzy consistency between indicators;

[0089] Step 2.5: Calculate the subjective weight vector obtained by fuzzy analytic hierarchy process using equations (30) and (31). :

[0090] (30)

[0091] (31)

[0092] In formula (30)-formula (31), represents the fuzzy difference factor, represents the subjective weight value of the i-th indicator;

[0093] Step 2.6: Calculate the combined weighted value vector using equations (32) and (33) :

[0094] (32)

[0095] (33)

[0096] In formula (32)-formula (33), Represents the combined weight value of the i-th indicator.

[0097] Furthermore, step 3 is performed as follows:

[0098] Step 3.1, define the current number of iterations as a, and initialize a=1;

[0099] Initialize the initial membership of the jth sample scene to the kth cluster under the ath iteration ;

[0100] Randomly initialize the cluster center of the kth cluster in the ath iteration ;

[0101] Step 3.2: Use equations (34) to (36) to construct the objective function for the ath iteration. :

[0102] (34)

[0103] (35)

[0104] (36)

[0105] In formula (34) to formula (36), represents the fuzzification parameter, Represents all indicator pairs in the jth sample scenario under the ath iteration The membership degree of Indicates the jth sample scene pair under the ath iteration The Euclidean distance of ; K represents the total number of clusters;

[0106] Step 3.3: Use formula (37) to obtain the membership of all indicators in the jth sample scenario to the kth cluster in the a+1th iteration: ;

[0107] (37)

[0108] In formula (37), Indicates all index pairs except cluster center in the jth sample scenario under the ath iteration The Euclidean distances of the centers of the remaining clusters;

[0109] Step 3.4: Use formula (38) to get the cluster center of the kth cluster in the a+1th iteration :

[0110] (38)

[0111] Step 3.5: If , then stop the iteration and get the cluster centers of K clusters in the a+1th iteration 、 、…、 、…、 and used as the divided sample scene cluster; otherwise, assign a+1 to a and return to 3.2 for sequential execution, where Indicates the convergence threshold.

[0112] Step 3.6: Use formula (39) to get the cluster center under the a+1th iteration Sample scene vector of And as the division result of typical distribution network scenarios:

[0113] (39)

[0114] In formula (39), Indicates that it belongs to the cluster center at the a+1th iteration The b-th sample scene, B represents the cluster center The total number of sample scenes.

[0115] An electronic device of the present invention includes a memory and a processor, and is characterized in that the memory is used to store a program that supports the processor to execute the method for dividing typical distribution network scenarios, and the processor is configured to execute the program stored in the memory.

[0116] The present invention provides a computer-readable storage medium, and the computer-readable storage medium stores a computer program, which is characterized in that when the computer program is run by a processor, the steps of the method for dividing typical distribution network scenarios are executed.

[0117] Compared with the prior art, the present invention has the following beneficial effects:

[0118] 1. The present invention constructs an index system for dividing typical scenarios of rural distribution networks, considers building an index system from four aspects: source, network, load and storage, and adopts a combined weighting method of subjective and objective weighting methods in the weighting method. This ensures that while taking into account the properties of the objective data itself, by drawing on experience, the influence of various factors of subjective weighting on the weighting results is reduced. A fuzzy C-means clustering algorithm is proposed as a method for dividing rural power grid scenarios, and the division results are representative and scientific.

[0119] 2. The present invention considers the four aspects of source, grid, load and storage in the power system as division indicators, ensuring the integrity of the division indicator system;

[0120] 3. The present invention adopts a combined weighting method of subjective and objective weighting methods to reduce the influence of various factors of subjective weighting on the weighting results. It can fully utilize the advantages of various weighting methods, avoid the limitations of a single method, and can be flexibly adjusted according to the characteristics of the evaluation object and the evaluation purpose. It has strong adaptability and practicality.

