Power distribution network reconstruction method, system, equipment and medium

Abstract

The invention discloses a power distribution network reconstruction method, system and device and a medium, and the method comprises the steps: describing the charging behavior of an electric vehicle cluster, carrying out the management of the charging behavior through a virtual electricity price mode, constructing a charging optimization model of a distributed Lagrange relaxation algorithm, and guiding the charging behavior of an electric vehicle; constructing a multi-target dynamic reconstruction model of the power distribution network according to the operation economy, the voltage quality and the service life of the flexible intelligent switch of the power distribution network; a niche technology is introduced to improve a standard particle swarm algorithm, and the power distribution network multi-target dynamic reconstruction model is solved through the improved particle swarm algorithm to obtain an optimal solution. According to the method, on one hand, the operation cost of the power distribution network is reduced by using the flexible intelligent switch, and the voltage quality is improved by optimizing the charging behavior of the electric vehicle, and on the other hand, the solving capability of the optimization model is improved by the proposed improved algorithm, and a high-quality optimal solution is provided, so that the economy and feasibility of a reconstruction scheme are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of flexible interconnected distribution network operation optimization, and in particular to a distribution network reconstruction method, system, equipment and medium. Background Art

[0002] With the rapid development of high-proportion renewable energy, including electric vehicles (EVs), photovoltaics (PV), wind turbines (WT), and various energy storage systems (ESS), the penetration of distributed flexible resources is increasing. While this has brought many benefits to the energy mix, the uncertainty of these distributed energy resources has also had many negative impacts on the safe and stable operation of the distribution system. Distribution network reconfiguration, as a technical means of optimizing the network, can improve the reliability and economic efficiency of the power system by adjusting the structure of the distribution system, thereby mitigating the adverse effects of the integration of large numbers of distributed flexible resources.

[0003] Existing research has explored the application of distributed generation (DG) and demand response (DR) in distribution network reconfiguration. It has been found that electric vehicles with smart charging modes and air conditioners with adjustable operating modes can achieve demand response and leverage time-of-use electricity prices to balance user electricity demand, significantly improving the economic efficiency of distribution network reconfiguration. Furthermore, with the increasing adoption of flexible smart switches in distribution networks, their application in distribution network reconfiguration will become a future trend. Previous research has analyzed the application of smart soft switches (SOPs) in active distribution system fault restoration, demonstrating their significant advantages in voltage support and power restoration during fault conditions. Furthermore, distribution systems equipped with SOPs can achieve more efficient energy utilization and improved voltage drop during steady-state operation.

[0004] However, comprehensive research on the integration of high-proportion renewable energy, demand-side response, and intelligent soft switches into distribution network reconfiguration remains limited. Furthermore, among existing algorithms for solving distribution network reconfiguration, traditional heuristic algorithms, such as genetic algorithms and particle swarm optimization, are widely used. However, these methods often yield locally optimal solutions, potentially failing to ensure overall optimality. Summary of the Invention

[0005] The present invention provides a distribution network reconstruction method, system, device and medium for utilizing demand-side response and flexible intelligent switches to further reduce the cost of distribution system reconstruction, improve the new energy absorption capacity, and manage EV charging behavior; further, by solving high-quality optimal solutions through particle algorithm, the economy and feasibility of the reconstruction scheme are improved.

[0006] In view of this, a first aspect of the present invention provides a distribution network reconstruction method, the method comprising:

[0007] By describing the charging behavior of electric vehicle clusters and managing it through virtual electricity pricing, a charging optimization model based on a distributed Lagrangian relaxation algorithm is constructed.

[0008] Under the guidance of the charging optimization model for electric vehicles, a multi-objective dynamic reconstruction model of the distribution network is constructed according to the economic efficiency of distribution network operation, voltage quality and life of flexible intelligent switches;

[0009] The standard particle swarm algorithm is improved by combining the Sigmoid function and the niche technology to obtain an improved particle swarm algorithm. The improved particle swarm algorithm is used to solve the multi-objective dynamic reconstruction model of the distribution network and obtain the distribution network flexible intelligent switch and branch interconnection strategy.

