Micro-grid connection path optimization method considering power supply recovery

Abstract

The invention discloses a micro-grid connection path optimization method considering power supply recovery, and the method comprises the steps: solving a constructed micro-grid connection path multi-objective optimization model through employing an NSGA-II algorithm, and obtaining a Pareto dominant solution of the micro-grid connection path multi-objective optimization model; selecting an optimal solution in a power supply recovery scene from the Pareto dominant solutions to obtain a feeder line needing to be installed and connected; and carrying out installation position optimization on the feeder line needing to be installed and connected to obtain an optimal installation position of the connection. According to the method, efficient load transfer is realized by optimizing the contact path, and the power supply reliability and the operation flexibility of the microgrid under the condition of island operation or main network power supply interruption are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of microgrid planning, and particularly relates to a method for optimizing the connection path of a microgrid considering power supply restoration. Background Art

[0002] With the rapid development of renewable energy and distributed energy resources (DERs), microgrid technology has gradually become an important part of modern power systems. A microgrid can not only supply power to a region through distributed energy resources but also has a flexible operation mode, enabling seamless switching between grid-connected and islanded modes. Therefore, in the face of large-scale power outages caused by natural disasters or power system failures, microgrid power supply restoration schemes have become one of the key means to improve the reliability and resilience of the power grid.

[0003] The core of microgrid power supply restoration lies in how to reasonably dispatch available resources in the shortest time to achieve the rapid restoration of important loads. Traditional power grid reconfiguration methods mainly focus on the restoration of large-scale power grids. However, due to its high dispersion and complex operating environment, a microgrid faces more uncertainties, such as unstable output of distributed energy resources and dynamic load changes. Therefore, the optimization of power supply restoration connections in a microgrid environment has become a current research hotspot.

[0004] In the process of microgrid power supply restoration, the reasonable selection and optimization of tie lines are crucial. Tie lines not only determine the power exchange capacity between the microgrid and the main grid and other microgrids but also affect the overall efficiency of restoration and the stability of the system. How to optimally allocate tie line resources while ensuring system stability and reliability has become an urgent problem to be solved. Summary of the Invention

[0005] To solve the deficiencies in the prior art, the present invention provides a method for optimizing the connection path of a microgrid considering power supply restoration, which realizes the efficient transfer of loads by optimizing the connection path, and improves the power supply reliability and operation flexibility of the microgrid in the case of island operation or interruption of main grid power supply.

[0006] To achieve the above object, the technical solution adopted by the present invention is: a method for optimizing the connection path of a microgrid, including: using the NSGA-II algorithm to solve the constructed multi-objective optimization model of the microgrid connection path to obtain the Pareto-optimal solutions of the multi-objective optimization model of the microgrid connection path; selecting the optimal solution in the power supply restoration scenario from the Pareto-optimal solutions to obtain the feeders that need to be installed with connections; optimizing the installation positions of the feeders that need to be installed with connections to obtain the best installation positions of the connections.

[0007] Further, the method for constructing the multi-objective optimization model of the microgrid connection path includes: collecting the operation data of the microgrid; calculating the accident load loss index, the cost index of constructing the connection line, and the power supply reliability index according to the operation data of the microgrid; and constructing the multi-objective optimization model of the microgrid connection path based on the accident load loss index, the cost index of constructing the connection line, and the power supply reliability index.

[0008] Further, the accident load loss index includes the non-fault power-off load, excluding the load that cannot be restored due to the occurrence of the fault and the circuit topology limitation, and the load that forms an unplanned island due to the existence of grid-forming distributed energy; the accident load loss index is: (1) Where represents the accident load loss index of the feeder , represents the evaluation index of the load quantity of the load with the comprehensive load priority that cannot be restored on the feeder after the fault occurs, represents the total load evaluation index in the microgrid, represents the total load evaluation index on the feeder , represents the evaluation index of the th load that is restored through the island form on the feeder , represents the number of grid-forming distributed energy installed on the feeder

[0009] Further, by assigning different weight factors to loads with different priorities and then multiplying them by the load quantities to be transferred respectively, the load evaluation matrix that comprehensively considers the load priority and the load quantity is obtained: (2) Where represents the value that comprehensively considers the load priority and the load quantity on the feeder , represents the priority of the load at the node, represents the load quantity at the node,

[0010] Further, the cost index of constructing the connection line is: (3) Where represents the cost index of constructing the connection line, represents the The cost of constructing a tie line for a feeder is related to the corresponding decision variable, which determines whether the tie line is selected or not.

[0011] Furthermore, the power supply reliability index is as follows: (4) where represents the power supply reliability index.

