Power distribution network station area three-phase tuning method based on source load prediction

Abstract

The invention provides a power distribution network area three-phase tuning method based on source load prediction, and relates to the technical field of power distribution networks, and the method comprises the steps: collecting the single-phase load data and grid-connected photovoltaic output data of a power distribution area user, and respectively building a Fourier GNN-based power distribution network user source load ultra-short-term prediction model; inputting the data into a power distribution network user source load ultra-short-term prediction model to obtain single-phase load and grid-connected photovoltaic output prediction data, and constructing a power distribution network station area three-phase tuning model by taking minimization of three-phase imbalance and voltage deviation as targets; inputting the single-phase load and grid-connected photovoltaic output prediction data into the power distribution network area three-phase tuning model to obtain an optimization result, and determining a power distribution network area three-phase tuning strategy according to the optimization result; according to the method, the problems of composite power quality disturbance and loss caused by three-phase imbalance are effectively solved, and the safety, reliability and economical efficiency of a power grid are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network, and in particular to a three-phase optimization method for a distribution network substation based on source-load prediction. Background Art

[0002] As a power grid that directly supplies power to ordinary user loads, the distribution network has the characteristics of complex network structure, diverse load equipment types, and multiple and wide load distribution points. Since a considerable part of the user loads have single-phase, asymmetric, unbalanced, and nonlinear power characteristics, low-voltage distribution networks generally have power quality problems such as unbalanced three-phase loads and excessive reactive power, which lead to unequal currents in each phase and non-zero neutral current. The imbalance of currents in each phase leads to increased line losses in distribution transformers and distribution lines, and reduced output of distribution transformers, which has a great impact on the safety and economic operation of the power grid. At the same time, there are a large number of single-phase sources and loads with unbalanced temporal and spatial distribution and different characteristics in the low-voltage distribution network. The large number of power electronic converters used in the low-voltage distribution network has aggravated the randomness, high-power disturbance, nonlinearity and dispersion of the distribution network, making the original power quality problems of the distribution network such as three-phase imbalance, voltage, and harmonics more prominent. In addition, unbalanced three-phase operation will cause voltage exceeding the limit, reduced transformer load capacity, increased power loss, and even endanger power supply safety. The resulting complex power quality problems will not only affect the safe operation of the distribution network, but also affect the distribution network losses, thereby affecting the safe, reliable and economical operation of the power grid. Summary of the invention

[0003] The purpose of the present invention is to provide a three-phase tuning method for distribution network substations based on source-load prediction, which can effectively deal with the composite power quality disturbance and loss problems caused by three-phase imbalance and improve the safety, reliability and economy of the power grid.

[0004] The technical solution of the present invention is:

[0005] In a first aspect, the present application provides a three-phase tuning method for a distribution network area based on source-load prediction, which comprises the following steps:

[0006] S1. Collect single-phase load data and grid-connected photovoltaic output data of users in the distribution area;

[0007] S2, through the single-phase load and grid-connected photovoltaic output data of users in the distribution station area, establish the distribution network user source-load ultra-short-term prediction model based on FourierGNN;

[0008] S3, inputting the single-phase load data and single-phase grid-connected photovoltaic output data of users in the distribution station area into the distribution network user source-load ultra-short-term prediction model to obtain single-phase load and grid-connected photovoltaic output prediction data;

[0009] S4. With the goal of minimizing three-phase imbalance and voltage deviation, a three-phase optimization model for the distribution network is constructed, and the single-phase load and grid-connected photovoltaic output forecast data are input into the three-phase optimization model for the distribution network to obtain the optimization result;

[0010] S5. Determine the three-phase optimization strategy for the distribution network substation based on the optimization results.

[0011] Furthermore, in step S2, the calculation process of establishing the distribution network user source-load ultra-short-term prediction model based on Fourier GNN includes:

[0012]

[0013] κ[i,j]=κ[ij]

[0014]

[0015] In the formula, represents discrete Fourier transform, (X*κ)[i] represents the convolution of X and κ, represents element-wise multiplication, is a complex matrix, X represents node features, κ represents Green kernel, i represents individual, j represents the individual dominating i, n represents the number of nodes, d represents the number of features, k represents the number of layers of Fourier graph neural network, and A ij represents the adjacency matrix element of the i-th row and j-th column, A is the adjacency matrix, W represents the weight, S A，W represents the Fourier graph operator FGO, FourierGNN(X, A) represents the ultra-short-term prediction model of distribution network user source and load based on Fourier GNN, σ is the activation function, b k is a complex-valued bias parameter.