[0010] Optionally, the method describes the charging behavior of the electric vehicle cluster and manages the charging behavior through virtual electricity prices, constructs a charging optimization model of a distributed Lagrangian relaxation algorithm, and guides the charging behavior of the electric vehicles, including:

[0011] By predicting the charging time sequence distribution of each electric vehicle, a disordered charging model of the electric vehicle is constructed, and the disordered charging behavior of the electric vehicle is determined based on the disordered charging model;

[0012] Managing the disorderly charging behavior of electric vehicles by means of virtual electricity prices, and constructing a load demand model under the influence of virtual electricity prices;

[0013] A centralized electric vehicle optimization scheduling model is constructed based on the load demand model, and the centralized electric vehicle optimization scheduling model is optimized by a distributed Lagrangian relaxation method to obtain a charging optimization model of a distributed Lagrangian relaxation algorithm.

[0014] Optionally, under the guidance of the charging optimization model for the electric vehicle, a multi-objective dynamic reconstruction model of the distribution network is constructed according to the economic efficiency of distribution network operation, voltage quality and life of flexible intelligent switches, including:

[0015] Under the guidance of the charging optimization model for electric vehicles, a multi-objective dynamic reconstruction model of the distribution network is constructed by combining the distribution network operation economy, node voltage performance, and flexible intelligent switch operation and investment cost;

[0016] The constraints of the multi-objective dynamic reconstruction model of the distribution network are set, including: system flow constraints, node voltage constraints, branch current constraints, network topology constraints, switch conversion times constraints in the distribution network, and operation constraints of flexible intelligent switches in the distribution system.

[0017] Optionally, the improved particle swarm algorithm is obtained by improving the standard particle swarm algorithm by combining the Sigmoid function and the niche technology. The improved particle swarm algorithm is used to solve the multi-objective dynamic reconstruction model of the distribution network to obtain a distribution network flexible intelligent switch and branch interconnection strategy, including:

[0018] After introducing the Sigmoid function to improve the position update formula of the standard particle swarm algorithm, the fitness of each individual in the group is adjusted through the niche technology, and the best particle is selected by the roulette wheel method to obtain the improved particle swarm algorithm.

[0019] The charging optimization model is solved by the improved particle swarm algorithm to obtain multiple groups of solutions, and a fuzzy membership decision method is introduced to select a group of compromise solutions from the multiple groups of solutions as the final solution of the improved particle swarm algorithm, which serves as the flexible intelligent switch and branch interconnection strategy of the distribution network.

[0020] A second aspect of the present invention provides a distribution network reconstruction system, the system comprising:

[0021] The first construction unit is used to construct a charging optimization model of a distributed Lagrangian relaxation algorithm by describing the charging behavior of the electric vehicle cluster and managing the charging behavior through a virtual electricity price;

[0022] The second construction unit is configured to construct a multi-objective dynamic reconstruction model of the distribution network based on the operation economy of the distribution network, voltage quality, and life of the flexible intelligent switch under the guidance of the charging optimization model for the electric vehicle;

[0023] The solving unit is used to improve the standard particle swarm algorithm by combining the Sigmoid function and the niche technology to obtain an improved particle swarm algorithm, and solve the multi-objective dynamic reconstruction model of the distribution network by the improved particle swarm algorithm to obtain the distribution network flexible intelligent switch and branch interconnection strategy.

[0024] Optionally, the first building unit is used to:

[0025] By predicting the charging time distribution of each electric vehicle, a disorderly charging model of electric vehicles is constructed to determine the disorderly charging behavior of electric vehicles;

[0026] Managing the disorderly charging behavior of electric vehicles by means of virtual electricity prices, and constructing a load demand model under the influence of virtual electricity prices;

[0027] A centralized electric vehicle optimization scheduling model is constructed based on the load demand model, and the centralized electric vehicle optimization scheduling model is optimized by a distributed Lagrangian relaxation method to obtain a charging optimization model of a distributed Lagrangian relaxation algorithm to guide the charging behavior of electric vehicles.

[0028] Optionally, the second building unit is used to:

[0029] Under the guidance of the charging optimization model for electric vehicles, a multi-objective dynamic reconstruction model of the distribution network is constructed by combining the distribution network operation economy, node voltage performance, and flexible intelligent switch operation and investment cost;

[0030] The constraints of the multi-objective dynamic reconstruction model of the distribution network are set, including: system flow constraints, node voltage constraints, branch current constraints, network topology constraints, switch conversion times constraints in the distribution network, and operation constraints of flexible intelligent switches in the distribution system.