[0121] 4. The present invention proposes a fuzzy C-means clustering algorithm as a rural power grid scene division method, ensuring that these scenes can fully reflect the overall situation of the scene. BRIEF DESCRIPTION OF THE DRAWINGS

[0122] Figure 1 is a flow chart of the method of the present invention;

[0123] Figure 2 This is a schematic diagram of the typical scenario division results of the distribution network of the present invention. DETAILED DESCRIPTION

[0124] In this embodiment, a comprehensive indicator system is constructed from the four aspects of power supply, power grid, load and energy storage, and a subjective and objective combined weighting model is established to weight the indicators. A method for dividing typical scenarios of rural distribution networks based on fuzzy C-means clustering theory is proposed. The evaluation criteria are defined for the division results of the clustering algorithm to ensure that these scenarios can truly and comprehensively reflect the overall situation of rural scenarios and have a certain degree of representativeness. Figure 1 As shown in the figure, the method includes: systematically constructing a comprehensive index system for multi-dimensional division of typical rural distribution network scenarios; establishing a combined weighting method model based on the entropy weight method and the fuzzy hierarchical analysis method; proposing a fuzzy C-means clustering algorithm for rural power grid scenario division; defining evaluation criteria for the division results of the clustering algorithm to provide a basis for judging the division results; specifically, the method is carried out in the following steps:

[0125] Step 1: Construct a comprehensive indicator system for multi-dimensional classification of typical rural distribution network scenarios:

[0126] Step 1.1: Construct an indicator system for the new energy development level:

[0127] 1.1.1 Using Equation (1) to construct the total photovoltaic installed capacity index for the jth sample scenario :

[0128] (1)

[0129] In formula (1), represents the centralized photovoltaic installed capacity of the jth sample scenario, represents the distributed photovoltaic installed capacity of the jth sample scenario, and m represents the total number of sample scenarios;

[0130] 1.1.2 Using Equation (2) to construct the PV installed capacity growth rate indicator for the jth sample scenario :

[0131] (2)

[0132] In formula (2), represents the total installed capacity of photovoltaic power generation in the jth sample scenario, Indicates the jth sample scene in y time period The total installed capacity of photovoltaic power generation equipment before; y represents the time period the number of

[0133] 1.1.3 Using Equation (3) to construct the new energy penetration index for the jth sample scenario :

[0134]

[0135] In formula (3), represents the wind power installed capacity of the jth sample scenario, represents the installed capacity of biomass energy in the jth sample scenario;

[0136] 1.1.4 Using Equation (4) to construct the distributed photovoltaic carrying capacity index for the jth sample scenario :

[0137] (4)

[0138] In formula (4), represents the distributed photovoltaic output coefficient of the jth sample scenario at time t, represents the load of the jth sample scenario at time t, Indicates that the substation in the jth sample scenario provides the maximum power.

[0139] Step 1.2: Construct an indicator system for the horizontal dimension of power grid construction:

[0140] 1.2.1 Using Equation (5) to construct the power supply voltage qualification rate index for the jth sample scenario :

[0141] (5)

[0142] In formula (5), represents the unqualified operation time of the grid voltage in the jth sample scenario, represents the total operation time of the power grid in the jth sample scenario;

[0143] 1.2.2 Using Equation (6) to construct the N-1 pass rate index for the jth sample scenario :

[0144] (6)

[0145] In formula (6), represents the total number of line transformers in the jth sample scenario that meet the N-1 condition, represents the total number of line transformer elements in the jth sample scenario;

[0146] 1.2.3 Using Equation (7) to construct the grid reliability index for the jth sample scenario :

[0147] (7)

[0148] In formula (7), Indicates the user of the jth sample scenario in the time period The average power outage time under Indicates the jth sample scene in the time period The number of periods under

[0149] 1.2.4 Using Equation (8) to construct the distribution automation coverage index for the jth sample scenario :

[0150] (8)

[0151] In formula (8), represents the number of automated devices in the jth sample scene, represents the total number of devices in the jth sample scene;

[0152] 1.2.5 Using Equation (9) to construct the line contact rate index for the jth sample scenario :

[0153] (9)

[0154] In formula (9), represents the total length of the line with tie switches in the jth sample scenario, represents the total length of the inner line of the jth sample scene;

[0155] 1.2.6 Using Equation (10) to construct the line overload ratio index for the jth sample scenario :

[0156] (10)

[0157] In formula (10), represents the number of overloaded lines in the jth sample scenario, Indicates the total number of lines of different calibers in the jth sample scene;

[0158] 1.2.7 Using Equation (11) to construct the single-radiation line ratio index for the jth sample scenario :

[0159] (11)

[0160] In formula (11), Indicates the number of radial lines in the power grid that rely on a single power source for power supply in the jth sample scenario.