[0016] Furthermore, in step S4, with the goal of minimizing three-phase imbalance and voltage deviation, the calculation process of constructing a three-phase tuning model for a distribution network substation includes:

[0017] Calculate the three-phase unbalance of the node:

[0018]

[0019]

[0020] Calculate the sum of the three-phase unbalance of the node voltage:

[0021]

[0022] Where U vf,i,t is the three-phase unbalance degree at node i at time t, V neg,i,t is the negative sequence voltage at node i at time t, V pos,i,tis the positive sequence voltage at node i at time t, α is the rotation factor, V a,i,t 、V b,i,t 、V c,i,t are the A, B, and C phase voltages at node i at time t respectively;

[0023] The voltage of node a after connecting to photovoltaic is:

[0024]

[0025] The voltage change of node a caused by the integration of photovoltaic power generation is:

[0026]

[0027] Where U a is the voltage at node a, U0 is the initial voltage, U k-1 is the voltage of node k-1, P b and Q b are the active power and reactive power consumed by the load at node b, respectively, P PVb is the active power generated by distributed photovoltaics, R k and X k are the resistance and reactance of the connection line between node k-1 and node k, respectively.

[0028] Furthermore, in step S5, the SPEA2 multi-objective optimization algorithm is used to determine the three-phase optimization strategy of the distribution network area according to the optimization result, and the calculation steps include:

[0029] Generate the initial population P t and empty external archive Let t = 0;

[0030] Calculate the population P t and external archives of populations The fitness value of the individual;

[0031] Sure like Then trim like Then put P t and The dominated solution in is added to until it equals is the maximum value of the external non-inferior solution set. If t>T, output the external file and the search was terminated;

[0032] For external archives A binary tournament method with replacement was used to decide whether individuals entered the mating pool;

[0033] For population Pt+1 Perform mutation and crossover with the mating pool, t=t+1, and repeat the above steps until the optimal solution is output.

[0034] Furthermore, the above optimal solution is equivalent to a minimum spanning tree that satisfies the objective function, and its calculation process includes:

[0035]

[0036] F(i)=D(i)+R(i)

[0037] In the formula, S(i) represents the quantitative strength value assigned to individual i in the population and external archives to represent the solutions dominated by the individual, R(i) represents the original fitness value of individual i, and P t represents the initial population, represents the external archive, j represents the individual that dominates i, D(i) represents the density value of individual i, is the distance between individual i and the kth adjacent individual in the target space, and F(i) represents the fitness value of individual i.

[0038] In a second aspect, the present application provides an electronic device, including:

[0039] A memory for storing one or more programs;

[0040] processor;

[0041] When the one or more programs are executed by the processor, a three-phase optimization method for a distribution network area based on source-load prediction as described in any one of the first aspects is implemented.

[0042] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a three-phase optimization method for a distribution network area based on source-load prediction as described in any one of the first aspects above.

[0043] Compared with the prior art, the present invention has at least the following advantages or beneficial effects:

[0044] (1) The present invention provides a three-phase optimization method for a distribution network substation based on source-load prediction. By collecting single-phase load data and grid-connected photovoltaic output data of users in the distribution substation, an ultra-short-term source-load prediction model for distribution network users based on Fourier GNN is established, which accurately fits the actual operating conditions of the substation and improves the operation of the distribution substation.