[0031] Optionally, the solving unit is used to:

[0032] After introducing the Sigmoid function to improve the position update formula of the standard particle swarm algorithm, the fitness of each individual in the group is adjusted through the niche technology, and the best particle is selected by the roulette wheel method to obtain the improved particle swarm algorithm.

[0033] The charging optimization model is solved by the improved particle swarm algorithm to obtain multiple groups of solutions, and a fuzzy membership decision method is introduced to select a group of compromise solutions from the multiple groups of solutions as the final solution of the improved particle swarm algorithm, which serves as the flexible intelligent switch and branch interconnection strategy of the distribution network.

[0034] A third aspect of the present invention provides a distribution network reconstruction device, the device comprising a processor and a memory:

[0035] The memory is used to store program code and transmit the program code to the processor;

[0036] The processor is configured to execute the steps of the distribution network reconstruction method as described in the first aspect above according to the instructions in the program code.

[0037] A fourth aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium is used to store program code, and the program code is used to execute the distribution network reconstruction method described in the first aspect.

[0038] It can be seen from the above technical solutions that the present invention has the following advantages:

[0039] A distribution network reconstruction method provided in an embodiment of the present invention first uses a virtual electricity price strategy to construct a charging optimization model of a distributed Lagrangian relaxation algorithm for the disorderly charging behavior of connected electric vehicles to manage the charging behavior of electric vehicles; then, based on the flow control of flexible intelligent switches and the dynamic reconstruction strategy of the distribution network, an operation architecture of the distribution network with large-scale electric vehicle access is constructed, and a multi-objective optimization mathematical model combining the number of switch operations, node voltage quality, and network loss is established; finally, to determine the optimal reconstruction scheme of the system, an improved particle swarm algorithm is obtained by improving the standard particle swarm algorithm by combining the Sigmoid function and the niche technology. The improved particle swarm algorithm is used to solve the multi-objective dynamic reconstruction model of the distribution network to obtain a flexible intelligent switch and branch connection strategy for the distribution network. On the one hand, the present invention uses flexible intelligent switches to reduce the operating cost of the distribution network and improves the voltage quality by optimizing the charging behavior of electric vehicles. On the other hand, the proposed improved algorithm improves the solving ability of the optimization model, solves the defects of the traditional heuristic algorithm in global optimization, provides a high-quality optimal solution, and thus improves the economy and feasibility of the reconstruction scheme. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0041] Figure 1 A schematic diagram of a flow chart of a distribution network reconstruction method provided by an embodiment of the present invention;

[0042] Figure 2 A flowchart of EV optimal scheduling based on the LR optimization algorithm and virtual electricity price provided by an embodiment of the present invention;

[0043] Figure 3 Flowchart of the improved particle swarm algorithm provided by the embodiment of the present invention;

[0044] Figure 4 A schematic structural diagram of a distribution network reconstruction system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0045] In order to make the purpose, features, and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0046] See also Figure 1 , a distribution network reconstruction method provided in an embodiment of the present invention includes:

[0047] Step 101: Construct a charging optimization model of a distributed Lagrangian relaxation algorithm by describing the charging behavior of an electric vehicle cluster and managing the charging behavior through a virtual electricity price.

[0048] Step 102: Under the guidance of the charging optimization model for electric vehicles, a multi-objective dynamic reconstruction model of the distribution network is constructed according to the economic efficiency of distribution network operation, voltage quality, and life of flexible intelligent switches.

[0049] Step 103: Improve the standard particle swarm algorithm by combining the Sigmoid function and the niche technology to obtain an improved particle swarm algorithm. Use the improved particle swarm algorithm to solve the multi-objective dynamic reconstruction model of the distribution network and obtain the distribution network flexible intelligent switch and branch interconnection strategy.

[0050] In one embodiment, step 101 includes:

[0051] Step 1011: Construct a disordered charging model of electric vehicles by predicting the charging time sequence distribution of each electric vehicle, and determine the disordered charging behavior of the electric vehicles based on the disordered charging model.

[0052] It should be noted that this embodiment makes the following assumptions: all electric vehicles (EVs) within the research scope obey the scheduling arrangement, and in 24 hours a day, each EV departs from the starting point and arrives at the destination, and returns to the starting point after a period of time, and each EV can obtain the ideal amount of electricity during the charging period.