[0161] Step 1.3: Construct an indicator system for load level dimension:

[0162] 1.3.1 Using Equation (12) to construct the load growth rate index for the jth sample scenario :

[0163] (12)

[0164] In formula (12), Indicates the jth sample scene in y time period Any time period before The highest power load under Indicates the jth sample scene in the time period The highest power load under

[0165] 1.3.2 Using Equation (13) to construct the distribution transformer capacity load ratio index for the jth sample scenario :

[0166] (13)

[0167] In formula (13), represents the capacity of the distribution transformer in the jth sample scenario, represents the load on the distribution transformer in the jth sample scenario;

[0168] 1.3.3 Using Equation (14) to construct the 10 kV household distribution transformer capacity index for the jth sample scenario :

[0169] (14)

[0170] In formula (14), represents the e-th 10kV distribution transformer capacity in the j-th sample scenario, represents the number of power users in the jth sample scenario; E represents the total capacity of 10kV distribution transformer;

[0171] 1.3.4 Using Equation (15) to construct the maximum peak-to-valley difference index for the jth sample scene :

[0172] (15)

[0173] In formula (15), Indicates the jth sample scene in the time period The maximum power load on the next day d is, Indicates the jth sample scene in the time period The minimum power load on the next day d.

[0174] Step 1.4: Construct an indicator system for energy storage construction levels:

[0175] 1.4.1 Using Equation (16) to construct the number of charging piles in the charging station for the jth sample scenario :

[0176] (16)

[0177] In formula (16), represents the total number of charging piles in the charging station of the jth sample scenario, represents the user living area of ​​the jth sample scenario, Indicates the base area;

[0178] 1.4.2 Using Equation (17) to construct the electric vehicle ownership rate index for the jth sample scenario :

[0179] (17)

[0180] In formula (17), represents the number of users who own electric vehicles in the jth sample scenario, represents the number of users in the jth sample scenario.

[0181] Step 2: Use the entropy weight method and fuzzy analytic hierarchy process to establish a calculation combination weight matrix model:

[0182] Step 2.1: Normalize the sample data using equations (18) to (20):

[0183] (18)

[0184] (19)

[0185] (20)

[0186] In formula (18)-formula (20), represents the sample data matrix obtained through the indicator system, represents the original data of the jth sample under the i-th indicator obtained from the survey, represents the normalized sample data corresponding to the original data of the jth sample under the i-th index obtained by calculation, The normalized sample data matrix obtained through calculation; Formula (18) represents the sample data matrix under the corresponding indicators of each county sample obtained from the survey; Formula (19) represents the normalization processing of the sample data of each county obtained from the survey; Formula (20) represents the sample data matrix after normalization processing.

[0187] Step 2.2: Use equations (21) to (24) to calculate the objective weight value obtained by the entropy weight method:

[0188] (twenty one)

[0189] (twenty two)

[0190] (twenty three)

[0191] (twenty four)

[0192] In formula (21) to formula (24), Indicates the normalized data ratio corresponding to the jth sample under the i-th indicator, Represents the information entropy value corresponding to the data under the i-th indicator, if Then the corresponding Also 0, It represents the objective weight value calculated by the entropy weight method corresponding to the i-th indicator, and is expressed as The form of the calculation objective weight value is a vector; Formula (21) represents the calculation of the proportion of the index value of the i-th sample under the j-th index; Formula (22) represents the calculation of the information entropy of the j-th index; Formula (23) represents the calculation of the entropy weight of the j-th index; Formula (24) represents the calculated objective weight value of the entropy weight method in the form of a vector.