[0045] (2) The present invention takes minimizing three-phase imbalance and minimum voltage deviation as the objective function, constructs a three-phase tuning model for the distribution network substation, determines the three-phase tuning strategy for the distribution network substation, significantly reduces the three-phase imbalance of the distribution network, greatly reduces the losses of the distribution transformer and the power supply line, and effectively improves the voltage quality;

[0046] (3) The present invention adopts the SPEA2 multi-objective optimization algorithm to achieve multi-objective optimization of the distribution network with distributed power access, further reducing the losses caused by three-phase imbalance and voltage fluctuations in the distribution network, and has certain reference significance for improving problems in more distribution substations. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0048] Figure 1 This is a step diagram of a three-phase optimization method for a distribution network area based on source-load prediction according to Embodiment 1 of the present invention;

[0049] Figure 2 This is a schematic diagram of source-load grid connection according to Embodiment 1 of the present invention;

[0050] Figure 3 This is a schematic diagram of a 21-node three-phase four-wire low-voltage distribution network system according to Embodiment 2 of the present invention;

[0051] Figure 4 This is a comparison chart of the actual photovoltaic output and the predicted value according to Embodiment 2 of the present invention;

[0052] Figure 5 A comparison chart of the actual load value and the predicted load value in Example 2 of the present invention;

[0053] Figure 6 This is a schematic structural block diagram of an electronic device according to Embodiment 3 of the present invention.

[0054] Icon: 101, memory; 102, processor; 103, communication interface. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations.

[0056] Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application for which protection is sought, but merely represents selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0057] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0058] It should be noted that, in this article, the term "comprises" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprises..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0059] In the description of this application, it is also necessary to explain that, unless otherwise clearly specified and limited, the terms "disposed" and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be the internal communication of two elements. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0060] In conjunction with the accompanying drawings, some embodiments of the present application are described in detail below. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0061] Example 1

[0062] See also Figure 1 , Figure 1 The figure shows a step diagram of a three-phase tuning method for a distribution network area based on source-load prediction provided in an embodiment of the present application.

[0063] In a first aspect, the present application provides a three-phase tuning method for a distribution network area based on source-load prediction, which comprises the following steps:

[0064] S1. Collect single-phase load data and grid-connected photovoltaic output data of users in the distribution area;

[0065] S2, through the single-phase load and grid-connected photovoltaic output data of users in the distribution station area, establish the distribution network user source-load ultra-short-term prediction model based on FourierGNN;

[0066] S3, inputting the single-phase load data and single-phase grid-connected photovoltaic output data of users in the distribution station area into the distribution network user source-load ultra-short-term prediction model to obtain single-phase load and grid-connected photovoltaic output prediction data;

[0067] S4. With the goal of minimizing three-phase imbalance and voltage deviation, a three-phase optimization model for the distribution network is constructed, and the single-phase load and grid-connected photovoltaic output forecast data are input into the three-phase optimization model for the distribution network to obtain the optimization result;

[0068] S5. Determine the three-phase optimization strategy for the distribution network substation based on the optimization results.

[0069] As a preferred implementation, in step S2, the calculation process of establishing the distribution network user source load ultra-short-term prediction model based on Fourier GNN includes:

[0070] Given a graph G = (X, A) with node features and the adjacency matrix Introduce a weight matrix To obtain in:

[0071]

[0072] κ[i,j]=κ[ij],

[0073] definition: As the Fourier graph operator FGO, according to the convolution theorem:

[0074]

[0075] get:

[0076]

[0077] Then we get the convolution equation:

[0078]

[0079] By stacking multiple layers of FGO, a K-layer Fourier graph neural network can be defined. Given the node features of the graph G = (X, A) and the adjacency matrix

[0080]

[0081] In the formula, represents discrete Fourier transform, (X*κ)[i] represents the convolution of X and κ, represents element-wise multiplication, is a complex matrix, X represents node features, κ represents Green kernel, i represents individual, j represents each individual that dominates i, n is the number of nodes, d is the number of features, k is the number of layers of Fourier graph neural network, and A ij represents the adjacency matrix element of the i-th row and j-th column, A is the adjacency matrix, W represents the weight, S A，W represents the Fourier graph operator FGO, FourierGNN(X, A) represents the ultra-short-term prediction model of distribution network user source and load based on FourierGNN, σ is the activation function, b k is a complex-valued bias parameter.