[0053] The behavior patterns of electric vehicle users in the region are described. Considering factors such as EV travel distance and charging time, and considering user needs, whether each electric vehicle needs to be connected to the power grid for charging after arriving at its destination, the charging time distribution of each EV is predicted, thereby obtaining an EV disorderly charging model. The EV disorderly charging model of this embodiment is expressed as follows:

[0054] (1)

[0055] (2)

[0056] (3)

[0057] (4)

[0058] (5)

[0059] (6)

[0060] in, is a probability density function expressed in logarithms, which represents the distribution of travel mileage of each electric vehicle. It represents the distance between the destination and the origin. According to statistical data, the expectation and standard deviation are generally set as ; The time required for an electric vehicle to depart from its starting point and return to its destination is also expressed by a probability density function, and both obey a normal distribution. and They represent the time when the EV departs and returns to the starting point, which can be easily obtained based on urban traffic statistics. and The expected values are 8.5 and 17.5, respectively, and the standard deviations All are set to 0.5; Indicates the SOC value required for an electric vehicle to travel from its starting point to its destination. (State of Charge: Commonly used in the battery field to indicate the remaining battery capacity. It is usually expressed as a percentage. For example, when the SOC is 50%, it means that the remaining battery capacity is 50% of its rated capacity. Accurately estimating SOC is very important for the rational use, management, and life prediction of batteries. It is widely used in scenarios where batteries are used, such as electric vehicles and energy storage systems.) is the SOC value consumed by the EV for every 1 km traveled. From a practical point of view, it is assumed that EVs do not have to be charged midway, that is, the power of each EV before departure meets the minimum capacity requirement for a single trip as described above. Therefore, based on the differences between different EVs and the one-way mileage, the initial SOC value of the i-th electric vehicle is obtained. As shown in formula (5), N represents the number of electric vehicles in the area, which is an N-dimensional random variable, where the i-th value Indicates the state of the i-th EV; when considering whether the i-th EV needs to be charged, we should start from its initial SOC value and one-way driving distance to judge whether the electric vehicle's power meets the requirements when it arrives at the destination and returns to the starting point. If it does not meet the requirements, as shown in formula (6), the EV should be charged. represents the SOC value of the i-th EV when it arrives at the destination or returns to the starting point, represents the SOC required for the i-th EV to travel one way, Indicates the self-discharge coefficient of the EV battery.

[0061] Step 1012: Manage the disorderly charging behavior of electric vehicles by means of virtual electricity prices, and construct a load demand model under the influence of virtual electricity prices.

[0062] It should be noted that this step formulates a corresponding electricity price strategy for the disorderly charging behavior of electric vehicles in the studied area, and uses virtual electricity prices in a scheduling cycle. In order to better demonstrate the guiding effect of this strategy on EV charging behavior, a complete scheduling cycle is discretized into multiple scheduling periods. , each time segment is 0.25h long, and the load demand model of the area under the influence of the virtual electricity price can be obtained. The load demand model expression of this embodiment is as follows:

[0063] (7)

[0064] (8)

[0065] (9)

[0066] in, express The virtual electricity price displayed when electric vehicles are connected to the grid, express The total load in the distribution system at this moment, express The base load level at the moment, express EV charging load connected to the grid at all times; when formulating the virtual electricity price, in order to make the virtual electricity price better follow the changing trend of the current peak and valley electricity prices of the grid, the amount of the virtual electricity price is set to a different proportion with the peak and valley electricity prices. is the adjustment coefficient, where , when the overall load is greater than, by adjusting the virtual electricity price, it can maximize the current electricity price when the load demand is high; Indicates the base level of load in the region, Indicates the electricity price during peak and valley periods. represents the average load value during the peak and valley period of electricity price; the partial derivative of formula (7) can be obtained , indicating that the virtual electricity price can correctly reflect the load level in the current period, and the demand side management of the load can be better achieved through electricity price guidance.

[0067] Step 1013: construct a centralized electric vehicle optimization scheduling model based on the load demand model, optimize the centralized electric vehicle optimization scheduling model through a distributed Lagrangian relaxation method, and obtain a charging optimization model of a distributed Lagrangian relaxation algorithm.

