Power distribution network dynamic reconstruction method and device based on decision-oriented time division, and medium

Abstract

The invention discloses a power distribution network dynamic reconstruction method and device based on decision-oriented time division, and a medium. The method comprises the following steps: constructing a power distribution network dynamic reconstruction model; the method comprises the following steps: designing a multi-moment decision interoperability calculation method for the dynamic reconstruction problem of the power distribution network, completing decision consistency evaluation at each moment, performing single-moment reconstruction at each moment, calculating a decision optimal solution deviation value between the moments on the premise of considering the time sequence, and constructing a similarity matrix; a similarity matrix is obtained, a spectral clustering algorithm is introduced, decision consistency and time continuity are fused, a corresponding degree matrix and a Laplacian matrix are obtained, and the decision-oriented time period division method is achieved. According to the method, it is ensured that segments can be reasonably divided according to load balancing requirements, the calculation speed is increased, optimization result similarity matrixes among all the time periods are calculated, then the time periods with high similarity are aggregated into one time period through the spectral clustering algorithm, and the method can be widely applied to the technical field of power system power distribution network reconstruction.

Description

Technical Field

[0001] The present invention relates to the technical field of power system distribution network reconstruction, and in particular to a distribution network dynamic reconstruction method, equipment and medium based on decision-oriented time period division. Background Art

[0002] With the continuous access of distributed photovoltaics, the scale and randomness of distribution networks are growing rapidly, the temporal and spatial imbalance of demand-side loads is prominent, light-loaded and heavy-loaded feeders coexist, and the safety and economy of distribution network operation are difficult to guarantee. In this context, dynamic reconstruction of distribution networks has received extensive attention. It can determine the status of segment switches and tie switches, and formulate dynamic adjustment plans for the operation mode of distribution networks in future periods to solve the negative impact of uncertainty on the demand side. However, the dynamic reconstruction of distribution networks is essentially a typical mixed integer programming problem (Mixed Integer Programming, MIP), and its solution space grows exponentially with the number of distribution network switches and the period to be reconstructed. This means that with the expansion of the scale of the distribution network and the increase of the reconstruction period, the time required for solution will increase sharply, and the dynamic reconstruction of distribution networks faces huge computational challenges. Therefore, how to effectively reduce the solution time while ensuring the quality of decision-making has become a research hotspot in the current dynamic reconstruction of distribution networks.

[0003] In current research, the solution of dynamic reconstruction of distribution network can be divided into two categories: artificial intelligence algorithm and mathematical optimization method. The former is represented by heuristic algorithms such as particle swarm optimization and deep reinforcement learning. These methods can quickly search for the local optimal solution of dynamic reconstruction of distribution network. However, the solution of artificial intelligence algorithm usually has a large deviation from the optimal solution, and even there is an infeasible phenomenon, which is difficult to apply to actual engineering. In contrast, mathematical optimization method can search for the global optimal solution under the premise of strictly satisfying constraints, but this process usually takes a lot of time. In this regard, a large number of studies have discussed model simplification. In particular, these simplification works focus on studying the time division method, by merging the reconstruction period to reduce the number of variables. This matches the actual demand that the operation mode of the distribution network will not change frequently. In fact, most dynamic reconstruction modeling has an upper limit constraint on the operation mode switching. This means that the reconstruction result will naturally be divided into several periods, which is defined as the optimal period division in this application. This type of simplification method is essentially to determine the optimal period division result in advance, which is generally acceptable in engineering.

[0004] However, the current research on time period division focuses on analyzing load similarity, that is, clustering time periods according to the Euclidean space distance of loads. This application refers to it as a load-oriented time period division method. This approach often relies on a hidden assumption: the optimal reconstruction decisions at similar load moments are mutually applicable. However, this assumption is not strictly applicable to the problem of dynamic reconstruction of distribution networks. In fact, according to multi-parameter planning theory, a small change in the value of a parameter is very likely to bring about a huge change in the results of mixed integer programming. Therefore, under load-oriented time period division, there is usually a significant gap between the dynamic reconstruction results of the distribution network and the optimal reconstruction. Obviously, if a new time period division method can be found to ensure that the optimal decisions at each moment in the time period are mutually applicable, the optimality of the simplified model can be improved. Summary of the invention

[0005] In order to solve at least one of the technical problems existing in the prior art to a certain extent, the purpose of the present invention is to provide a method, device and medium for dynamic reconstruction of a distribution network based on decision-oriented time period division, so as to ensure that a more reasonable time period division result can be obtained, improve the solution speed of the dynamic reconstruction model of the distribution network, and divide the time period by a spectral clustering algorithm to reduce the number of time periods and improve the solution speed.