[0082] As a preferred implementation, in step S4, the three-phase imbalance problem caused by photovoltaic and load access is suppressed, such as Figure 2 The source-load grid connection schematic diagram shown in the figure aims to minimize the three-phase imbalance and voltage deviation. The calculation process of constructing the three-phase optimization model of the distribution network area includes:

[0083] The optimization goal minf1 is to minimize the sum of the three-phase imbalance of the system node voltage. f1 is the sum of the three-phase imbalance of each node at all times. The three-phase imbalance of the node is calculated as follows:

[0084]

[0085]

[0086] Calculate the sum of the three-phase unbalance of the node voltage:

[0087]

[0088] Where U vf,i,t is the three-phase unbalance degree at node i at time t, V neg,i,t is the negative sequence voltage at node i at time t, V pos,i,t is the positive sequence voltage at node i at time t, α is the rotation factor, V a,i,t 、V b,i,t 、V c,i,t are the A, B, and C phase voltages at node i at time t respectively;

[0089] The voltage over-limit caused by the access of distributed photovoltaics is also an important issue affecting the safety and stability of the distribution network. The minimum voltage deviation is taken as the optimization target min f2, where f2 is the voltage change of the node. The voltage of node a after access to photovoltaics is:

[0090]

[0091] If no control measures are taken, the PV grid connection will cause the voltage at point a to increase, and the voltage change of node a caused by the integration of PV is:

[0092]

[0093] Where U a is the voltage at node a, U0 is the initial voltage, U k-1 is the voltage of node k-1, P b and Q b are the active power and reactive power consumed by the load at node b, respectively, P PVb is the active power generated by distributed photovoltaics, R k and X k are the resistance and reactance of the connection line between node k-1 and node k, respectively.

[0094] As a preferred implementation, in step S5, the SPEA2 multi-objective optimization algorithm is used to determine the three-phase optimization strategy of the distribution network area according to the optimization result, and the calculation steps include:

[0095] Generate the initial population P t and empty external archive Let t = 0;

[0096] Calculate the population P t and external archives of populations The fitness value of the individual;

[0097] Sure like Then trim like Then put P t and The dominated solution in is added to until it equals is the maximum value of the external non-inferior solution set. If t>T, output the external file and the search was terminated;

[0098] For external archives A binary tournament method with replacement was used to decide whether individuals entered the mating pool;

[0099] For population P t+1Perform mutation and crossover with the mating pool, t=t+1, and repeat the steps until the optimal solution is output.

[0100] As a preferred implementation, the optimal solution is equivalent to a minimum spanning tree that satisfies the objective function, and the calculation process includes:

[0101] In order to avoid individuals dominated by the same external archive members having the same fitness value, individual i in the population and the external archive is assigned a strength value S(i) to indicate the number of solutions dominated by the individual:

[0102]

[0103] On the basis of S(i), the original fitness value R(i) of individual i is equal to the sum of the strength values ​​of all individuals that dominate the individual, that is:

[0104]

[0105] The original fitness assignment process provides a non-inferior ranking, but non-inferior ranking alone is not enough. Density information must be introduced to distinguish individuals with the same original fitness value. SPEA2 uses the k-nearest neighbor method to calculate the density value D(i) of individual i:

[0106]

[0107] It can be obtained that the fitness value F(i) of individual i is the sum of the original fitness value and the density value:

[0108] F(i)=D(i)+R(i)

[0109] In the formula, S(i) represents the quantitative strength value assigned to individual i in the population and external archives to represent the solutions dominated by the individual, R(i) represents the original fitness value of individual i, and P t represents the initial population, represents the external archive, j represents the individual that dominates i, D(i) represents the density value of individual i, is the distance between individual i and the kth adjacent individual in the target space, and F(i) represents the fitness value of individual i.

[0110] Example 2

[0111] Embodiment 2 of the present invention adopts a 21-node three-phase four-wire low-voltage distribution network system, such as Figure 3 As shown. The line self-impedance is Z L=0.65+j0.412Ω / km, the mutual impedance is 0.01 times the self-impedance, the line length between each node is 50m, the distributed photovoltaic single-phase access, the access nodes are 4, 8, 10, 12, 13, 16, 17, 20, and phase-changing switches are installed at the photovoltaic power generation access nodes and nodes 6, 14, 15. The load and photovoltaic access phase sequence on sunny days are shown in Table 1 and Table 2. The predicted values ​​and actual values ​​of photovoltaic and load are shown in Figure 4 , Figure 5 The performance of the prediction model on the dataset is shown in Table 3.

