A forward-backward power flow optimization calculation method, medium, and equipment for active distribution networks

Abstract

This invention discloses a forward-backward iteration power flow optimization calculation method, medium, and equipment for active distribution networks, relating to the fields of distribution network analysis, operation, and control technology. It constructs an active distribution network based on the physical connection structure of the network and hierarchically divides it into multi-layer networks. The number of iterations is predicted based on the parameters and corresponding correlation coefficients of the multi-layer network to obtain the minimum number of iterations. The PV node sensitivity impedance matrix is ​​calculated based on the attributes of all nodes in the network. After performing forward-backward iteration on the active distribution network, it is determined whether the current iteration count has reached the minimum number of iterations. If it has, the iteration stops and the calculation result is obtained based on node voltage convergence or the upper limit of the iteration count; otherwise, the iteration continues. By predicting the minimum number of iterations, employing an efficient PV node sensitivity impedance matrix construction method, and dynamically optimized relaxation factors, the efficiency and robustness of forward-backward iteration power flow calculation for active distribution networks are significantly improved while ensuring computational accuracy.

Description

Technical Field

[0001] This invention relates to the field of power distribution network analysis, operation and control technology, and in particular to a forward-backward power flow optimization calculation method, medium and equipment for active power distribution networks. Background Technology

[0002] The statements in this section are merely background information relating to this disclosure and do not necessarily constitute prior art.

[0003] Power flow calculation in the power distribution network of a nuclear power plant is fundamental for the economic and safe operation analysis, network reconfiguration, and fault handling of the distribution network. Compared with transmission networks, distribution networks exhibit characteristics such as a radial, hierarchical structure and a high impedance ratio. During power flow calculations, the Jacobian matrix can exhibit singular characteristics, leading to convergence difficulties with traditional PQ decomposition and Newton-Laurel methods. The forward-backward substitution method, however, demonstrates good adaptability to radial and high-impedance-ratio distribution networks, exhibiting good convergence and high accuracy, and has been widely applied in power flow calculations.

[0004] With the integration of distributed generation into the nuclear power plant distribution network, the traditional distribution network transforms into an active network. When calculating power flow using the forward-backward substitution method in active networks, a new approach is needed to handle PV nodes. The current mainstream solution is to derive the sensitivity impedance matrix of PV nodes using the Thevenin equivalent circuit. With the help of The reactive power output of PV nodes is corrected during the iteration process. This scheme expands the applicability of the traditional forward-backward substitution method and can effectively solve the power flow of active distribution networks.

[0005] However, in practical applications, the inventors discovered that the solution still has the following problems:

[0006] First, the PV node sensitivity impedance matrix The generation speed is slow. The dimension is equal to the number of PV nodes, and its diagonal elements PV node Sum of all branch impedances on the path to the equilibrium node, and off-diagonal elements. PV node and PV nodes The sum of impedances on the common path to the equilibrium node. Traditional methods for determining... It is necessary to inject unit reactive power into each PV node in the network, and then use forward and backward iterations to obtain the matrix elements. Each PV node needs to go through In the next forward-backward iteration, when the PV node is not at the network end, the network topology needs to be changed by adding virtual connection points to turn the PV node into an end node. The physical meaning of this method is clear, but its acquisition process significantly affects the overall computation speed.

[0007] Secondly, due to the reactive power output constraints at PV nodes, reactive power cannot be continuously injected to maintain the set voltage after reaching the limit. Therefore, the calculation program needs to convert a PV node to a PQ node when its reactive power output exceeds the limit during the iteration process. Since the reactive power output obtained through the modified equation is not linearly convergent, oscillations will occur during the iteration process, leading to false limit exceedances. Converting a node to a PQ node when a false limit exceedance occurs will cause deviations in the calculation results. Furthermore, this oscillation also causes voltage oscillations in subsequent steps. Although no impact on power flow convergence has yet been observed, it significantly increases the number of iterations.