[0012] Furthermore, the objective function of the multi-objective optimization model for the microgrid tie line path is: (5) where the first layer represents minimizing the accident load loss index, the tie line construction cost index, and maximizing the power supply reliability, and the second layer represents considering the worst-case scenario where distributed energy is completely out of service. represents the number of feeders in the microgrid that need to install tie lines.

[0013] Furthermore, selecting the optimal solution in the power supply restoration scenario from the Pareto-optimal solutions includes: First, obtaining a set of dominant solutions from the Pareto-optimal solutions according to the tie line construction cost index and the power supply reliability index, and then obtaining an optimal solution considering the maximum restoration of the non-fault power loss load in the power supply restoration scenario from these dominant solutions according to the accident load loss index: (6) where represents a 0-1 decision variable, taking 1 means the tie line is selected, and 0 means it is not selected. represents a row vector of the accident load loss index considering the load priority for different feeders.

[0014] Furthermore, optimizing the installation location of the feeders that need to install ties includes: establishing a fault occurrence matrix , which is used to represent the impact of faults occurring in different feeders on each load point, and excluding the fault power loss load from the fault occurrence matrix, that is, the load that cannot be restored due to the fault point being in its own load section during power supply restoration. In this way, the 1 elements in the fault occurrence matrix represent the load that needs to be restored through tie connection; and excluding the branches that have no direct relationship with the load power supply from the matrix; numbering each branch and load node, and the fault occurrence matrix is: , The fault occurrence matrix The row number and column number respectively correspond to the branch number and the load node number; when the matrix element is 0, it means that the branch fault corresponding to the row number has no impact on the load or the corresponding load cannot be restored due to fault isolation, and when the element is 1, it means that the branch fault will cause the load to be interrupted; establish a connection restoration matrix , which is used to represent the non-faulty power-loss loads that are transferred through tie lines after a fault occurs; when not considering the transfer margin of the tie line, the connection restoration matrix corresponds to the tie line installation position respectively; the connection restoration matrix The 1 element in it indicates that when the corresponding branch has a fault, the load node realizes power supply restoration through the tie line for transfer; the connection restoration matrix The 0 element in it indicates that the installation of the tie line has no impact on the restoration of the load, including that the load is not disconnected from the network due to the fault, or the load is powered off but cannot be restored through this connection method, and the specific reason is judged according to the fault occurrence matrix Make a judgment; use the method of formula (2) to comprehensively consider the load priority and load quantity, and then multiply by different connection restoration matrices to obtain the impact of the installation positions of different tie lines on load restoration when faults occur at different positions, and select the tie line installation position with the strongest restoration ability.

[0015] Furthermore, optimize the installation position of the feeder that needs to be connected, including: for a feeder, the situations that require load transfer include two categories: all load transfers caused by the microgrid island operation and partial load transfers caused by local feeder faults; the load quantity that can be restored in the former case has little correlation with the specific installation position of the tie line, and in the latter case, the most serious power-off situation is that the fault at the tie line installation position causes the tie switch not to close and the downstream load cannot be restored; in this case, the installation position of the first tie line is optimized according to the microgrid island operation state to ensure that high-priority loads are restored first; and the second tie line is installed at the end of the feeder to adapt to different fault types; among them, the first tie line refers to the tie line connected to the feeder with an external connection; when a feeder needs to add tie lines to the two feeders below it at the same time, according to the first row elements of different tie line installation positions, that is, the microgrid islanding fault restoration matrix considering the tie line transfer margin, optimize the load quantities of different priorities that can achieve power supply restoration, and obtain the tie line installation position.

[0016] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: By optimizing the configuration and path selection of the microgrid tie lines, the present invention significantly improves the power supply reliability and operation flexibility of the microgrid in the case of island operation or main grid power supply interruption. By maximizing the load restoration ability and power supply reliability, the present invention effectively reduces the power-off range of loads in the non-faulty area after a fault and improves the overall power supply guarantee level of the system.

[0017] In addition, during the optimization process of the present invention, the volatility of the output of distributed energy resources (DER) is fully considered. Through the reasonable configuration of connection paths, the efficient transfer of load is realized, enhancing the adaptability of the system to uncertain conditions. At the same time, the optimization scheme takes into account the balance between the construction cost of tie lines and the operating efficiency of the system, ensuring that the optimization results have practical value in terms of economy and feasibility. By efficiently utilizing resources and achieving stable power supply, the present invention can preferentially restore critical loads and meet the power supply reliability requirements of important loads. At the same time, through the dynamic balance of energy supply and demand within the microgrid, the safety and robustness of the system are further improved, providing a reliable and efficient solution for the power supply restoration of the microgrid under complex operating scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 FIG. is a schematic diagram of the main process of a method for optimizing the connection path of a microgrid considering power supply restoration provided by an embodiment of the present invention; Figure 2 FIG. is a schematic diagram of a cross-section of a certain feeder in an embodiment of the present invention; Figure 3 is Figure 2 a schematic diagram of the connection restoration matrix of the cross-section of the feeder shown; Figure 4 FIG. is a schematic diagram of a cross-section of a certain feeder where multiple tie lines need to be installed in an embodiment of the present invention; Figure 5 FIG. is a schematic diagram of the implementation process of a method for optimizing the connection path of a microgrid considering power supply restoration provided by an embodiment of the present invention; Figure 6 FIG. is a schematic diagram of a microgrid in an embodiment of the present invention; Figure 7 is Figure 6 a schematic diagram after installing tie lines. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and should not be used to limit the protection scope of the present invention.