[0006] The first technical solution adopted by the present invention is:

[0007] A method for dynamic reconstruction of a distribution network based on decision-oriented time period division comprises the following steps:

[0008] Construct a distribution network dynamic reconstruction model that comprehensively considers network loss and load balancing index minimization;

[0009] Design a multi-time decision interoperability calculation method for the dynamic reconstruction problem of distribution network, complete the consistency evaluation of decisions at each moment, and calculate the deviation value of the optimal solution of the decision between each moment under the premise of considering the time sequence by performing single-time reconstruction at each moment, and construct a similarity matrix;

[0010] By obtaining the similarity matrix, introducing the spectral clustering algorithm, integrating decision consistency and time continuity, the corresponding degree matrix and Laplace matrix are obtained to realize the decision-oriented time period division method.

[0011] Furthermore, the objective function of the distribution network dynamic reconstruction model optimization is:

[0012]

[0013] Where T is the number of hours in the cycle of dynamic reconstruction of the distribution network; c 1,t is the network loss of the distribution network at time t; c 2,t is the load balancing index of the distribution network at time t; E1 and E2 are the network loss cost coefficient and the load balancing index weight coefficient respectively.

[0014] Furthermore, the network loss is calculated as follows:

[0015]

[0016] In the formula, is the distribution network branch set; r k is the impedance value of branch k; w k,t is the switch closing state of branch k at time t, which takes 1 when closed and 0 when open. If the branch is not equipped with a switch, it is set to the normally closed state, that is, always 1; I k,t is the current value flowing through branch k at time t;

[0017] The calculation formula of the load balancing index is:

[0018]

[0019]

[0020] In the formula, I LB,t is the load deviation value at time t, I LB,max is the global load deviation value in the initial state; I k,max is the maximum current that can flow through branch k.

[0021] Furthermore, the constraints of the distribution network dynamic reconstruction model include distributed photovoltaic output constraints, mid-segment load constraints, power flow constraints, node voltage branch current constraints, radial network constraints, and switch operation restriction constraints.

[0022] Furthermore, the multi-moment decision interoperability calculation method for the dynamic reconstruction problem of the distribution network is designed to complete the consistency evaluation of the decisions at each moment. By performing a single-moment reconstruction at each moment, the deviation value of the optimal solution of the decision between each moment is calculated under the premise of considering the time sequence, and a similarity matrix is ​​constructed, including:

[0023] definition The optimal value when the dynamic reconstruction of the distribution network in period t only considers the optimal target of load balancing index is expressed as follows:

[0024]

[0025] In the formula, g(F; t) represents the target value matrix of all feasible decisions of the distribution network at time t;

[0026] The results of reconstruction optimization at different time periods are generally different. m Substitute the switch status results of the time period into t nIn the network under the period of t, the solution is obtained according to the dynamic reconstruction model of the distribution network, but the result is obviously inferior to that under t n The optimal solution for direct reconstruction optimization in the time period; based on the consideration of time series, the normalized difference between the two is defined as t n The deviation value of the optimal solution decision in the tm period is expressed by δ(t1t2):

[0027]

[0028]

[0029]

[0030] In the formula, Indicates that after considering the timing, n Take t in the period m The value of the optimal solution decision for the time period; Indicates that at t n Take t in the period m The value of the optimal solution decision for the time period; Indicates the importance of timing;

[0031] Since in general δ(t m t n )≠δ(t n t m ), in order to facilitate subsequent analysis and the division of time periods by spectral clustering, a symmetrical similarity is constructed by summing up, which is specifically defined as follows:

[0032] w(t m ,t n )=δ(t m t n )+δ(t n t m )

[0033] w(t m ,t n ) indicates that at t m and t n Based on the similarity of decision analysis in each time period, a similarity matrix is ​​constructed according to the similarity:

[0034]

[0035] Furthermore, the spectral clustering algorithm is a clustering method based on spectral graph theory, which is divided into two stages: graph construction and graph segmentation; the graph is composed of vertices and edges, wherein the vertices are represented by sample points, which are represented as time period numbers in the dynamic reconstruction of the distribution network, and the edges represent the similarity between sample points, which are represented as the similarity of decisions in different time periods in the dynamic reconstruction of the distribution network, and clustering is to cluster vertices with high similarity, that is, to aggregate the time periods with high decision similarity.