[0008] Third, due to factors such as network size, network complexity, and convergence accuracy, the number of iterations for power flow calculation in active distribution networks is generally more than that of traditional algorithms. Traditional algorithms often do not perform special processing in the convergence check stage, and often do not set a minimum number of iterations. Each iteration traverses all nodes. For PQ nodes, it checks whether the absolute value of the last two voltage differences meets the convergence accuracy. For PV nodes, it checks whether the absolute value of the last voltage difference from the set voltage meets the convergence tolerance. The unrestricted multiple traversal process increases additional overhead and affects the overall efficiency of the algorithm. Summary of the Invention

[0009] To overcome the shortcomings of the prior art, this invention provides a forward-backward generation power flow optimization calculation method, medium, and equipment for active distribution networks. Through innovative minimum iteration count prediction, efficient PV node sensitivity impedance matrix construction method, and dynamically optimized relaxation factor, the efficiency and robustness of forward-backward generation power flow calculation for active distribution networks are significantly improved while ensuring calculation accuracy.

[0010] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions:

[0011] In a first aspect, the present invention provides a forward-backward power flow optimization calculation method for active distribution networks, comprising:

[0012] An active distribution network is constructed based on the physical connection structure of the active distribution network, and the active distribution network is layered to obtain a multi-layer network.

[0013] Based on the parameter data and corresponding correlation coefficients of the multilayer network, the number of iterations is predicted to obtain the minimum number of iterations;

[0014] Based on the attributes of all nodes in the multilayer network, calculate the PV node sensitivity impedance matrix;

[0015] Initial values ​​are assigned to the voltages of all nodes in the active distribution network. After performing a current forward shift and voltage backward shift on the active distribution network, the difference matrix between the voltage amplitude of the PV node and the set voltage is calculated. If there are elements in the difference matrix that do not meet the convergence condition, the reactive power correction is calculated, and the reactive power output value of the PV node is corrected based on the reactive power correction.

[0016] Determine whether the corrected reactive power output of the PV node has reached the limit. If so, take the limit value of the reactive power output of the PV node and then determine whether the current iteration number has reached the minimum iteration number. Otherwise, directly determine whether the current iteration number has reached the minimum iteration number.

[0017] If the target is reached, the iteration stops and the calculation result is obtained based on the node voltage convergence or the upper limit of the number of iterations. If the target is not reached, the iteration calculation continues.

[0018] In a further technical solution, the parameters include the number of levels, the number of PV nodes, and the convergence accuracy.

[0019] A further technical solution, the formula for calculating the minimum number of iterations is:

[0020]

[0021] in, Minimum number of iterations The correlation coefficient represents the number of levels. Indicates the number of levels. The correlation coefficient represents the number of PV nodes. Indicates the number of PV nodes. The correlation coefficient, representing the convergence accuracy, Indicates the convergence accuracy.

[0022] A further technical solution involves calculating the PV node sensitivity impedance matrix as follows:

[0023] Iterate through all node attributes in the active distribution network to obtain the number of PV nodes and the list of PV nodes;

[0024] Create a PV node sensitivity impedance matrix based on the number of PV nodes, and assign initial values ​​to all elements in the PV node sensitivity impedance matrix.

[0025] Based on the PV node list and the number of PV nodes, the impedance between each element in the PV node list is calculated, and the impedance value is used as the value of each element in the PV node sensitivity impedance matrix.

[0026] A further technical solution yields the following specific self-impedance of the PV node:

[0027] Assign initial values ​​to the self-impedance of the PV nodes, create a path list between the PV nodes and the balancing nodes, and add the PV nodes to the path list.

[0028] Query the node attributes of the active distribution network and backtrack upwards layer by layer to accumulate the positive sequence impedance of the branches;

[0029] Check if the node level is 1. If it is, end the backtracking; otherwise, continue backtracking and accumulating.

[0030] A further technical solution yields the PV node mutual impedance as follows: calculate the paths from one PV node and another PV node to the balancing node, calculate the intersection of the two paths, and the sum of the edge weights of the intersection is the PV node mutual impedance.

[0031] A further technical solution involves introducing a relaxation factor with the number of iterations as a variable during the iteration process, and calculating the reactive power correction amount of the PV node based on the relaxation factor.

[0032] A further technical solution is that the formula for the reactive power correction is expressed as:

[0033]

[0034] in, This represents the reactive power correction amount. Represents the sensitivity impedance matrix. Represents the difference matrix. This represents the voltage amplitude at the PV node.

[0035] In a second aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the forward-backward power flow optimization calculation method for an active distribution network as described in the first aspect.