[0020] As Figure 1 shown, a method for optimizing the connection path of a microgrid considering power supply restoration includes: using the NSGA-II algorithm to solve the constructed multi-objective optimization model of the microgrid connection path to obtain the Pareto-optimal solutions of the multi-objective optimization model of the microgrid connection path; selecting the optimal solution in the power supply restoration scenario from the Pareto-optimal solutions to obtain the feeders that need to install connections; and optimizing the installation positions of the connections for the feeders that need to install connections to obtain the best installation positions of the connections.

[0021] Due to the fluctuating output of distributed energy resources (DER), a microgrid fault can cause a large number of loads in non-fault areas to lose power. Therefore, it is necessary to optimize interconnection paths to achieve efficient load transfer and enhance the power supply reliability of the microgrid.

[0022] This invention aims to address the issue of optimizing microgrid interconnection in power restoration scenarios, enhancing power supply reliability and operational flexibility during isolated operation or main grid outages. By optimizing the interconnection line configuration within the microgrid, it maximizes load restoration capabilities and power supply reliability while minimizing interconnection line construction costs. This invention further balances the generation and load demands of distributed energy sources (such as photovoltaics, wind power, and energy storage) within the microgrid, achieving efficient energy resource utilization and stable power supply, ensuring prioritized restoration of critical loads, and safeguarding the safe and stable operation of the overall system.

[0023] Faults at the point of common connection (PCC) or the main busbar are undoubtedly the most impactful fault types for microgrids. They can cause the microgrid to disconnect from the grid without any preparation, forming an island area, which may cause serious power shortages. In this case, the microgrid needs to reconnect to the grid through a tie line to maintain power and voltage stability. When constructing a microgrid, considering the N-1 principle, a normally open tie line connecting to other microgrids or other buses will be installed at a certain feeder. Figure 4 As shown, a tie line connects the end of a feeder in microgrid 1 to microgrid 2, a crucial means of maintaining grid stability. However, as microgrids scale, the loads carried by other feeders gradually increase, inevitably connecting more and more DERs and load nodes to the microgrid. This leads to increasingly poor stability for this single tie approach. In the event of a grid failure that shuts down the main busbar or a failure without a tie feeder, a large number of non-fault-related loads may be unable to recover. In this context, adding ties between different feeders in the same microgrid is an effective way to enhance power supply reliability. Consider adding tie lines in the following situations: (1) The types of DERs carried by the feeders are single and cannot form a complementary relationship. Consider increasing connections with other feeders to increase power supply reliability and restoration robustness. (2) The feeder carries a large load but has a small number of DER connections, resulting in a heavy reliance on the busbar for power supply. Consider increasing connections with other buses or feeders. (3) The load priority is high and higher power supply reliability is required, so consider increasing the number of contacts; (4) If the feeder does not have a DER with fault ride-through capability, an interconnecting line should be added.

[0024] The present invention realizes the collaborative optimization of tie-line selection and specific installation locations by establishing a two-layer optimization model. In the first-layer optimization, the NSGA-II algorithm is used to solve the constructed multi-objective optimization model of the microgrid tie-line path to globally optimize the configuration of the tie-line, so as to improve the overall performance of the system power supply restoration; in the second-layer optimization, combined with the fault occurrence matrix, the specific installation locations of the tie-lines are further optimized to improve the practical feasibility and implementation effect of the scheme.

[0025] To address the impact of the output uncertainty of distributed energy resources (DERs) on the system power supply restoration, the present invention introduces a max-min optimization problem into the constructed multi-objective optimization model of the microgrid tie-line path, thereby enhancing the robustness of the method to randomness and volatility. In addition, the present invention fully considers actual constraints such as the transfer margin of the microgrid and insufficient line capacity during the optimization process, and specifically optimizes the configuration and installation locations of the tie-lines to ensure that important loads can be preferentially restored during the power supply restoration process.

[0026] Compared with the existing tie-line optimization methods, the present invention pays more attention to the handling of key issues in complex microgrid operation scenarios. By comprehensively considering multiple constraint conditions and actual operation requirements, the accuracy and robustness of the optimization method are improved, providing a more efficient and reliable solution for microgrid power supply restoration.