[0193] Step 2.3: Use equations (25) to (27) to construct the fuzzy judgment matrix :

[0194] (25)

[0195] (26)

[0196] (27)

[0197] In formula (25) to formula (27), represents the fuzzy judgment matrix, Represents the fuzzy judgment matrix elements, index i and index i * Compared with the importance; Formula (25) indicates that the sum of the importance of two indicators corresponding to each other in the fuzzy judgment matrix is ​​1; Formula (26) indicates that the importance of the indicator itself is always 0.5; Formula (27) represents the constructed fuzzy judgment matrix.

[0198] Table 1

[0199]

[0200] Step 2.4: Use equations (28) and (29) to construct the fuzzy consistency matrix:

[0201] (28)

[0202] (29)

[0203] In formula (28)-formula (29), represents the auxiliary element of fuzzy consistency for calculating index i, represents the fuzzy consistency matrix element, represents the fuzzy consistency matrix; Formula (28) represents the calculation of the fuzzy consistency auxiliary element To represent the fuzzy consistency matrix elements ; Formula (28) represents the fuzzy consistency matrix The elements in should satisfy the conditions; Formula (29) represents the constructed fuzzy consistency matrix.

[0204] Step 2.5: Use equations (30) and (31) to calculate the subjective weight value obtained by the fuzzy analytic hierarchy process:

[0205] (30)

[0206] (31)

[0207] In formula (30)-formula (31), represents the fuzzy difference factor, represents the subjective weight value of the i-th indicator calculated by the fuzzy analytic hierarchy process, Represents the calculation of the subjective weight value vector; Formula (30) represents the calculation of the subjective weight value expression obtained by the fuzzy hierarchical analysis method; Formula (31) represents the calculated fuzzy hierarchical analysis method subjective weight value in the form of a vector.

[0208] Step 2.6: Calculate the combined weighted value using equations (32) and (33):

[0209] (32)

[0210] (33)

[0211] In formula (32)-formula (33), represents the subjective weight value of the i-th indicator calculated by the fuzzy analytic hierarchy process, It represents the objective weight value calculated by the entropy weight method corresponding to the i-th indicator, It represents the combined weight value of the i-th indicator calculated by the combined weighting method. Represents the combined weight value vector; Formula (32) represents the combined weight value obtained by the subjective and objective weighting method calculated using the geometric mean; Formula (33) represents the combined weight value calculated in the form of a vector.

[0212] Step 3: Propose the fuzzy C-means clustering algorithm as the rural power grid scenario division model:

[0213] Step 3.1, define the current number of iterations as a, and initialize a=1;

[0214] Initialize the initial membership of the jth sample scene to the kth cluster under the ath iteration ;

[0215] Randomly initialize the cluster center of the kth cluster in the ath iteration ;

[0216] Step 3.2: Use equations (34) to (36) to construct the objective function for the ath iteration. :

[0217] (34)

[0218] (35)

[0219] (36)

[0220] In formula (34) to formula (36), represents the fuzzification parameter, Represents all indicator pairs in the jth sample scenario under the ath iteration The membership degree of Indicates the jth sample scene pair under the ath iteration The Euclidean distance between the cluster centers is 0.000, K is the total number of clusters, Equation (34) indicates that the algorithm iteratively updates the membership of the cluster center and each data point by optimizing an objective function, which is usually defined as: Equation (35) indicates the Euclidean distance between sample points, and Equation (36) indicates the nature of the membership itself, and the sum of the membership of all samples is always 1.

[0221] Step 3.3: Use formula (37) to obtain the membership of all indicators in the jth sample scenario under the a+1th iteration to the kth cluster: ;

[0222] (37)

[0223] In formula (37), Indicates all index pairs except cluster center in the jth sample scenario under the ath iteration The Euclidean distances of the centers of the remaining clusters;

[0224] Step 3.4: Use formula (38) to get the cluster center of the kth cluster in the a+1th iteration :

[0225] (38)

[0226] Step 3.5: If , then stop the iteration and get the a+1th iteration 、 、…、 、…、 as the divided sample scene cluster; otherwise, assign a+1 to a and return to 3.2 for sequential execution, where Indicates the convergence threshold.