[0036] Thirdly, the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps in the forward-backward power flow optimization calculation method for an active distribution network as described in the first aspect.

[0037] The above one or more technical solutions have the following beneficial effects:

[0038] For power flow calculation in distribution networks, based on the typical network structure of an active distribution network in a nuclear power plant's auxiliary power system and the known forward-backward substitution method for power flow calculation in active distribution networks, an optimized process for the forward-backward substitution method is proposed. This method adds a minimum iteration count prediction step after network layering and a minimum iteration count check step before determining whether node voltages meet convergence criteria. This approach fully leverages the inherent radial structure of distribution networks, improving the efficiency of generating the PV node sensitivity impedance matrix.

[0039] This invention independently introduces the process for obtaining PV node sensitivity impedance matrix, PV node self impedance, and PV node mutual impedance, enabling rapid calculation and value determination.

[0040] This invention reduces the probability of reactive power output of PV nodes oscillating during iteration by introducing a dynamic relaxation factor, thereby reducing the number of iterations, and provides a specialized dynamic range for the relaxation factor.

[0041] This invention reduces convergence check overhead and improves overall computation speed by constructing a prediction model with the lowest number of iterations. Attached Figure Description

[0042] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0043] Figure 1 This is a flowchart of a forward-backward power flow optimization calculation method for an active distribution network according to an embodiment of the present invention;

[0044] Figure 2 This is a schematic diagram of a typical active power distribution network structure according to an embodiment of the present invention;

[0045] Figure 3 This is a flowchart illustrating the minimum iteration count prediction process according to an embodiment of the present invention;

[0046] Figure 4 This is a flowchart of the PV node sensitivity impedance matrix calculation according to an embodiment of the present invention;

[0047] Figure 5 This is a flowchart of the PV node self-impedance calculation according to an embodiment of the present invention;

[0048] Figure 6 This is a flowchart of the PV node mutual impedance calculation according to an embodiment of the present invention;

[0049] Figure 7 This is a flowchart illustrating the dynamic calculation of the relaxation factor according to an embodiment of the present invention;

[0050] Figure 8 This is a diagram showing the dynamic range of relaxation factors in an embodiment of the present invention. Detailed Implementation

[0051] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0052] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0053] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0054] Example 1

[0055] like Figure 1 As shown in the figure, this embodiment discloses a forward-backward power flow optimization calculation method for active distribution networks, which includes the following steps:

[0056] S1: Construct an active distribution network based on the physical connection structure of the active distribution network, and divide the active distribution network into layers to obtain a multi-layer network;

[0057] In this embodiment, the active power distribution network includes a slack node, multiple PQ nodes (constant power load nodes), multiple PV nodes (distributed power generation access points), and branches connecting the nodes. Using the slack node as the root node, all nodes within the network are hierarchically divided according to their distance from the slack node.

[0058] like Figure 2 As shown, in a typical active distribution network structure, Grades 0-3 represent the power supply levels of the active distribution network, with Grade 0 being the top power supply level and Grade 3 being the bottom power supply level. Codes 1 and 6 represent the first-level power supply level, which is an intermediate power supply level; codes 2 and 8 represent the second-level power supply level, which is also an intermediate power supply level; codes 3 and 7 represent the second-level power supply level, which is the bottom power supply level; and codes 4, 5, 9, and 10 represent the third-level power supply level, which is also the bottom power supply level. PV represents distributed generation points, with codes 5, 9, and 10 indicating distributed generation points connected to the distribution network; code 11 represents the 0th-level power supply level, which is the top power supply level.

[0059] S2: Based on the parameter data and corresponding correlation coefficients of the multilayer network, predict the number of iterations to obtain the minimum number of iterations;

[0060] In this embodiment, the purpose of performing minimum iteration count prediction is to avoid performing convergence checks if the minimum number of iterations has not been reached, thereby reducing unnecessary overhead and improving computational efficiency. The parameter data for the multilayer network includes the number of layers, the number of PV nodes, and the convergence accuracy.

[0061] like Figure 3 As shown, the steps of the minimum iteration count prediction algorithm are as follows:

[0062] S201: Obtain the number of levels in the multi-layer network and set its correlation coefficient; specifically, obtain the number of levels in the active distribution network. And set its correlation coefficient. Correlation coefficient This refers to the first-order coefficient of the number of network layers when calculating the minimum number of iterations.