[0027] A method for constructing a multi-objective optimization model of the microgrid tie-line path includes: (1) collecting microgrid operation data; (2) calculating the accident load loss index, the cost index of constructing the tie-line, and the power supply reliability index according to the microgrid operation data; (3) constructing a multi-objective optimization model of the microgrid tie-line path based on the accident load loss index, the cost index of constructing the tie-line, and the power supply reliability index.

[0028] Since the main purpose of adding connections is to enhance power supply reliability and achieve power supply restoration under fault conditions, the first goal to be considered is the maximum restoration of non-fault power-loss loads. The accident load loss index (CLLI) is introduced but with certain improvements. Instead of considering all power-loss loads in the index, only non-fault power-loss loads are considered, while ignoring those loads that cannot complete power supply restoration due to the occurrence of faults and circuit topology limitations, as well as those loads that are capable of forming unplanned islands due to the existence of surrounding grid-forming DERs. This can make the goal of power supply restoration clearer and more definite. At the same time, load priorities are considered within this index to ensure that high-priority loads have more advantages in power supply restoration.

[0029] Due to the existence of the second-layer optimization for the specific installation locations of the tie-lines, in the first-layer optimization, only the case where the main busbar exits operation due to a fault at the PCC is considered. In this case, the accident load loss index is considered as: (1) Among them, represents the accident load loss index of the feeder , represents the load quantity evaluation index of the comprehensive load priority that cannot be restored on the feeder after a fault occurs, represents the total load evaluation index in the microgrid, represents the feeder on the total load evaluation index, represents the feeder on the th load evaluation index that completes power supply restoration in the form of an island, represents the number of network-forming distributed energy sources installed on the feeder .

[0030] Among them, by assigning different weight factors to loads with different priorities and then multiplying by the respective load quantities to be transferred, a load evaluation matrix to be restored that comprehensively considers load priority and load quantity is obtained: (2) Among them, represents the value that combines load priority and load quantity on the feeder , represents the priority of the load at the node, represents the load quantity at the node,

[0031] Loads are classified into level 1, level 2, and level 3 loads according to their importance, and different weight factors are assigned. The specific information is shown in Table 1: Table 1 Priority and Weight Factors of Different Loads Priority and Weight Factors of Different Loads

[0032] Although the method of connecting each feeder of the microgrid by means of liaison is reliable, considering the cost of building liaison lines, it is also necessary to be restricted by the goal of minimizing the cost of building liaison lines, that is, adding the fewest liaison lines. The cost index of building liaison lines is: (3) Among them, represents the cost index of building liaison lines, represents the The cost of constructing a tie line for a feeder is the decision variable corresponding to which determines whether the tie line is selected.

[0033] Power supply reliability is also a very important evaluation index. The goal of maximizing power supply reliability means that the larger the proportion of the load connected by the tie line in the entire microgrid load, the better. This requirement is different from the maximum restoration of the non-fault power outage load. Due to the uncertainty of DER output and the randomness of fault occurrence, all loads in the microgrid should be considered in this goal. The power supply reliability index is: (4) where represents the power supply reliability index.

[0034] To ensure the robustness during actual power supply restoration, the present invention adopts the max-min optimization modeling method, that is, the tie line is selected and optimized under the worst conditions. Due to the uncertainty of DER output, except for the DER at the planned island with the ability to form an island, all other DERs are considered to be disconnected from the power grid by the island protection action after being affected by a fault, that is, the tie line selection and optimization are carried out under the condition that DERs do not output power at all during the power supply restoration stage. A multi-objective optimization model for the microgrid tie line path as shown in formula (5) is established. The objective function of the multi-objective optimization model for the microgrid tie line path is: (5) where the first layer represents minimizing the accident load loss index, the tie line construction cost index and maximizing the power supply reliability, and the second layer represents considering the worst case where distributed energy does not output power at all. represents the number of feeders in the microgrid where tie lines need to be installed.

[0035] After applying the NSGA-II algorithm to the established multi-objective optimization model for the microgrid tie line path, a set of Pareto-optimal solutions is obtained, that is, a Pareto-optimal set. The final decision to be applied should select an optimal solution from the optimal set according to specific goals. The present invention obtains a set of optimal solutions through the two objective functions of tie line construction cost and power supply reliability, and then obtains an optimal decision considering the maximum restoration of the non-fault power outage load in a power supply restoration scenario according to the accident load loss degree index.

[0036] The result obtained through algorithm optimization is a set of solution vectors, and each solution contains its own fitness value. Formula (1) gives the form of the accident load loss index evaluation matrix considering load priority. Multiplying each solution by the evaluation matrix obtains a set of evaluation indexes considering load priority and the maximum restoration of accident load. The solution corresponding to the maximum value among them is the optimal solution considering all objective functions: (6) Among them, represents a 0-1 decision variable. Taking 1 means that the tie line is selected, and 0 means it is not selected. represents a row vector of the accident load loss index considering the load priority of different feeders.