[0036] Furthermore, the degree matrix and the Laplace matrix are obtained in the following manner:

[0037] Since the similarity matrix W is a skew-symmetric matrix, the degree matrix D is obtained according to the similarity matrix, where D is a diagonal matrix, the diagonal value is the sum of the same edge in the corresponding similarity matrix, and the other values ​​are 0;

[0038] Calculate the Laplace matrix L based on the similarity matrix and degree matrix:

[0039] L=DW

[0040] Normalize the Laplace matrix and calculate the eigenvalues ​​and eigenvectors V of the Laplace matrix L e ;

[0041] Sort the eigenvalues ​​and obtain k eigenvalues ​​and the eigenvectors corresponding to the eigenvalues ​​in order to form a matrix V = {v1, v2,, v k}, perform K-means clustering on the matrix V to obtain the corresponding classification results, that is, the corresponding time period division results.

[0042] Furthermore, the distribution network dynamic reconstruction method also includes a verification step:

[0043] The distribution network system is simulated and verified based on the time period division method, and the distribution network dynamic reconstruction model is solved to increase the speed of dynamic reconstruction of the distribution network and improve the economy and reliability of distribution network operation.

[0044] The second technical solution adopted by the present invention is:

[0045] An electronic device comprises a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement a distribution network dynamic reconstruction method based on decision-oriented time period division as described above.

[0046] The third technical solution adopted by the present invention is:

[0047] A computer-readable storage medium stores at least one instruction, at least one program, a code set or an instruction set, wherein the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by a processor to implement a distribution network dynamic reconstruction method based on decision-oriented time period division as described above.

[0048] The fourth technical solution adopted by the present invention is:

[0049] A computer program product or a computer program includes computer instructions stored in a computer-readable storage medium. A processor of a computer device can read the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the above method.

[0050] The beneficial effects of the present invention are as follows: the present invention proposes a method for dynamic reconstruction of distribution networks based on decision-oriented time period division to cluster time periods, ensure that the segments can be reasonably divided according to load balancing requirements and improve the calculation speed, and calculate the similarity matrix of the optimization results between each time period, and then use the spectral clustering algorithm to aggregate the high similarity into one time period. The present invention designs a multi-time decision interoperability calculation method for the problem of dynamic reconstruction of distribution networks, and completes the consistency evaluation of decisions at each time. At the same time, the spectral clustering algorithm is introduced to integrate decision consistency and time continuity to realize a decision-oriented time period division method. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the embodiments of the present invention or the drawings of related technical solutions in the prior art are introduced below. It should be understood that the drawings introduced below are only for the convenience of clearly describing some embodiments of the technical solutions of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0052] Figure 1 It is a schematic flow chart of a method for dynamic reconstruction of a distribution network based on decision-oriented time period division in an embodiment of the present invention;

[0053] Figure 2 It is a schematic diagram of the process of dividing time periods based on the decision-oriented spectral clustering algorithm in an embodiment of the present invention;

[0054] Figure 3 is a typical three-feeder network connection diagram in an embodiment of the present invention;

[0055] Figure 4 is a typical three-feeder network node load variation diagram in an embodiment of the present invention;

[0056] Figure 5It is a schematic diagram of the variance value of the load between feeders at different reconstruction times in different modes of a typical three-feeder network in an embodiment of the present invention. DETAILED DESCRIPTION

[0057] The embodiments of the present invention are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and are not to be construed as limitations of the present invention. For the step numbers in the following embodiments, they are only provided for the convenience of explanation, and the order between the steps is not limited in any way, and the execution order of each step in the embodiment can be adaptively adjusted according to the understanding of those skilled in the art.

[0058] In the description of the present invention, it should be understood that descriptions involving orientations, such as up, down, front, back, left, right, etc., and orientations or positional relationships indicated are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.