[0063] S202: Obtain the number of PV nodes in the multi-layer network and set its correlation coefficient; specifically, obtain the number of PV nodes in the active distribution network. And set its correlation coefficient. Correlation coefficient This refers to the first-order coefficient of the number of PV nodes when calculating the minimum number of iterations.

[0064] S203: Obtain the convergence accuracy and calculate its correlation coefficient; Obtain convergence accuracy Calculate its correlation coefficient Specifically, when season ;when season .

[0065] S204: Calculate the minimum number of iterations based on the data obtained above. Specifically, calculate the minimum number of iterations according to the following formula. , is represented as:

[0066]

[0067] in, Minimum number of iterations; The correlation coefficient representing the number of levels typically ranges from [value range missing]. , Indicates the number of levels; The correlation coefficient, representing the number of PV nodes, typically ranges from [value range missing]. , Indicates the number of PV nodes; The correlation coefficient represents the convergence accuracy; Indicates the convergence accuracy.

[0068] The number of iterations and network layers required for the final convergence of forward and backward power flow calculations in active distribution networks PV node count and convergence accuracy Based on the above variables, this invention constructs an iteration number prediction model, namely the minimum iteration number calculation formula. This prediction model is only related to the basic network data, is obtained once before the iteration calculation, and is not dynamically adjusted during the iteration process, so as not to increase the additional computational overhead.

[0069] S3: Calculate the PV node sensitivity impedance matrix based on the attributes of all nodes in the multilayer network;

[0070] In this embodiment, during the iteration process, the reactive power output correction of the PV node is obtained by using the difference between the iterated voltage amplitude and the set voltage amplitude, and the PV node sensitivity impedance matrix. This correction is then injected into the node, allowing it to be treated as a PQ node during the iteration process. This is crucial for the forward-backward substitution method to obtain the power flow of the active distribution network. The following section presents a weighted distance method for rapidly generating the PV node sensitivity impedance matrix, based on the node distance calculation method in weighted graph theory.

[0071] like Figure 4 As shown, the steps for generating the PV node sensitivity impedance matrix using the weighted distance method are as follows:

[0072] S301: Traverse the node attributes of all nodes in the active distribution network to obtain the number of PV nodes and the PV node list; specifically, traverse the node attributes of all nodes in the network structure of the active distribution network to find the number of PV nodes. And store the PV nodes to form a PV node list. .

[0073] S302: Create a PV node sensitivity impedance matrix based on the number of PV nodes, and assign initial values ​​to all elements in the PV node sensitivity impedance matrix; specifically, based on the number of PV nodes... create The PV node sensitivity impedance matrix is ​​given, and all elements in the matrix are initialized to 0.

[0074] S303: Based on the PV node list and the number of PV nodes, calculate the impedance between each element in the PV node list, and use the impedance value as the value of each element in the PV node sensitivity impedance matrix. Specifically, based on the obtained PV node list... and the number of PV nodes ,Pick The Middle The and the first There are elements, among which , ,when When to obtain the first The self-impedance of each element The obtained values ​​are then filled into the PV node sensitivity impedance matrix. The Line number column, when When to obtain the first The and the first The mutual impedance of each element The calculated values ​​are then filled into the PV node sensitivity impedance matrix. The Line number List and No. Line number List.

[0075] like Figure 5 As shown, the self-impedance of a PV node is a diagonal element of the PV node sensitivity-impedance matrix, and its determination is an essential step in forming the PV node sensitivity-impedance matrix. Self-impedance The specific steps for obtaining this information are as follows:

[0076] (1) Assign initial values ​​to the self-impedance of the PV nodes, create a path list between the PV nodes and the slack nodes, and add the PV nodes to the path list; specifically, set the self-impedance of the PV nodes... And create a list of paths from PV nodes to balance nodes. Add PV nodes to .

[0077] (2) Query the node attributes of the active distribution network and backtrack upwards layer by layer to accumulate the positive sequence impedance of the branches; specifically, query the parent node layer by layer and add the positive sequence impedance of the branches between the node and the parent node. Accumulate to And add the parent node name to the path list. end.