[0068] It should be noted that this step first establishes a Lagrangian relaxation model for electric vehicle charging behavior. Specifically, the EV charging behavior is described by the LR centralized model, where LR is logistic regression (LR). Then, it is discretized. The established centralized electric vehicle optimization scheduling model is as follows:

[0069] (10)

[0070] (11)

[0071] (12)

[0072] (13)

[0073] (14)

[0074] Among them, Equations (10)-(14) are Lagrangian centralized models. As shown in Equation (10), under the premise of meeting electricity demand, the optimization target of scheduling electric vehicle clusters is EV charging cost Minimum, formula (12) represents the constraints satisfied by the SOC of each electric vehicle during charging, formula (13) represents the time constraints that electric vehicles can respond to scheduling, assuming that each EV can respond to the adjustment of the virtual electricity price at the end of the period of access to the grid and at the end of the period before the end of charging, formula (14) represents the constraints on the real-time transmission power of the distribution network, At any given moment, the total load of the distribution network (including base load and electric vehicle charging load) should not exceed the system's load limit; Indicates that the i-th electric car is The charging state at the moment, its value satisfies the description of formula (11), is the charging power of the EV, which is assumed to be a constant value; and Represent EV battery capacity and charging efficiency, represents the expected charging SOC of the i-th electric vehicle, and represents the time when the i-th EV is connected to the grid twice. represents the time when the i-th EV finishes charging, Indicates the upper limit of load that the power distribution system can bear.

[0075] Then, based on the centralized electric vehicle optimization scheduling model, the distributed Lagrangian relaxation method is used to optimize it. The optimization process is as follows:

[0076] (15)

[0077] (16)

[0078] (17)

[0079] (18)

[0080] (19)

[0081] (20)

[0082] (twenty one)

[0083] a) The optimization objective function remains unchanged and is converted into a distributed LR optimization model, which can improve the solution speed of the model, as shown in Equations (15)-(20). The optimization algorithm flow chart is shown in the attached Figure 2 ;

[0084] b) The Lagrange multiplier Add it to formula (14), transform the objective function of the original problem without systematic constraints, and obtain the objective function expressed by the Lagrangian relaxation algorithm, as shown in formula (15). On this basis, the problem is decomposed into the charging sub-problem corresponding to each electric vehicle, and the objective function is obtained as shown in formula (16). At this time, the two constraints of formulas (12)-(13) can be combined, as shown in formula (17). By solving N problems containing formulas (15)-(17) in parallel, the charging status of each electric vehicle at any time in a scheduling cycle can be obtained;

[0085] c) Dualize the original problem to obtain the equation (18), while treating the Lagrange multipliers in the relaxation function as variables;

[0086] d) Verify the accuracy of the solution by substituting the EV charging state from b) into both the original and dual problems. If the difference between the two meets the verification requirements, the solution to the dual problem can be considered to represent the result of the original problem. Otherwise, an iterative update should be performed.

[0087] e) Considering the non-smoothness of the dual problem, the subgradient method is used to update the Lagrange multiplier, as shown in Equation (19). It represents the number of iterations. The initial value of the Lagrange multiplier is set, and then it is iterated along the subgradient direction until the multiplier meets the requirements shown in Equation (21). The iteration ends and the infimum of the dual problem is the optimal solution of the original problem.

[0088] in, represents the number of constraints, Indicates the step size of the vth round of iteration, It represents the sub-gradient of the vth iteration, which is the ratio of the sub-gradient of the iteration to its first-order norm, that is, , represents the Lagrange multiplier in the vth iteration process at time t, represents the iteration rate of the Lagrange multiplier.

[0089] In one embodiment, step 102 includes:

[0090] Step 1021: Under the guidance of the charging optimization model for electric vehicles, a multi-objective dynamic reconstruction model of the distribution network is constructed in combination with the economic efficiency of distribution network operation, node voltage performance, and flexible intelligent switch operation and investment cost.