[0059] In the description of the present invention, "several" means one or more, "more" means more than two, "greater than", "less than", "exceed" etc. are understood as not including the number itself, and "above", "below", "within" etc. are understood as including the number itself. If there is a description of "first" or "second", it is only used for the purpose of distinguishing the technical features, and cannot be understood as indicating or implying the relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features.

[0060] In the description of the present invention, unless otherwise clearly defined, terms such as setting, installing, connecting, etc. should be understood in a broad sense, and technicians in the relevant technical field can reasonably determine the specific meanings of the above terms in the present invention based on the specific content of the technical solution.

[0061] Example 1

[0062] like Figure 1 As shown, this embodiment provides a method for dynamic reconstruction of a distribution network based on decision-oriented time period division, comprising the following steps:

[0063] S1: Construct a dynamic reconstruction model of distribution network that comprehensively considers network loss and load balancing index minimization.

[0064] The objective function of the distribution network dynamic reconstruction model optimization is:

[0065]

[0066] Where T is the number of hours in the cycle of dynamic reconstruction of the distribution network; c 1,t is the network loss of the distribution network at time t; c 2,t is the load balancing index of the distribution network at time t; E1 and E2 are the network loss cost coefficient and the load balancing index weight coefficient respectively.

[0067] Specifically, the calculation formula for network loss is as follows:

[0068]

[0069] In the formula, is the distribution network branch set; r k is the impedance value of branch k; w k,t is the switch closing state of branch k at time t, which takes 1 when closed and 0 when open. If the branch is not equipped with a switch, it is set to the normally closed state, that is, always 1; I k,t is the current value flowing through branch k at time t.

[0070] The calculation formula of the load balancing index is:

[0071]

[0072]

[0073] In the formula, I LB,t is the load deviation value at time t, I LB,max is the global load deviation value in the initial state; I k,max is the maximum current that can flow through branch k.

[0074] In some embodiments, the constraints of the distribution network dynamic reconstruction model include distributed photovoltaic output constraints, mid-segment load constraints, flow constraints, node voltage branch current constraints, radial network constraints, and switch operation restriction constraints.

[0075] 1) The calculation formula for the distributed photovoltaic output constraint is:

[0076]

[0077] In the formula, and They represent the actual power and available power of distributed photovoltaic node i at time t respectively; ε represents the set of distributed photovoltaic nodes.

[0078] 2) The formula for the mid-section load constraint is as follows:

[0079]

[0080]

[0081] In the formula, They represent the reduction of active / reactive load of node i at time t respectively; Indicates the power reduction rate; Represents a collection of nodes; Represents the set of nodes powered by the upper power grid.

[0082] 3) The calculation formula of the power flow constraint is:

[0083]

[0084]

[0085]

[0086]

[0087]

[0088]

[0089]

[0090] Where: P i,t and Q i,t They represent the active power and reactive power injected into node i at time t respectively; P ij,t and Q ij,t They represent the active power flow and reactive power flow from node i to node j at time t respectively; and represents the square of the current in branch ij and the square of the voltage at node i at time t; g i and b i represents the conductance and susceptance from node i to ground; w ij,t represents the switch state of node i to j at time t, 0 represents disconnection and 1 represents merging; and They represent the active power demand and reactive power demand of load node i at time t respectively; M represents a relatively large number; Represents a set of branches containing switches.

[0091] 4) The calculation formulas for the node voltage and branch current constraints are as follows:

[0092]

[0093]

[0094]

[0095] Where: Vi,t represents the voltage value of node i at time t; I ij,t represents the branch current value of branch ij at time t; I ij,t,max and I ij,t,min Represents the upper and lower limits of branch current of branch ij at time t; V i,t,max and V i,t,min Represents the upper and lower limits of the voltage at node i at time t.

[0096] 5) The calculation formula of the radial network constraint is as follows:

[0097]

[0098]

[0099]

[0100]

[0101] Where: wz ij,t ,wf ij,t is an introduced auxiliary variable, whose value is greater than or equal to 0.

[0102] 6) The calculation formula of the switch operation constraint is as follows:

[0103]

[0104] In the formula, It is a segmentation variable, which indicates the set of time periods covered by time period a; A indicates the limit on the number of actions per day.