[0078] (3) Determine if the node level is 1. If yes, end the calculation; otherwise, return to step (2). Specifically, obtain the node level number. If the level number is 1, it means that the parent node of this node is a slack node, and the positive sequence impedance of the branch between it and the slack node is accumulated to... Then the backtracking ends.

[0079] like Figure 6 As shown, the mutual impedances of PV nodes are the off-diagonal elements of the PV node sensitivity impedance matrix, and their determination is an essential step in forming the PV node sensitivity impedance matrix. Mutual Impedance The specific steps for obtaining the value are as follows:

[0080] (1) Obtain PV nodes and PV nodes And assign initial values ​​to their mutual impedances; specifically, let the PV nodes... and PV nodes mutual impedance The initial value is 0.

[0081] (2) Calculate the intersection of the paths from the two PV nodes to the equilibrium node; calculate the intersection of the paths from the PV nodes respectively. PV nodes Path to the balanced node ,path and calculate the path and path intersection ;

[0082] (3) Determine the intersection Does it exist? If it does not exist, then let... If it exists, then equals intersection The sum of the positive sequence impedances of all branches between nodes.

[0083] The above technical solution allows for the determination of the PV node sensitivity impedance matrix. A weighted distance method was proposed. Specifically, it is known that... diagonal elements PV node To the equilibrium node The sum of the branch impedances on the intermediate path, and the off-diagonal elements. PV node and PV nodes To the equilibrium node The problem involves calculating the impedance along a common path, transforming the actual problem into calculating the distance between two nodes in a weighted graph in graph theory. Because of the inherent radial tree structure of a distribution network, there is only one unique path between any two nodes, eliminating the need to traverse the network to find the shortest path. Furthermore, since the preceding steps have already layered the network, the hierarchy attributes of each node and its parent node are known; only the PV nodes need to be considered. Start by backtracking from the parent node to the balanced node. Once finished, you will receive [the reward / award]. to path Based on this, using impedance as the edge weight, PV nodes With balance node The distance including edge weights is The diagonal elements. In calculating... For non-diagonal elements, first obtain the two PV nodes respectively. and With balance node Path between , , then ask and intersection Finally, the path obtained using the aforementioned method The sum of the edge weights on the top is Off-diagonal elements.

[0084] S4: Assign initial values ​​to the voltage of all nodes in the active distribution network. After performing a current forward push and voltage backward substitution on the active distribution network, calculate the difference matrix between the voltage amplitude of the PV node and the set voltage. If the difference matrix contains elements that do not meet the convergence condition, calculate the reactive power correction amount and correct the reactive power output value of the PV node based on the reactive power correction amount.

[0085] In this embodiment, initial voltage values ​​are assigned to all nodes of the active distribution network, typically the rated voltage. And assign initial values ​​to the reactive power output of all PV nodes, generally as follows: Current forward calculation starts from the last layer of the network, calculates the injected current at the end node of the branch and accumulates the current of each branch in the next level until the first layer, to obtain the branch current distribution of the entire distribution network under the current node voltage; voltage backward calculation starts from the first layer of the network, uses the branch current and branch impedance to calculate the voltage drop on the branch, calculates the voltage of the next layer node, until the last layer.

[0086] After completing one iteration (i.e., current forward and voltage backward), calculate the difference matrix between the PV node voltage amplitude and the set voltage after that iteration. .like If any element does not meet the convergence condition, the reactive power correction is calculated using the following formula, and this correction is accumulated to the reactive power output value of the PV node:

[0087]

[0088] in, This represents the reactive power correction amount. Represents the difference matrix. This represents the voltage amplitude at the PV node.

[0089] S5: Determine whether the corrected reactive power output value of the PV node has reached the limit. If so, take the limit value of the reactive power output value of the PV node and then determine whether the current iteration number has reached the minimum iteration number. Otherwise, directly determine whether the current iteration number has reached the minimum iteration number.

[0090] Determine whether the corrected reactive power output of the PV node reaches the limit. If so, the reactive power output of the PV node is set to the limit. The PV node sensitivity impedance matrix is ​​then used. Delete the relevant rows and columns, and then determine whether the current iteration count has reached the minimum iteration count; otherwise, directly determine whether the current iteration count has reached the minimum iteration count.