[0227] Step 3.6: Use formula (39) to redefine the cluster center under the a+1th iteration Sample scene vector of

[0228] (39)

[0229] In formula (39), Indicates that it belongs to the cluster center at the a+1th iteration The b-th sample scene, B represents the cluster center The total number of sample scenes;

[0230] Step 3.6. Finally, a multi-dimensional classification system of typical scenarios of rural distribution networks is obtained based on four aspects: new energy development level, grid structure construction level, load level and energy storage construction level. The classification results of typical scenarios of rural distribution networks are obtained by combining the weighted method model and the fuzzy C-means clustering algorithm.

[0231] In this embodiment, an electronic device includes a memory and a processor, wherein the memory is used to store a program that supports the processor to execute the above method, and the processor is configured to execute the program stored in the memory.

[0232] In this embodiment, a computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above method are executed.

[0233] like Figure 2 The following briefly shows the partitioning results:

[0234] Category 1, yellow part: green development grid interconnection type, mainly in the northern Anhui plain scene. The northern Anhui scene is rich in light resources. The sufficient light resources provide good natural conditions for photovoltaic power generation and are suitable for the development of the photovoltaic industry. In the context of the "dual carbon" goals, the policy response to the development of photovoltaics provides a broad market space for the photovoltaic industry.

[0235] Category 2, red area: Mature load-intensive grids. This central Anhui scenario includes cities like Hefei and Chaohu. These areas have experienced rapid economic development in recent years, with accelerated industrialization and urbanization. This rapid economic growth has driven rapid development in surrounding towns and cities, significantly increasing electricity demand from businesses and residents. This has led to a continued increase in power load demand. This, coupled with a dense population, increased residential electricity demand, and frequent industrial and commercial activities, has resulted in relatively high electricity consumption, especially during peak periods such as summer and winter.

[0236] Category 3, Green: Re-electrification-enabled development. The Southern Anhui scenario is dominated by low mountains and hills, concentrated in the central and southern parts, while plains are found in the north. Overall, the terrain is higher in the south and lower in the north, with significant surface undulations in the central and southern parts. This terrain provides an essential growing environment for Southern Anhui's unique agricultural products, such as camellia and citrus. Greater agricultural development and mountainous conditions result in relatively lower load levels in the Southern Anhui scenario. Vigorously developing clean energy sources such as biomass energy offers significant long-term development potential.

[0237] Indicators are used to verify the division results to provide a reference for the quality of the division results.

Claims

1. A method for dividing typical scenarios of distribution networks based on fuzzy C-means clustering theory, characterized by: The steps are as follows: Step 1: Construct a comprehensive indicator system for multi-dimensional classification of typical distribution network scenarios; Step 1.1: Construct an indicator system for the new energy development level; Step 1.2: Construct an indicator system for the horizontal dimension of power grid construction; Step 1.3: Construct an indicator system for the load level dimension; Step 1.4: Construct an indicator system for energy storage construction level dimensions; Step 2: Obtain the original data of the sample scenarios under each indicator system, and use the entropy weight method and fuzzy hierarchical analysis method to obtain the combined weight vector ; Step 3: Use the fuzzy C-means clustering algorithm and combine it with the combined weight vector , all sample scenarios are divided, and the division results of typical distribution network scenarios are obtained.