[0037] In fact, the three objective functions of tie optimization can all be key factors affecting the final decision, which depends on which objective is more important in the actual situation. Due to the scenario of power supply restoration of the considered microgrid, the maximum restoration of the accident load is used as the key factor to make the final decision. If a large number of DERs with unstable output are connected to the microgrid, then the power supply stability objective can be the key influencing factor for the final decision.

[0038] When a fault occurs on the feeder where a tie needs to be added, different fault locations will cause different load nodes to lose power, and the tie installation location will restore different load nodes. Since the load amounts and load priorities of different load nodes are different, in order to maximize the restoration of non-fault power-loss loads considering the load priority, simply imitating the "hand-in-hand" structure in the distribution network, that is, the method of installing the tie at the end of the feeder is likely to result in the inability to complete the power supply restoration of high-priority loads far from the end of the feeder due to insufficient transfer margin. Therefore, it is necessary to traverse and optimize all fault occurrence locations and tie line installation locations. However, due to the capacity limit of the tie line and the limited transfer margin, when a fault occurs at the PCC and the main bus exits operation, the tie line cannot restore all non-fault power-loss loads. Therefore, when a fault occurs at the PCC, comprehensive optimization should be carried out in combination with the restoration margin of the tie line, and the faults on the local feeder can be considered to be fully restored through the load transfer method. At the same time, due to the fact that the topological structure of the power grid will have two different types of feeders - general type feeders and feeders in the middle position, the specific tie installation locations for these two types of feeders are also different.

[0039] In addition to the fault at the PCC and the main bus exiting operation resulting in the microgrid operating in island mode, various fault forms may also occur in the microgrid. When a fault occurs on a feeder not connected to the main tie, non-fault power-loss loads may appear, and different installation locations of the tie line also have different restoration amounts for non-fault power-loss loads. As Figure 3 shown, when a fault occurs at the DG connection point and the switch operates, the first tie installation location will cause the downstream load to be unable to be transferred through the tie line, while the second installation method can achieve the restoration of non-fault power-loss loads. For this reason, a Fault Incidence Matrix (FIM) is established A, which is used to represent the impact on each load node when different feeders fail. Different from other methods, the failed power-off loads are excluded from the fault occurrence matrix, that is, the loads that cannot be restored due to the fault point being on their own load section during power supply restoration. In this way, the 1 elements in the fault occurrence matrix represent the loads that need to be restored through connection. And the branches that have no direct relation with load power supply are excluded from the matrix to increase the sparsity of the matrix. As Figure 2 shown, each branch and load node are numbered. Among them, branch 6 is the secondary bus, and any fault occurring at any position on it will affect all downstream lines.

[0040] Fault occurrence matrix is: , Fault occurrence matrix The row numbers and column numbers of which respectively correspond to the branch numbers and load node numbers; The elements in it are 0 or 1. When the matrix element is 0, it means that the branch fault corresponding to the row number has no impact on the load or the corresponding load cannot be restored due to fault isolation. When the element is 1, it means that the branch fault will cause the load to be interrupted.

[0041] At the same time, a connection restoration matrix is established, which is used to represent the non-fault power-off loads that can be transferred through the connection line after a fault occurs. As Figure 3 shown, except at the bus outlet, there are 7 positions on the feeder where connections can be installed. When not considering the transfer margin of the connection, considering the connection restoration matrices at two typical positions, the two connection restoration matrices respectively correspond to the two connection installation positions in the figure. The 1 element in the matrix means that when the corresponding branch fails, the load node can be restored through the connection line. The 0 element means that the installation of the connection line has no impact on the restoration of the load, which may be because the load has not been disconnected from the network due to the fault, or the load is powered off but cannot be restored through this connection method. The specific situation needs to be judged in combination with the fault occurrence matrix for judgment.

[0042] The loads carried by the six load nodes in the figure are different, and they also have different load priorities. Using the method of formula (2), considering their load priorities and load amounts comprehensively, and then multiplying by different tie restoration matrices, the influence of the installation positions of different tie lines on load restoration when faults occur at different positions can be obtained. The installation position of the tie line with the strongest restoration ability can be selected. If the maximum transfer capacity of the tie line is not considered, the fault restoration capabilities of the two tie installation positions for the feeder head are the same. However, for faults after L2, obviously the second installation method can restore more loads. Therefore, the second installation position is a better solution. However, in fact, after the microgrid experiences a disconnection fault, it is impossible for the tie line to fully restore the non-faulty power-off loads, that is, the first row of the tie restoration matrix cannot all be 1 elements. At the same time, the load amounts and priorities at different positions are different, which leads to significant differences in the load restoration capabilities of different tie line installation positions for different faults.