[0112] The model is measured by the average absolute error, root mean square error, and average percentage error of photovoltaic and load. The average absolute errors of photovoltaic and load are 0.2191 and 0.0527, and the root mean square errors are 0.4894 and 0.0783, respectively. It can be seen that the predicted values ​​of photovoltaic and load are not much different from the actual values. The ultra-short-term prediction model of distribution network user source and load based on FourierGNN takes into account the spatiotemporal characteristics of photovoltaic and load, and pays attention to the time-varying dependence between variables. The prediction method in this paper is more accurately in line with the actual operation of the substation.

[0113] Table 1 Single-phase load access phase

[0114]

[0115] Table 2 Photovoltaic access phase

[0116]

[0117]

[0118] Table 3 Performance of prediction models on the dataset

[0119]

[0120] After adopting the method of the present invention, the three-phase imbalance of the system has changed significantly, and the results are shown in Table 4. Before the three-phase four-wire system was treated, the system network loss was 39.523kWh, the network loss rate was 4.986%, and the system three-phase imbalance was 553.667%. After adopting the optimization method of this article, the system network loss was reduced by 13.212kWh, and the system three-phase imbalance was reduced by 429.935%. It can be seen that the three-phase imbalance has been significantly improved, and no voltage over-limit phenomenon occurred after adopting the method of this article.

[0121] Table 4 Comparison of results before and after treatment

[0122] type Grid loss / kWh Network loss rate / % System three-phase imbalance / % Voltage over limit Before governance 39.523 4.986 553.667 Appear After governance 26.311 3.386 123.732 Not present

[0123] In summary, the present invention collects single-phase load data of users in the distribution station area and single-phase grid-connected photovoltaic output data of users in the distribution station area, and establishes an ultra-short-term prediction model of source and load of distribution network users based on Fourier GNN, which more accurately fits the actual operating conditions of the substation area. The method provided by the present invention is more effective in reducing network losses and reducing voltage imbalance, effectively improving voltage quality, reducing distribution network operating losses, and improving the operation of distribution stations.

[0124] Example 3

[0125] See also Figure 6 , Figure 6 A schematic structural block diagram of an electronic device provided in an embodiment of the present application.

[0126] An electronic device includes a memory 101, a processor 102 and a communication interface 103, wherein the memory 101, the processor 102 and the communication interface 103 are electrically connected to each other directly or indirectly to realize data transmission or interaction. For example, these components can be electrically connected to each other through one or more communication buses or signal lines. The memory 101 can be used to store software programs and modules, and the processor 102 executes various functional applications and data processing by executing the software programs and modules stored in the memory 101. The communication interface 103 can be used to communicate signaling or data with other node devices.

[0127] Among them, the memory 101 can be, but is not limited to, a random access memory (RAM), a read only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable read-only memory (EEPROM), etc.

[0128] The processor 102 may be an integrated circuit chip with signal processing capability. The processor 102 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0129] It is understood that the structure shown in the figure is only for illustration, and a three-phase tuning method for a distribution network area based on source-load prediction may also include more or fewer components than those shown in the figure, or have a different configuration than those shown in the figure. Each component shown in the figure may be implemented by hardware, software, or a combination thereof.

[0130] In the embodiments provided in the present application, it should be understood that the disclosed method can also be implemented in other ways. The embodiments described above are merely schematic, for example, the flowchart or block diagram in the accompanying drawings shows the possible implementation architecture, functions and operations of the method and computer program product according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or the flowchart, and the combination of the boxes in the block diagram and / or the flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0131] In addition, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0132] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0133] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

[0134] It will be apparent to those skilled in the art that the present application is not limited to the details of the exemplary embodiments described above, and that the present application can be implemented in other specific forms without departing from the spirit or essential features of the present application. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the present application is defined by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims be included in the present application. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A three-phase tuning method for distribution network area based on source-load prediction, characterized in that: The following steps are involved: S1. Collect single-phase load data and grid-connected photovoltaic output data of users in the distribution area; S2, through the single-phase load and grid-connected photovoltaic output data of users in the distribution station area, establish the distribution network user source-load ultra-short-term prediction model based on Fourier GNN; S3, inputting the single-phase load data and single-phase grid-connected photovoltaic output data of users in the distribution station area into the distribution network user source-load ultra-short-term prediction model to obtain single-phase load and grid-connected photovoltaic output prediction data; S4. With the goal of minimizing three-phase imbalance and voltage deviation, a three-phase optimization model for the distribution network is constructed, and the single-phase load and grid-connected photovoltaic output forecast data are input into the three-phase optimization model for the distribution network to obtain the optimization result; S5. Determine the three-phase optimization strategy for the distribution network substation based on the optimization results.