[0091] It should be noted that active network loss can reflect the economic efficiency of system operation, and the degree of deviation of the node voltage at both ends of the bus can be used to evaluate the relationship between the load size connected to the bus and the power injected by the grid, thereby reflecting the real-time operating status of the bus. Since distribution network reconstruction is essentially a change in the switch state in the network, frequent operation of the switch may reduce its service life. The number of switch conversions should be minimized. When considering the cost of flexible intelligent switches, their investment cost is converted into part of the operating cost. Combining the above four indicators: distribution network operation economy, node voltage performance, and flexible intelligent switch operation and investment costs, the objective function of distribution network reconstruction is established, and the expression is as follows:

[0092] (twenty two)

[0093] (twenty three)

[0094] (twenty four)

[0095] (25)

[0096] in, Indicates the system active power loss (network loss), Indicates the total number of branches contained in the distribution network, is the switch state of the i-th branch, 0 for open and 1 for closed. and They represent the active and reactive power flowing through the i-th branch, represents the voltage at the first end of the busbar of the ith branch, Indicates the charging period of electric vehicles, Indicates the reference period of the system in the process of calculating network loss; Indicates the voltage offset of the entire system. is the total number of system nodes, and Respectively represent the rated voltage and actual voltage of the i-th node; represents the total cost of running the SOP, Indicates the set of branches connected to SOP in the system. Indicates the operating cost of SOP per unit time, Indicates the power loss of the distribution network branch flow during the time period when passing through the SOP in scenario c; and Respectively State variables of section switches and tie switches in the period distribution system, and They represent the number of section switches and tie switches in the system respectively. It is worth noting that the number of switch operations in the system is a multiple of 2.

[0097] Step 1022: Set the constraints of the multi-objective dynamic reconstruction model of the distribution network, including: system flow constraints, node voltage constraints, branch current constraints, network topology constraints, switch conversion times constraints in the distribution network, and operation constraints of flexible intelligent switches in the distribution system.

[0098] It should be noted that, based on the objective function in step 1021, the relevant constraints for distribution network reconstruction, including system power flow constraints, node voltage constraints, branch current constraints, network topology constraints, switch switching times constraints in the distribution network, and operation constraints of flexible intelligent switches in the distribution system, are expressed as follows:

[0099] (26)

[0100] (27)

[0101] (28)

[0102] (29)

[0103] (30)

[0104] (31)

[0105] (32)

[0106] (33)

[0107] (34)

[0108] (35)

[0109] (36)

[0110] (37)

[0111] (38)

[0112] Where, Equation (26) represents the system power flow constraint, and They represent the active power and reactive power injected by the i-th node, and are the active power and reactive power injected by the electric vehicle at the i-th node, and Represent the voltage amplitude of node i, j respectively, represents the admittance matrix of the distribution network branch; Equation (27) represents the node voltage constraint, and They represent the upper and lower limits of the voltage amplitude of the i-th node respectively; Equation (28) represents the branch current constraint, represents the upper limit of the current that can pass through the i-th branch; Equation (29) represents the topological structure constraint of the network. After the distribution network is reconfigured, the network status needs to be checked to ensure that there are no islands or loops in the grid. represents the combination of switch states in the network, is the set of switches in the system that contain positions and state variables; Equation (30) represents the switching number constraint of the switches in the distribution network, represents the total number of system switching changes, Indicates the maximum number of switching changes that the system can withstand; Equations (31)-(38) represent the operating constraints of the SOP in the distribution system. It is worth noting that since Equations (31)-(32) contain quadratic terms, the optimization algorithm cannot directly solve them. Therefore, the rotation cone convexification method is used to convert them into the form of Equations (37)-(38), where, and 、 and They represent the active power and reactive power flowing out of the kth smart flexible switch under scenario c, and 、 and They represent the upper and lower limits of reactive power that the converters at both ends of the kth SOP can carry, i and j are the two branches connected by the SOP, Indicates the loss coefficient of SOP, generally taken as 0.02, and They represent the design capacity of the converter at both ends of the i-th SOP respectively.

[0113] In one embodiment, step 103 includes:

[0114] Step 1031: After introducing the Sigmoid function to improve the position update formula of the standard particle swarm algorithm, the fitness of each individual in the group is adjusted by the niche technology, and the best particle is selected by the roulette wheel method to perform the operation, thereby obtaining the improved particle swarm algorithm.

[0115] It should be noted that since the basic particle swarm algorithm updates the position based on decimal numbers, and the distribution network reconstruction mainly examines the state of the switch, which is a 0-1 variable, the present invention first introduces the Sigmoid function to improve the position update formula of the standard particle swarm algorithm. The calculation speed and position update formula of the improved second-order particle swarm algorithm are:

[0116] (39)

[0117] (40)

[0118] Among them, represents the introduced Sigmoid function, rand represents the random generation of a real number between [0,1], and Represent the current speed and position of the particle, is the inertia weight, are learning factors, They are all random numbers generated by rand. represents the position of the global extreme point, Indicates the extreme point position of an individual.