[0043] If there are two or more feeders that need to add ties on the same side of the feeder equipped with an external tie, then the feeder closer to this feeder is called the intermediate feeder. Considering the construction cost of the tie line and the interval limit at the tie installation location of the feeder, generally, the tie line heads of all feeders will not be installed on the same feeder. Instead, the intermediate feeder is used for transfer to save costs. For the intermediate feeder, there are two different positions on the line where ties are installed, which can achieve more load transfer when a local fault occurs in the feeder.

[0044] However, for a small-scale feeder with only 8 load nodes and no branches, there are 28 installation positions for the two tie lines. The situation is more complex for feeders with multi-level buses and multi-branches. For a feeder, the situations that require load transfer are nothing more than two categories: full load transfer caused by the island operation of the microgrid and partial load transfer caused by a local fault in the feeder. The load amount that can be restored in the former case has little correlation with the specific installation position of the tie line. In the latter case, the most serious situation of power loss is that a fault occurs at the tie line installation position, resulting in the tie switch not being able to close and the downstream loads not being able to be restored. In this case, in order to ensure the robustness of the microgrid against various faults, the installation position of the first tie (connected to the feeder equipped with an external tie) is optimized according to the island operation state of the microgrid, striving to restore high-priority loads first; while the second feeder focuses on the local fault situation and is installed at the end of the feeder to adapt to different fault types. This not only ensures the maximum restoration of high-priority loads when the microgrid experiences a disconnection fault but also enables the full restoration of non-faulty power-off loads during a local fault. As Figure 6As shown in the figure, if the feeder 5 needs to add connections to the two feeders below it at the same time, a connection can be added between the feeder 2 and the feeder 5, and then a connection can be added between the feeders 2 and 3. According to the first row elements of different connection installation positions (i.e., the microgrid islanding fault recovery matrix considering the connection transfer margin), the load quantities of different priorities that can achieve power supply recovery are optimized to obtain the connection installation positions connected to the feeder 2; then the tie line connected to the feeder 3 is installed at the end of the feeder to ensure power supply reliability to the greatest extent. The flowchart of the two-layer optimization algorithm is as Figure 5 shown.

[0045] The present invention models the optimization process of the tie line as a two-layer optimization model. In the first layer, the maximum recovery of the non-fault power loss load, the minimization of the tie line construction cost, the maximization of the power supply reliability, and the load priority are considered to make a choice for the installation of the tie line, and a microgrid tie line optimization method based on the non-dominated sorting genetic algorithm II (NSGA-II) is proposed. As a multi-objective optimization algorithm, the NSGA-II algorithm can optimize multiple conflicting objective functions at the same time, has excellent convergence and diversity maintenance capabilities, and is suitable for multi-objective problems in the power supply recovery process. At the selected feeder where the connection is installed, the specific installation positions of the tie line are optimized according to the fault occurrence matrix by traversing all the fault occurrence positions. After the above steps, the optimal connection installation plan in the microgrid power supply recovery scenario can be obtained.

[0046] The method of the present invention is further described below through a specific embodiment. The accident load loss degree, power supply reliability, and tie line construction cost parameters during the tie line selection optimization are shown in Table 2. The large-scale microgrid model is as Figure 6 shown. Select the typical feeders for analysis. The structure of feeder 1 in the microgrid is special, with a secondary bus at the end of the feeder. The load information of feeder 1 is shown in Table 3. The load information of feeder 2 in the microgrid is shown in Table 4.

[0047] Table 2 Simulation Parameters for Tie-Line Selection Optimization Simulation Parameters for Tie-Line Selection Optimization

[0048] Table 3 Load Levels and Priorities of Feeder 1 Load Levels and Priorities of Feeder 1

[0049] Table 4 Load Levels and Priorities of Feeder 2 Load Levels and Priorities of Feeder 2

[0050] The NSGA-II algorithm is used to optimize whether to install tie lines. The operating parameters of the microgrid are input into the algorithm, including the power supply reliability index and the construction cost of tie lines, to obtain a set of Pareto front solutions, and each solution in them cannot be surpassed by other solutions in terms of the fitness of all objectives. Then, the accident load loss index matrix considering load priorities is used to optimize the front solutions, and finally the unique optimal solution [1, 1, 0, 1] in the power supply restoration scenario is obtained.