[0105] The formula equivalent to relaxation using the big M method is as follows:

[0106]

[0107]

[0108] 1-M(1-z t )≤W t ≤Mz t ,t∈(T-1);

[0109]

[0110] Where: is an auxiliary binary variable, 0 means that the switch of branch k has no action from time t to time t+1, and 1 means that it has action; W t is an auxiliary continuous variable, indicating the number of switches that operate from time t to time t+1; tIndicates whether all switches are actuated from time t to time t+1, 1 indicates that a switch is actuated, and 0 indicates that a switch is not actuated.

[0111] S2: Design a multi-moment decision interoperability calculation method for the dynamic reconstruction problem of distribution network, complete the consistency evaluation of decisions at each moment, perform single-moment reconstruction at each moment, calculate the deviation value of the optimal solution of the decision between each moment while considering the temporal sequence, and construct a similarity matrix based on this.

[0112] Definition F t * The optimal value when the dynamic reconstruction of the distribution network in period t only considers the optimal target of load balancing index is expressed as follows:

[0113] F t * =argming(F; t)

[0114] Where g(F; t) represents the target value matrix of all feasible decisions of the distribution network at time t.

[0115] The results of reconstruction optimization at different time periods are generally different. m Substitute the switch status results of the time period into t n In the network under the period t, the solution can be obtained according to the model in the previous section, but the result is obviously inferior to that under the period t n The optimal solution for direct reconstruction optimization in the time period can be defined as the deviation value of the optimal solution decision of the tm period in the tn period based on the consideration of time series, which is expressed as δ(t1t2):

[0116]

[0117]

[0118]

[0119] In the formula, Indicates that after considering the timing, n Take t in the period m The value of the optimal solution decision for the time period; Indicates that at t n Take t in the period m The value of the optimal solution decision for the time period; Indicates the importance of timing.

[0120] Since in general δ(t m t n )≠δ(t n t m), in order to facilitate subsequent analysis and the division of time periods by spectral clustering, a symmetrical similarity is constructed by summing up, which is specifically defined as follows:

[0121] w(t m ,t n )=δ(t m t n )+δ(t n t m )

[0122] w(t m ,t n ) indicates that at t m and t n Based on the similarity of decision analysis under different time periods, it is obvious that when w(t m ,t n ) is smaller, the higher the similarity between the two periods is, and the greater the probability of obtaining the optimal solution when the same decision is taken, otherwise the similarity is smaller. Further, the similarity matrix is ​​constructed based on the similarity as follows:

[0123]

[0124] From the above formula, we can see that when n∈T, w(t n ,t n )=0, W is a skew-symmetric matrix.

[0125] S3: The similarity matrix is ​​obtained through step S2, and the spectral clustering algorithm is introduced to integrate the decision consistency and time continuity to obtain the corresponding degree matrix and Laplace matrix to realize the decision-oriented time period division method.

[0126] Spectral clustering is a clustering method based on spectral graph theory. The algorithm is divided into two stages: graph construction and graph segmentation. The graph is composed of "vertices + edges". The samples are represented by a graph, and clustering is completed by dividing the vertices in the graph. The vertices are generally represented by sample points, which are represented as time periods in the dynamic reconstruction of the distribution network. The edges represent the similarity between sample points, which are represented as the similarity of decisions in different time periods in the dynamic reconstruction of the distribution network. Clustering is to cluster vertices with high similarity, that is, to aggregate the number of time periods with high decision similarity.

[0127] The spectral clustering algorithm is divided into two stages: graph construction and graph segmentation. In a network, its node set is defined as V and its edge set is defined as E, then the graph formed by this network is G = (V, E). Spectral clustering is an algorithm based on graph theory, which can find the most suitable clustering solution in various spatial sample shapes. Unlike previous clustering methods, it is more adaptable to the division of high-dimensional data, and can also achieve good results in processing multi-class community division. When using spectral clustering, it is first required to obtain the adjacency matrix of the network graph, that is, the skew-symmetric similarity matrix W. The purpose of the algorithm is to calculate and divide the similarity matrix into different classes. The points within each class have high similarity, while the points between classes have low similarity. The spectral clustering algorithm needs to cut the graph itself to achieve the purpose of division. It has several different cutting criteria, and standardized cutting is generally used.