[0119] Next, considering that the conventional particle swarm algorithm is prone to fall into a local range during the calculation process and the optimal solution obtained is not general, based on the above-mentioned improvement of the position update formula of the standard particle swarm algorithm by introducing the Sigmoid function, the fitness of each individual in the group is adjusted by introducing the niche technology, and the best particle is selected for operation using the roulette method, thereby improving the global optimization ability of the algorithm:

[0120] (41)

[0121] (42)

[0122] (43)

[0123] Among them, formula (41) represents the fitness sharing function, which is used to dynamically adjust the fitness of individuals. represents the number of individuals in the microhabitat, represents the sharing function in the niche, as shown in Equation (42), represents the Euclidean distance between individuals in a small habitat, represents the niche radius of the i-th particle, which can change dynamically according to the distance between particles i and j, as shown in Equation (43).

[0124] Step 1032: Solve the charging optimization model using the improved particle swarm algorithm to obtain multiple solutions, and introduce a fuzzy membership decision method to select a compromise solution from the multiple solutions as the final solution of the improved particle swarm algorithm, which serves as the distribution network flexible intelligent switch and branch interconnection strategy.

[0125] It should be noted that after the Pareto optimal solution is obtained by using the high-quality population retention strategy mentioned above, this step introduces a fuzzy membership decision method to select a set of compromise solutions as the final solution of the improved particle swarm algorithm, as the distribution network flexible intelligent switch and branch interconnection strategy. It can be understood that the fuzzy membership decision method is used when determining the configuration of the distribution network flexible intelligent switch and the branch interconnection strategy. This method selects a set of compromise solutions from many possible solutions and uses this set of compromise solutions as the optimal solution finally found by the improved particle swarm algorithm to guide the installation of the distribution network flexible intelligent switch and the connection method of the branch, thereby achieving the optimized operation of the distribution network. The expression of this method is as follows:

[0126] (44)

[0127] (45)

[0128] Wherein, Equation (44) represents the introduced DMF function, represents the adaptability of the i-th objective function, and are the maximum and minimum values of the objective function respectively. At the same time, formula (45) is used as the standardized definition function of satisfaction. The closer the value is to 1, the better the calculation result is. Indicates the number of objective functions.

[0129] The DMF function is used to evaluate the voltage stability of each node during the distribution network reconstruction process. By calculating the DMF value, the voltage fluctuations at each node after reconstruction can be quantified, providing a basis for optimizing the distribution network structure. The DMF function comprehensively considers factors such as node voltage amplitude, phase, and load changes, and can fully reflect the voltage stability of the distribution network.

[0130] It should be further explained that the improvement process of the basic particle swarm optimization (PSO) can be referred to in the specific implementation. Figure 3 The process shown, Figure 3 IMPSO is the improved particle swarm optimization algorithm.

[0131] A distribution network reconstruction method provided in an embodiment of the present invention first uses a virtual electricity price strategy to construct a charging optimization model of a distributed Lagrangian relaxation algorithm for the disorderly charging behavior of connected electric vehicles to manage the charging behavior of electric vehicles; then, based on the flow control of flexible intelligent switches and the dynamic reconstruction strategy of the distribution network, an operation architecture of the distribution network with large-scale electric vehicle access is constructed, and a multi-objective optimization mathematical model combining the number of switch operations, node voltage quality, and network loss is established; finally, to determine the optimal reconstruction scheme of the system, an improved particle swarm algorithm is obtained by improving the standard particle swarm algorithm by combining the Sigmoid function and the niche technology. The improved particle swarm algorithm is used to solve the multi-objective dynamic reconstruction model of the distribution network to obtain a flexible intelligent switch and branch connection strategy for the distribution network. On the one hand, the present invention uses flexible intelligent switches to reduce the operating cost of the distribution network and improves the voltage quality by optimizing the charging behavior of electric vehicles. On the other hand, the proposed improved algorithm improves the solving ability of the optimization model, solves the defects of the traditional heuristic algorithm in global optimization, provides a high-quality optimal solution, and thus improves the economy and feasibility of the reconstruction scheme.