[0051] For this optimization result, tie lines need to be added between Feeder 5 and Feeder 1, Feeder 2, and Feeder 4. For Feeder 4, due to the limitation of the line topology, there is only one location to install the tie line, so there is no need to optimize the installation location of the tie line. For Feeder 1, due to the existence of its secondary bus, the fault occurrence matrix and the tie line restoration matrix are relatively special. The feeder is modeled in the form of an undirected graph, with load nodes as points in the graph and branches with sectional switches connecting each load as edges in the graph. The first point is selected as the bus node, which is the power source of the network. By gradually removing the edges in the graph, each fault occurrence matrix is generated through connectivity checks. Then, through the traversal of the tie line installation locations and by synthesizing the non-fault power loss loads reflected in the fault occurrence matrix, the tie line restoration matrix at different tie line installation locations is obtained. It should be noted that since there are many non-fault power loss loads when the microgrid experiences a disconnection fault and full restoration cannot be achieved, only some elements in the first row of each tie line restoration matrix can be 1, and the specific positions of the 1 elements depend on the installation location of the tie line, the maximum transfer capacity of the tie line, and the load to be transferred. According to the maximum transfer margin of the tie line, the loads are searched according to the principle of breadth-first search to obtain the elements in the first row of each different tie line restoration matrix. For the elements in other positions of the matrix, they reflect the local faults of the feeder, and it can be considered that the non-fault power loss loads can be fully restored. After multiplying each tie line restoration matrix by the matrix of the load to be restored considering the comprehensive load priorities (9×1) and adding up all the elements, the restoration ability of different tie line restoration matrices for the power loss loads in different faults can be obtained, and the tie line installation location with the strongest restoration ability is selected, which completes the optimization of the tie line installation location.

[0052] For feeder 1, to simplify the simulation process, both edge (5, 6) and edge (5, 7) are added to the graph (i.e., first assume that there is no connection between the branches where CB5 (circuit breaker 5) is connected to CB6 and CB7), and then the 6th and 7th rows of the fault occurrence matrix and the connection restoration matrix are merged into one row, obtaining a set of 8×6 matrices, which are the fault occurrence matrix and the connection restoration matrix that can represent the actual network topology. Assuming the restoration margin of the tie line is 10 MW, after traversing the installation positions of the tie lines and different types of faults, the optimal installation position of the tie line considering the load priority under various faults is found to be at L4.

[0053] For feeder 2, which is a feeder located in the middle position, the installation position of the tie line connected to feeder 5 is optimized according to the above steps. Similarly, taking the maximum transfer margin provided by the tie line to feeder 2 during the microgrid islanding fault as 10 MW, it is obvious that all the non-fault power-off loads on the feeder (i.e., all the loads on the feeder) cannot be fully restored. After a breadth-first search with the restoration margin as the radius, the first row of the connection restoration matrix with the tie line installed at the end of the feeder changes from [1, 1, 1, 1, 1, 1, 1, 1, 1] to [0, 0, 0, 0, 1, 1, 1, 1, 1].

[0054] The connection restoration matrices at other positions are the same. After optimizing the matrix of the load to be restored considering the comprehensive load priority, the installation position of the tie line that can achieve the maximum restoration of high-priority loads in the power supply restoration scenario is found to be at L14, and the evaluation value (the ability to restore loads during various types of faults) is 124.8, which is higher than 110.9 when the tie line is installed at the end of the feeder. The tie line connected to feeder 4 is installed at the end of the feeder to ensure its power supply reliability to the greatest extent. The microgrid with tie lines installed is as Figure 7 shown.

[0055] The above are only the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the technical principle of the present invention, several improvements and deformations can be made, and these improvements and deformations should also be regarded as the protection scope of the present invention.

Claims

1. A method for optimizing the microgrid connection path, characterized in that, Including: Solving the multi-objective optimization model of the microgrid connection path constructed by using the NSGA-II algorithm to obtain the Pareto-optimal solutions of the multi-objective optimization model of the microgrid connection path; Selecting the optimal solution in the power supply restoration scenario from the Pareto-optimal solutions to obtain the feeders where connections need to be installed; Optimizing the installation positions of the feeders where connections need to be installed to obtain the best installation positions of the connections.

2. The microgrid connection path optimization method according to claim 1, wherein The method for constructing the multi-objective optimization model of the microgrid connection path includes: Collecting the operation data of the microgrid; Calculating the accident load loss index, the cost index of constructing connection lines, and the power supply reliability index according to the operation data of the microgrid; Constructing the multi-objective optimization model of the microgrid connection path according to the accident load loss index, the cost index of constructing connection lines, and the power supply reliability index.

3. The microgrid connection path optimization method according to claim 2, characterized in that The accident load loss index includes the non-fault power loss load, excluding the load that cannot complete power supply restoration due to the occurrence of faults and circuit topology limitations, and the load that forms an unplanned island due to the existence of network-forming distributed energy sources; the accident load loss index is: （1） Among them, represents the accident load loss index of the feeder , represents the load quantity evaluation index of the comprehensive load priority that cannot be restored on the feeder after a fault occurs, represents the total load evaluation index in the microgrid, represents the total load evaluation index of the feeder , represents the total load evaluation index of the feeder on the th load evaluation index for power supply restoration completed in the form of an island, represents the number of grid-forming distributed energy sources installed on the feeder .