2. A method for dividing typical scenarios of distribution networks based on fuzzy C-means clustering theory according to claim 1, characterized in that: The step 1.1 is carried out as follows: Step 1.1.1 Use formula (1) to construct the total photovoltaic installed capacity index for the jth sample scenario: : (1) In formula (1), represents the centralized photovoltaic installed capacity of the jth sample scenario, represents the distributed photovoltaic installed capacity of the jth sample scenario, and m represents the total number of sample scenarios; Step 1.1.2 Use formula (2) to construct the photovoltaic installed capacity growth rate index for the jth sample scenario: : (2) In formula (2), represents the total installed capacity of photovoltaic power generation in the jth sample scenario, Indicates the jth sample scene in y time period The total installed capacity of photovoltaic power generation equipment before; y represents the time period the number of Step 1.1.3 Use formula (3) to construct the new energy penetration index of the jth sample scenario : (3) In formula (3), represents the wind power installed capacity of the jth sample scenario, represents the installed capacity of biomass energy in the jth sample scenario; Step 1.1.4 Use formula (4) to construct the distributed photovoltaic carrying capacity index of the jth sample scenario : (4) In formula (4), represents the distributed photovoltaic output coefficient of the jth sample scenario at time t, represents the load of the jth sample scenario at time t, Indicates that the substation in the jth sample scenario provides the maximum power.

3. The method for dividing typical scenarios of distribution network based on fuzzy C-means clustering theory according to claim 2 is characterized in that: The step 1.2 is carried out as follows: Step 1.2.1 Use formula (5) to construct the power supply voltage qualification rate index of the jth sample scenario : (5) In formula (5), represents the unqualified operation time of the grid voltage in the jth sample scenario, represents the total operation time of the power grid in the jth sample scenario; Step 1.2.2 Use formula (6) to construct the N-1 pass rate index of the jth sample scenario : (6) In formula (6), represents the total number of line transformers in the jth sample scenario that meet the N-1 condition, represents the total number of line transformer elements in the jth sample scenario; Step 1.2.3 Use formula (7) to construct the grid reliability index of the jth sample scenario : (7) In formula (7), Indicates the user of the jth sample scenario in the time period The average power outage time under Indicates the jth sample scene in the time period The number of periods under Step 1.2.4 Use formula (8) to construct the distribution automation coverage index of the jth sample scenario : (8) In formula (8), represents the number of automated devices in the jth sample scene, represents the total number of devices in the jth sample scene; Step 1.2.5 Use formula (9) to construct the line contact rate index of the jth sample scenario : (9) In formula (9), represents the total length of the line with tie switches in the jth sample scenario, represents the total length of the inner line of the jth sample scene; Step 1.2.6 Use formula (10) to construct the line overload ratio index of the jth sample scenario : (10) In formula (10), represents the number of overloaded lines in the jth sample scenario, Indicates the total number of lines of different calibers in the jth sample scene; Step 1.2.7 Use equation (11) to construct the single radiation line ratio index of the jth sample scenario: : (11) In formula (11), Indicates the number of radial lines in the power grid that rely on a single power source for power supply in the jth sample scenario.

4. A method for dividing typical scenarios of distribution networks based on fuzzy C-means clustering theory according to claim 3, characterized in that: The step 1.4 is carried out as follows: Step 1.3.1 Use formula (12) to construct the load growth rate index of the jth sample scenario : (12) In formula (12), Indicates the jth sample scene in y time period Any time period before The highest power load under Indicates the jth sample scene in the time period The highest power load under Step 1.3.2 Use formula (13) to construct the distribution transformer capacity load ratio index of the jth sample scenario : (13) In formula (13), represents the capacity of the distribution transformer in the jth sample scenario, represents the load on the distribution transformer in the jth sample scenario; Step 1.3.3 Use formula (14) to construct the 10kV per household distribution transformer capacity index for the jth sample scenario: : (14) In formula (14), represents the e-th 10kV distribution transformer capacity in the j-th sample scenario, represents the number of power users in the jth sample scenario; E represents the total capacity of 10kV distribution transformer; Step 1.3.4 Use formula (15) to construct the maximum peak-to-valley difference index of the jth sample scene : (15) In formula (15), Indicates the jth sample scene in the time period The maximum power load on the next day d is, Indicates the jth sample scene in the time period The minimum power load on the next day d.

5. A method for dividing typical scenarios of distribution networks based on fuzzy C-means clustering theory according to claim 4, characterized in that: The step 1.4 is carried out as follows: Step 1.4.1 Use formula (16) to construct the number of charging piles in the charging station of the jth sample scenario: : (16) In formula (16), represents the total number of charging piles in the charging station of the jth sample scenario, represents the user living area of ​​the jth sample scenario, Indicates the base area; Step 1.4.2 Use Equation (17) to construct the electric vehicle ownership rate index for the jth sample scenario: : (17) In formula (17), represents the number of users who own electric vehicles in the jth sample scenario, represents the number of users in the jth sample scenario.