[0128] See also Figure 2 After calculating the similarity matrix W at different times in the dynamic reconstruction of the distribution network through the process, the graph can be cut through the following steps:

[0129] Step 1: Calculate the degree matrix D, where D is a diagonal matrix. The diagonal values ​​are the sum of the same edge in the corresponding similarity matrix, and the rest of the values ​​are 0.

[0130] Step 2: Calculate the corresponding Laplace matrix, which is calculated as follows:

[0131] L=DW

[0132] Step 3: Normalize the Laplace matrix. The calculation formula is as follows:

[0133]

[0134] Step 4: Calculate the eigenvalues ​​and eigenvectors V of the Laplacian matrix L e ;

[0135] Step 5: Take out the eigenvectors corresponding to the smallest first k eigenvalues ​​to form a matrix V = {v1,v2,,v k}, and perform K-means clustering on the matrix V to obtain the corresponding classification results, and then the corresponding time period division results can be obtained.

[0136] S4: Based on the method of dividing the time periods in step S2 and step S3, the distribution network system is simulated and verified, and the model of step S1 is solved to improve the speed of dynamic reconstruction of the distribution network and enhance the economy and reliability of the distribution network operation.

[0137] The following is a detailed explanation with reference to the accompanying drawings and specific methods. To verify the effectiveness of the method proposed in the embodiment of the present invention, a typical three-feeder network is used as an example for simulation verification, where the network wiring diagram is shown in FIG. Figure 3 shown.

[0138] The typical three-feeder network connection relationship and line impedance information are shown in Table 1:

[0139] Table 1 Typical three-feeder network connection relationship and line impedance information

[0140]

[0141] Before reconstructing the network, the load of each load point in the three-feeder network at 24 moments is generated by random numbers. The results are as follows: Figure 4 shown.

[0142] Considering that the change of the operating state of the power grid in actual operation is limited, generally controlled within six times, this paper sets the number of time segment division to 5, and sets the value of the number of actions per day A to 4 to ensure that it does not exceed this value. The results of the time segment division by the FCM algorithm and the decision spectrum clustering are shown in Table 2:

[0143] Table 2 Results of time division by different algorithms

[0144]

[0145] By dynamically reconfiguring a typical three-feeder network system, the result of dynamic reconstruction without time division is set as the optimal reconfiguration (OPR) result. By comparing no reconfiguration (NR), OPR, and load-based reconfiguration (LBR), the superiority of decision-based reconfiguration (DBR) is determined. The results are shown in Table 3:

[0146] Table 3 Typical 3-feeder network reconstruction results

[0147]

[0148]

[0149] The results show that distribution network reconstruction can effectively reduce the network loss and load balancing index of the distribution network. In terms of computational efficiency, compared with OPR, LBR and DBR under time division are more efficient, and the calculation time is shortened by more than 4000s. In terms of reconstruction objectives, both methods are weaker than the improvement effect of the optimal decision. Among them, the network loss and load deviation improvement effect of DBR reached 92.5% and 93.6% of the optimal decision, which is significantly higher than 65.8% and 42.1% of LBR. In order to more intuitively reflect the advantages of DBR in load balancing, the load variance of each feeder under different reconstruction methods is shown as follows Figure 5 As shown. Obviously, DBR and ORP have similar balancing effects, and their load variance is significantly reduced. However, due to the limitations of time period division, LBR aggravates the load variance in some time periods. Taking the interval from time 10 to time 13 as an example, both OPR and DBR divide it into one time period for optimization, and the load is significantly reduced. However, LBR divides it into two time periods, 4-11 and 12-17, respectively, resulting in an increase in the load deviation within the interval. Furthermore, according to the segmentation of the reconstruction scheme, DBR and OPR have 3 segments that are completely consistent, and the other two segments are not much different, while the LBR segmentation is completely different. This further illustrates the superiority of decision-making guidance.

[0150] Example 2

[0151] An embodiment of the present invention further provides an electronic device, the electronic device comprising a processor and a memory, the memory storing at least one instruction, at least one program, a code set or an instruction set, the at least one instruction, the at least one program, the code set or the instruction set being loaded and executed by the processor to implement the following Figure 1 and Figure 2 A distribution network dynamic reconstruction method based on decision-oriented time period division is shown.