4. The microgrid connection path optimization method according to claim 3, wherein By assigning different weight factors to loads with different priorities and then multiplying them by the respective load quantities to be transferred, a load restoration evaluation matrix that comprehensively considers load priority and load quantity is obtained: （2） Among them, represents the feeder with the combined value of load priority and load quantity represents the load priority at the node represents the load quantity at the node represents the feeder with the total number of load nodes 5. The microgrid connection path optimization method according to claim 4, wherein The cost index of constructing connection lines is: （3） Among them, represents the cost index for constructing the tie line, represents the cost for constructing the tie line for the th feeder, is the decision variable corresponding to and determines whether the tie line is selected.

6. The microgrid connection path optimization method according to claim 5, wherein The power supply reliability index is: （4） Among them, represents the power supply reliability index.

7. The microgrid connection path optimization method according to claim 6, wherein The objective function of the multi-objective optimization model of the microgrid connection path is: （5） Among them, the first layer represents minimizing the accident load loss index, the interconnection line construction cost index, and maximizing the power supply reliability, and the second layer represents considering the worst-case scenario where distributed energy does not output power at all. It represents the number of feeders that need to install interconnection lines in the microgrid.

8. The microgrid connection path optimization method according to claim 7, characterized in that Selecting the optimal solution in the power supply restoration scenario from the Pareto-optimal solutions includes: First, obtaining a set of dominant solutions from the Pareto-optimal solutions according to the cost index of constructing connection lines and the power supply reliability index, and then obtaining an optimal solution that maximally restores the non-fault power loss load in the power supply restoration scenario according to the accident load loss index: （6） Among them, represents a 0-1 decision variable. Taking 1 means that the tie line is selected, and taking 0 means it is not selected. represents a row vector of the accident load loss index considering the load priority of different feeders.

9. The microgrid connection path optimization method according to claim 8, wherein Optimizing the installation positions of the feeders where connections need to be installed includes: Establish a fault occurrence matrix , which is used to represent the impact of faults occurring in different feeders on each load point, and exclude the fault-powered-off loads in the fault occurrence matrix, that is, the loads that cannot be restored due to the fault point being located on its own load section during power supply restoration. In this way, the 1 elements in the fault occurrence matrix represent the loads that need to be restored through connection; and exclude the branches that have no direct relationship with load power supply from the matrix; Number each branch and load node, and the fault occurrence matrix is as follows: ， Fault occurrence matrix The row number and column number respectively correspond to the branch number and the load node number; when the matrix element is 0, it means that the branch fault corresponding to the row number has no impact on the load or the corresponding load cannot be restored due to fault isolation, and when the element is 1, it means that the branch fault will cause the load interruption; Establish a connection restoration matrix , which is used to represent the non-faulty power-loss loads that are supplied through tie lines after a fault occurs; when not considering the transfer margin of the tie line, the connection restoration matrix corresponds to the installation positions of the ties respectively; the connection restoration matrix The 1 element in it indicates that when a fault occurs in the corresponding branch, the load node realizes power supply restoration through the tie line; the 0 element in the connection restoration matrix indicates that the installation of the tie line has no effect on the restoration of the load, including that the load is not disconnected from the grid due to the fault, or the load loses power but cannot be restored through this connection method. The specific reason is judged according to the fault occurrence matrix ; Using the method of formula (2) to comprehensively consider load priority and load quantity, and then multiplying by different connection restoration matrices to obtain the influence of the installation positions of different connection lines on load restoration when faults occur at different positions, and selecting the connection installation position with the strongest restoration ability.

10. The microgrid connection path optimization method according to claim 9, wherein Optimizing the installation positions of the feeders where connections need to be installed includes: For a feeder, the situations where load transfer is required include two categories: all load transfer caused by microgrid island operation and partial load transfer caused by local feeder faults; the load quantity that can be restored in the former case has little correlation with the specific installation position of the connection line, while in the latter case, the most serious power loss situation is that the connection switch cannot be closed due to a fault at the connection line installation position, and the downstream load cannot be restored; in this case, the installation position of the first connection line is optimized according to the microgrid island operation state to ensure that high-priority loads are restored first; while the second connection line is installed at the end of the feeder to adapt to different fault types; where the first connection line refers to the connection line connected to the feeder with an external connection. When a feeder needs to add tie lines to two feeders below it simultaneously, according to the first-row elements of different tie installation positions, that is, the microgrid islanding fault recovery matrix considering the tie transfer margin, the load quantities with different priorities that can achieve power supply recovery are optimized to obtain the installation positions of the tie lines.

Images