6. A method for dividing typical scenarios of distribution network based on fuzzy C-means clustering theory according to claim 5, characterized in that: The second step is carried out as follows: Step 2.1: Use equations (18) to (20) to normalize the sample data and obtain the normalized sample data matrix : (18) (19) (20) In formula (18)-formula (20), Represents the sample data matrix under the multi-dimensional comprehensive indicator system, represents the original data of the jth sample scenario under the i-th indicator, Express Normalized sample data, where ; Step 2.2: Calculate the objective weight vector obtained by the entropy weight method using equations (21) to (24): : (21) (22) (23) (24) In formula (21) to formula (24), Indicates the normalized data ratio corresponding to the jth sample scenario under the i-th indicator, represents the information entropy value corresponding to the sample data of all sample scenarios under the i-th indicator, represents the objective weight value of the i-th indicator; Step 2.3: Use equations (25) to (27) to construct the fuzzy judgment matrix : (25) (26) (27) In formula (25) to formula (27), express The i-th index and the The importance between the indicators; express Middle The importance between the first indicator and the i-th indicator; Step 2.4: Use equations (28) and (29) to construct the fuzzy consistency matrix : (28) (29) In formula (28)-formula (29), represents the fuzzy consistency auxiliary quantity of the i-th indicator, Indicates the relationship between the i-th index and the Fuzzy consistency between indicators; Step 2.5: Calculate the subjective weight vector obtained by fuzzy analytic hierarchy process using equations (30) and (31). : (30) (31) In formula (30)-formula (31), represents the fuzzy difference factor, represents the subjective weight value of the i-th indicator; Step 2.6: Calculate the combined weighted value vector using equations (32) and (33) : (32) (33) In formula (32)-formula (33), Represents the combined weight value of the i-th indicator.

7. A method for dividing typical scenarios of distribution network based on fuzzy C-means clustering theory according to claim 6, characterized in that: Described step 3 is carried out as follows: Step 3.1, define the current number of iterations as a, and initialize a=1; Initialize the initial membership of the jth sample scene to the kth cluster under the ath iteration ; Randomly initialize the cluster center of the kth cluster in the ath iteration ; Step 3.2: Use equations (34) to (36) to construct the objective function for the ath iteration. : (34) (35) (36) In formula (34) to formula (36), represents the fuzzification parameter, Represents all indicator pairs in the jth sample scenario under the ath iteration The membership degree of Indicates the jth sample scene pair under the ath iteration The Euclidean distance of ; K represents the total number of clusters; Step 3.3: Use formula (37) to obtain the membership of all indicators in the jth sample scenario to the kth cluster in the a+1th iteration: ; (37) In formula (37), Indicates all index pairs except cluster center in the jth sample scenario under the ath iteration The Euclidean distances of the centers of the remaining clusters; Step 3.4: Use formula (38) to get the cluster center of the kth cluster in the a+1th iteration : (38) Step 3.5: If , then stop the iteration and get the cluster centers of K clusters in the a+1th iteration 、 、…、 、…、 and used as the divided sample scene cluster; otherwise, assign a+1 to a and return to 3.2 for sequential execution, where represents the convergence threshold; Step 3.6: Use formula (39) to get the cluster center under the a+1th iteration Sample scene vector of And as the division result of typical distribution network scenarios: (39) In formula (39), Indicates that it belongs to the cluster center at the a+1th iteration The b-th sample scene, B represents the cluster center The total number of sample scenes.

8. An electronic device comprising a memory and a processor, characterized in that: The memory is used to store a program that supports the processor to execute the method for dividing typical distribution network scenarios according to any one of claims 1 to 7, and the processor is configured to execute the program stored in the memory.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for dividing typical scenarios of a distribution network according to any one of claims 1 to 7 are executed.

Images