2. A three-phase tuning method for distribution network area based on source-load prediction as claimed in claim 1, characterized in that: In step S2, the calculation process of establishing the distribution network user source load ultra-short-term prediction model based on Fourier GNN includes: κ[i,j]=κ[ij] In the formula, represents discrete Fourier transform, (X*κ)[i] represents the convolution of X and κ, represents element-wise multiplication, is a complex matrix, X represents node features, κ represents Green kernel, i represents individual, j represents the individual dominating i, n represents the number of nodes, d represents the number of features, k represents the number of layers of Fourier graph neural network, and A ij represents the adjacency matrix element of the i-th row and j-th column, A is the adjacency matrix, W represents the weight, S A，W represents the Fourier graph operator FGO, FourierGNN(X, A) represents the ultra-short-term prediction model of distribution network user source and load based on Fourier GNN, σ is the activation function, b k is the complex-valued bias parameter.

3. A three-phase tuning method for a distribution network area based on source-load prediction as described in claim 1, characterized in that: In step S4, the calculation process of constructing the three-phase optimization model of the distribution network area with the goal of minimizing the three-phase imbalance and voltage deviation includes: Calculate the three-phase unbalance of the node: Calculate the sum of the three-phase unbalance of the node voltage: Where U vf,i,t is the three-phase unbalance degree at node i at time t, V neg,i,t is the negative sequence voltage at node i at time t, V pos,i,t is the positive sequence voltage at node i at time t, α is the rotation factor, V a,i,t 、V b,i,t 、V c,i,t are the A, B, and C phase voltages at node i at time t respectively; The voltage of node a after connecting to photovoltaic is: The voltage change of node a caused by the integration of photovoltaic power generation is: Where U a is the voltage at node a, U0 is the initial voltage, U k-1 is the voltage of node k-1, P b and Q b are the active power and reactive power consumed by the load at node b, respectively, P PVb is the active power generated by distributed photovoltaics, R k and X k are the resistance and reactance of the connection line between node k-1 and node k, respectively.

4. A three-phase tuning method for distribution network area based on source-load prediction as claimed in claim 1, characterized in that: In step S5, the SPEA2 multi-objective optimization algorithm is used to determine the three-phase optimization strategy of the distribution network area according to the optimization result, and the calculation steps include: Generate the initial population P t and empty external archive Let t = 0; Calculate the population P t and external archives of populations The fitness value of the individual; Sure like Then trim like Then put P t and The dominated solution in until it equals is the maximum value of the external non-inferior solution set. If t>T, output the external file and the search was terminated; For external archives A binary tournament method with replacement was used to decide whether individuals entered the mating pool; For population P t+1 Perform mutation and crossover with the mating pool, t=t+1, and repeat the above steps until the optimal solution is output.

5. A three-phase tuning method for distribution network area based on source-load prediction as claimed in claim 4, characterized in that: The optimal solution is equivalent to a minimum spanning tree that satisfies the objective function, and the calculation process includes: F(i)=D(i)+R(i) In the formula, S(i) represents the quantitative strength value assigned to individual i in the population and external archives to represent the solutions dominated by the individual, R(i) represents the original fitness value of individual i, and P t represents the initial population, represents the external archive, j represents the individual that dominates i, D(i) represents the density value of individual i, is the distance between individual i and the kth adjacent individual in the target space, and F(i) represents the fitness value of individual i.

6. An electronic device, characterized in that: include: A memory for storing one or more programs; processor; When the one or more programs are executed by the processor, a three-phase optimization method for a distribution network area based on source-load prediction as described in any one of claims 1 to 5 is implemented.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, a three-phase optimization method for a distribution network area based on source-load prediction is implemented as described in any one of claims 1-5.

Images