[0152] It is understood that the memory may include a random access memory (RAM) or a read-only memory (ROM). Optionally, the memory includes a non-transitory computer-readable storage medium. The memory may be used to store instructions, programs, codes, code sets, or instruction sets. The memory may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function, instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data created according to the use of the server, etc.

[0153] The processor may include one or more processing cores. The processor uses various interfaces and lines to connect the various parts of the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory, and calling data stored in the memory. Optionally, the processor can be implemented in at least one hardware form of digital signal processing (DSP), field programmable gate array (FPGA), and programmable logic array (PLA). The processor can integrate one or a combination of a central processing unit (CPU) and a modem. Among them, the CPU mainly processes the operating system and application programs; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor, but implemented separately through a chip.

[0154] Since the electronic device is an electronic device corresponding to a method for dynamic reconstruction of a distribution network based on decision-oriented time period division in an embodiment of the present invention, and the principle of solving the problem by the electronic device is similar to that of the method, the implementation of the electronic device can refer to the implementation process of the above-mentioned method embodiment, and the repeated parts will not be repeated.

[0155] Example 3

[0156] The embodiment of the present invention further provides a computer-readable storage medium, wherein the storage medium stores at least one instruction, at least one program, code set or instruction set, and the at least one instruction, the at least one program, the code set or instruction set is loaded and executed by a processor to implement the following Figure 1 and Figure 2 A distribution network dynamic reconstruction method based on decision-oriented time period division is shown.

[0157] Those skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, and the storage medium includes a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically erasable rewritable read-only memory (EEPROM), a compact disc (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0158] Since the storage medium is a storage medium corresponding to a method for dynamic reconstruction of a distribution network based on decision-oriented time period division in an embodiment of the present invention, and the principle of solving the problem by the storage medium is similar to that of the method, the implementation of the storage medium can refer to the implementation process of the above-mentioned method embodiment, and the repeated parts will not be repeated.

[0159] Example 4

[0160] In some possible implementations, various aspects of the method of the embodiment of the present invention may also be implemented in the form of a program product, which includes a program code. When the program product is run on a computer device, the program code is used to enable the computer device to execute the steps of a distribution network dynamic reconstruction method based on decision-oriented time period division according to various exemplary embodiments of the present application described above in this specification. Among them, the executable computer program code or "code" for executing each embodiment can be written in a high-level programming language such as C, C++, C#, Smalltalk, Java, JavaScript, Visual Basic, structured query language (e.g., Transact-SQL), Perl, or in various other programming languages.

[0161] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0162] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0163] The above embodiments are only for illustrating the technical concept and features of the present invention, and their purpose is to enable ordinary technicians in the field to understand the content of the present invention and implement it accordingly, and they cannot be used to limit the protection scope of the present invention. Any equivalent changes or modifications made based on the essence of the content of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for dynamic reconstruction of distribution network based on decision-oriented time period division, characterized in that: The following steps are involved: Construct a distribution network dynamic reconstruction model that comprehensively considers network loss and load balancing index minimization; Design a multi-time decision interoperability calculation method for the dynamic reconstruction problem of distribution network, complete the consistency evaluation of decisions at each moment, and calculate the deviation value of the optimal solution of the decision between each moment under the premise of considering the time sequence by performing single-time reconstruction at each moment, and construct a similarity matrix; By obtaining the similarity matrix, introducing the spectral clustering algorithm, integrating decision consistency and time continuity, the corresponding degree matrix and Laplace matrix are obtained to realize the decision-oriented time period division method.

2. A method for dynamic reconstruction of a distribution network based on decision-oriented time period division according to claim 1, characterized in that: The objective function of the distribution network dynamic reconstruction model optimization is: Where T is the number of hours in the cycle of dynamic reconstruction of the distribution network; c 1,t is the network loss of the distribution network at time t; c 2,t is the load balancing index of the distribution network at time t; E1 and E2 are the network loss cost coefficient and the load balancing index weight coefficient respectively.

3. A method for dynamic reconstruction of distribution network based on decision-oriented time period division according to claim 1, characterized in that: The network loss is calculated as follows: In the formula, is the distribution network branch set; r k is the impedance value of branch k; w k,t is the switch closing state of branch k at time t, which takes 1 when closed and 0 when open. If the branch is not equipped with a switch, it is set to the normally closed state, that is, always 1; I k,t is the current value flowing through branch k at time t; The calculation formula of the load balancing index is: In the formula, I LB,t is the load deviation value at time t, I LB,max is the global load deviation value in the initial state; I k,max is the maximum current that can flow through branch k.