[0091] To avoid voltage oscillations in PV nodes after reactive power output correction during iteration, and to prevent false exceedances of limits in PV node reactive power output that could lead to calculation deviations, this invention introduces a relaxation factor and provides a dynamic adjustment algorithm with the number of iterations as the variable. The relaxation factor that is dynamically adjusted with the number of iterations is more effective than a fixed relaxation factor.

[0092] The relaxation factor, with the number of iterations as the variable, is calculated using the following formula:

[0093]

[0094] in, For the first The relaxation factor required for the next iteration, a constant. The initial value of the relaxation factor is a constant. This is the attenuation coefficient.

[0095] like Figure 7 As shown, the steps for calculating the relaxation factor with the number of iterations as the variable are as follows:

[0096] For relaxation factor and decay coefficient Assign initial values, initial values ​​for relaxation factors ;

[0097] Based on the current iteration number initial value of relaxation factor attenuation coefficient The value, expressed by the formula Calculate the relaxation factor .

[0098] To facilitate quick querying and value retrieval, this invention also provides a dynamic range for the relaxation factor. Specifically, when and The value of makes the relaxation factor fall into Figure 8 The effect is best when in the shade shown.

[0099] The following is a simple example of dynamic optimization of the relaxation factor:

[0100] Known In the During the next iteration, it contains The PV node voltage obtained by network back-substitution of PV nodes With set voltage The difference is:

[0101]

[0102] The reactive power correction for PV nodes is:

[0103] In the formula, Initial value of relaxation factor With the number of iterations The function.

[0104] make The value remains 1 throughout the iteration process, meaning no relaxation technique is used. It can be seen that both the rate of change and the magnitude of change exhibit an exponential decay trend with the increase of the number of iterations, eventually approaching 0. In other words, the probability of oscillation decreases as the number of iterations increases. The relaxation effect should also gradually tighten with the increase of the number of iterations. Based on the above observations, an optimization function for the relaxation factor is proposed. In the formula This is the attenuation coefficient.

[0105] S6: If the condition is met, stop the iteration and obtain the calculation result based on the node voltage convergence or the upper limit of the number of iterations. If the condition is not met, continue the iterative calculation.

[0106] In this embodiment, if the minimum number of iterations is reached, it is determined whether the node voltage has converged. If it has converged, the iteration stops and the iteration calculation result is output. If it has not converged, it is determined whether the upper limit of the number of iterations has been reached. If the upper limit of the number of iterations has been reached, the calculation ends and the calculation result is obtained. If the upper limit of the number of iterations has not been reached, the iteration calculation continues (returning to the current forward and voltage back substitution in S4). If the minimum number of iterations has not been reached, the iteration calculation continues (returning to the current forward and voltage back substitution in S4).

[0107] In summary, this invention proposes an optimized forward-backward substitution method for active distribution network power flow calculation. During the iteration process, a dynamic relaxation technique is introduced in the reactive power output correction stage of PV nodes, and a minimum iteration count prediction model is also proposed. Specifically, for power flow calculation in nuclear power plant auxiliary power systems, this invention proposes a rapid generation method for the PV node sensitivity impedance matrix generation stage in the forward-backward substitution method for active distribution network power flow calculation. During the iteration process, a dynamic relaxation technique is proposed in the reactive power output correction stage of PV nodes to suppress oscillations in reactive power output of PV nodes, reduce the probability of false limit violations, and reduce the number of iterations. A minimum iteration count prediction model is proposed to reduce convergence check overhead and improve the overall efficiency of the algorithm.

[0108] Compared with the existing main process of forward-backward generation power flow calculation in active distribution networks, this invention adds a minimum iteration count prediction step after network layering, and adds a minimum iteration count check step before determining whether the node voltage meets the convergence criteria. Based on the optimized main process, it proposes new methods in the following aspects: (1) In the process of obtaining the PV node sensitivity impedance matrix (1) A weighted distance method was proposed; (2) When correcting the reactive power output of PV nodes, a relaxation technique to prevent false limit exceedance was adopted, a relaxation factor was introduced, and a dynamic optimization method for the relaxation factor was proposed. (3) A minimum iteration count check was added to the main process, and a minimum iteration count prediction model was proposed.

[0109] Example 2

[0110] The purpose of this embodiment is to provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the method of Embodiment 1.

[0111] Example 3

[0112] The purpose of this embodiment is to provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the steps of the method of Embodiment 1.