4. The method for dynamic reconstruction of distribution network based on decision-oriented time period division according to claim 1, characterized in that: The constraints of the distribution network dynamic reconstruction model include distributed photovoltaic output constraints, mid-segment load constraints, power flow constraints, node voltage branch current constraints, radial network constraints, and switch operation restriction constraints.

5. The method for dynamic reconstruction of distribution network based on decision-oriented time period division according to claim 1, characterized in that: The method for calculating the interoperability of multi-time decisions for the dynamic reconstruction problem of the distribution network is designed to complete the consistency evaluation of the decisions at each moment. By performing single-moment reconstruction at each moment, the deviation value of the optimal solution of the decision between each moment is calculated under the premise of considering the time sequence, and a similarity matrix is ​​constructed, including: Definition F t * The optimal value when the dynamic reconstruction of the distribution network in period t only considers the optimal target of load balancing index is expressed as follows: In the formula, g(F; t) represents the target value matrix of all feasible decisions of the distribution network at time t; Reconstruction optimization is performed at different time periods; if t m Substitute the switch status results of the time period into t n In the network under the period of t, the solution is obtained according to the dynamic reconstruction model of the distribution network, but the result is obviously inferior to that under t n The optimal solution for direct reconstruction optimization in the time period; based on the consideration of time series, the normalized difference between the two is defined as t n Take t in the period m The deviation value of the optimal solution decision in the time period is expressed as δ(t1|t2): In the formula, Indicates that after considering the timing, n Take t in the period m The value of the optimal solution decision for the time period; Indicates that at t n Take t in the period m The value of the optimal solution decision for the time period; Indicates the importance of timing; Since in general δ(t m |t n )≠δ(t n |t m ), in order to facilitate subsequent analysis and the division of time periods by spectral clustering, a symmetrical similarity is constructed by summing up, which is specifically defined as follows: w(t m ,t n )=δ(t m |t n )+δ(t n |t m ) w(t m ,t n ) indicates that at t m and t n Based on the similarity of decision analysis in each time period, a similarity matrix is ​​constructed according to the similarity:

6. A method for dynamic reconstruction of distribution network based on decision-oriented time period division according to claim 1, characterized in that: The spectral clustering algorithm is a clustering method based on spectral graph theory. The algorithm is divided into two stages: graph construction and graph segmentation. The graph is composed of vertices and edges. The vertices are represented by sample points, which are represented as time period numbers in the dynamic reconstruction of the distribution network. The edges represent the similarity between sample points, which are represented as the similarity of decisions in different time periods in the dynamic reconstruction of the distribution network. Clustering is to cluster vertices with high similarity, that is, to aggregate the time periods with high decision similarity.

7. The method for dynamic reconstruction of distribution network based on decision-oriented time period division according to claim 1, characterized in that: The degree matrix and Laplacian matrix are obtained by: Since the similarity matrix W is a skew-symmetric matrix, the degree matrix D is obtained according to the similarity matrix, where D is a diagonal matrix, the diagonal value is the sum of the same edge in the corresponding similarity matrix, and the other values ​​are 0; Calculate the Laplace matrix L based on the similarity matrix and degree matrix: L=DW Normalize the Laplace matrix and calculate the eigenvalues ​​and eigenvectors V of the Laplace matrix L e ; Sort the eigenvalues ​​and obtain k eigenvalues ​​and eigenvectors corresponding to the eigenvalues ​​in order to form a matrix V = {v1, v2, ..., v k }, perform K-means clustering on the matrix V to obtain the corresponding classification results, that is, the corresponding time period division results.

8. The method for dynamic reconstruction of distribution network based on decision-oriented time period division according to claim 1, characterized in that: The method for dynamic reconstruction of the distribution network also includes a verification step: The distribution network system is simulated and verified based on the time period division method, and the distribution network dynamic reconstruction model is solved to increase the speed of dynamic reconstruction of the distribution network and improve the economy and reliability of distribution network operation.

9. An electronic device, characterized in that: The electronic device includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement the method described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: The storage medium stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement the method according to any one of claims 1 to 8.

Images