[0113] The steps and methods involved in the apparatuses of Embodiments 2 and 3 above correspond to those in Embodiment 1. For specific implementation details, please refer to the relevant description section of Embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood as including any medium capable of storing, encoding, or carrying an instruction set for execution by a processor and enabling the processor to perform any of the methods in this invention.

[0114] Those skilled in the art will understand that the modules or steps of the present invention described above can be implemented using general-purpose computer devices. Optionally, they can be implemented using computer-executable program code, thereby allowing them to be stored in a storage device for execution by a computer device, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. The present invention is not limited to any particular combination of hardware and software.

[0115] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0116] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A forward-backward power flow optimization calculation method for active distribution networks, characterized in that, include: An active distribution network is constructed based on the physical connection structure of the active distribution network, and the active distribution network is layered to obtain a multi-layer network. Based on the parameter data and corresponding correlation coefficients of the multilayer network, the number of iterations is predicted to obtain the minimum number of iterations; Based on the attributes of all nodes in the multilayer network, calculate the PV node sensitivity impedance matrix; Initial values ​​are assigned to the voltages of all nodes in the active distribution network. After performing a current forward shift and voltage backward shift on the active distribution network, the difference matrix between the voltage amplitude of the PV node and the set voltage is calculated. If there are elements in the difference matrix that do not meet the convergence condition, the reactive power correction is calculated, and the reactive power output value of the PV node is corrected based on the reactive power correction. Determine whether the corrected reactive power output of the PV node has reached the limit. If so, take the limit value of the reactive power output of the PV node and then determine whether the current iteration number has reached the minimum iteration number. Otherwise, directly determine whether the current iteration number has reached the minimum iteration number. If the target is reached, the iteration is stopped and the calculation result is obtained based on the node voltage convergence or the upper limit of the number of iterations. If the target is not reached, the iteration calculation continues. The formula for calculating the minimum number of iterations is: in, Minimum number of iterations The correlation coefficient represents the number of levels. Indicates the number of levels. The correlation coefficient represents the number of PV nodes. Indicates the number of PV nodes. The correlation coefficient, representing the convergence accuracy, Indicates the convergence accuracy; The specific steps for calculating the PV node sensitivity impedance matrix are as follows: Iterate through all node attributes in the active distribution network to obtain the number of PV nodes and the list of PV nodes; Create a PV node sensitivity impedance matrix based on the number of PV nodes, and assign initial values ​​to all elements in the PV node sensitivity impedance matrix. Based on the PV node list and the number of PV nodes, the impedance between each element in the PV node list is calculated, and the impedance value is used as the value of each element in the PV node sensitivity impedance matrix. A relaxation factor with the number of iterations as the variable is introduced during the iteration process, and the reactive power correction of the PV node is calculated based on the relaxation factor.

2. The method for forward-backward power flow optimization calculation in an active distribution network as described in claim 1, characterized in that, The parameters include the number of levels, the number of PV nodes, and the convergence accuracy.

3. The method for forward-backward power flow optimization calculation in an active distribution network as described in claim 1, characterized in that, The specific self-impedance of the PV node is as follows: Assign initial values ​​to the self-impedance of the PV nodes, create a path list between the PV nodes and the balancing nodes, and add the PV nodes to the path list. Query the node attributes of the active distribution network and backtrack upwards layer by layer to accumulate the positive sequence impedance of the branches; Check if the node level is 1. If it is, end the backtracking; otherwise, continue backtracking and accumulating.

4. The forward-backward power flow optimization calculation method for active distribution networks as described in claim 1, characterized in that, The PV node mutual impedance is obtained by calculating the paths from one PV node to the balancing node and the other PV node, respectively, and then calculating the intersection of the two paths. The sum of the edge weights of the intersection is the PV node mutual impedance.

5. The method for forward-backward power flow optimization calculation in an active distribution network as described in claim 1, characterized in that, The formula for the reactive power correction is expressed as follows: in, This represents the reactive power correction amount. Represents the sensitivity impedance matrix. Represents the difference matrix. This represents the voltage amplitude at the PV node.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps in the forward-backward power flow optimization calculation method for an active distribution network as described in any one of claims 1-5.

7. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the forward-backward power flow optimization calculation method for an active distribution network as described in any one of claims 1-5.

Images