[0132] The above is a distribution network reconstruction method provided in an embodiment of the present invention, and the following is a distribution network reconstruction system provided in an embodiment of the present invention.

[0133] See also Figure 4 , a distribution network reconstruction system provided in an embodiment of the present invention includes:

[0134] The first construction unit 201 is configured to construct a charging optimization model of a distributed Lagrangian relaxation algorithm by describing the charging behavior of an electric vehicle cluster and managing the charging behavior through a virtual electricity price;

[0135] The second construction unit 202 is configured to construct a multi-objective dynamic reconstruction model of the distribution network based on the operation economy of the distribution network, voltage quality, and life of the flexible intelligent switch under the guidance of the charging optimization model for the electric vehicle;

[0136] The solving unit 203 is used to improve the standard particle swarm algorithm by combining the Sigmoid function and the niche technology to obtain an improved particle swarm algorithm, solve the distribution network multi-objective dynamic reconstruction model through the improved particle swarm algorithm, and obtain the distribution network flexible intelligent switch and branch interconnection strategy.

[0137] Furthermore, an embodiment of the present invention also provides a distribution network reconstruction device, the device including a processor and a memory:

[0138] The memory is used to store program code and transmit the program code to the processor;

[0139] The processor is configured to execute the steps of the distribution network reconstruction method as described in the above method embodiment according to the instructions in the program code.

[0140] Furthermore, an embodiment of the present invention also provides a computer-readable storage medium, which is used to store program code, and the program code is used to execute the distribution network reconstruction method described in the above method embodiment.

[0141] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0142] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in one or more embodiments of this specification are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards, and corresponding operation entrances must be provided for users to choose to authorize or refuse.

[0143] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be through some interface, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0144] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0145] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0146] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the method of the present invention. The aforementioned storage medium includes various media that can store program code, such as USB flash drives, mobile hard drives, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0147] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A distribution network reconstruction method, characterized in that: include: By describing the charging behavior of electric vehicle clusters and managing it through virtual electricity pricing, a charging optimization model based on a distributed Lagrangian relaxation algorithm is constructed. Under the guidance of the charging optimization model for electric vehicles, a multi-objective dynamic reconstruction model of the distribution network is constructed according to the economic efficiency of distribution network operation, voltage quality and life of flexible intelligent switches; The standard particle swarm algorithm is improved by combining the Sigmoid function and the niche technology to obtain an improved particle swarm algorithm. The improved particle swarm algorithm is used to solve the multi-objective dynamic reconstruction model of the distribution network and obtain the distribution network flexible intelligent switch and branch interconnection strategy.

2. The distribution network reconstruction method according to claim 1, characterized in that: The method describes the charging behavior of electric vehicle clusters and manages the charging behavior through virtual electricity prices, constructs a charging optimization model of a distributed Lagrangian relaxation algorithm, and guides the charging behavior of electric vehicles, including: By predicting the charging time sequence distribution of each electric vehicle, a disordered charging model of the electric vehicle is constructed, and the disordered charging behavior of the electric vehicle is determined based on the disordered charging model; Managing the disorderly charging behavior of electric vehicles by means of virtual electricity prices, and constructing a load demand model under the influence of virtual electricity prices; A centralized electric vehicle optimization scheduling model is constructed based on the load demand model, and the centralized electric vehicle optimization scheduling model is optimized by a distributed Lagrangian relaxation method to obtain a charging optimization model of a distributed Lagrangian relaxation algorithm.

3. The distribution network reconstruction method according to claim 1, characterized in that: Under the guidance of the charging optimization model for electric vehicles, a multi-objective dynamic reconstruction model of the distribution network is constructed according to the economic efficiency of distribution network operation, voltage quality and life of flexible intelligent switches, including: Under the guidance of the charging optimization model for electric vehicles, a multi-objective dynamic reconstruction model of the distribution network is constructed by combining the distribution network operation economy, node voltage performance, and flexible intelligent switch operation and investment cost; The constraints of the multi-objective dynamic reconstruction model of the distribution network are set, including: system flow constraints, node voltage constraints, branch current constraints, network topology constraints, switch conversion times constraints in the distribution network, and operation constraints of flexible intelligent switches in the distribution system.

4. The distribution network reconstruction method according to claim 1, characterized in that: The improved particle swarm algorithm is obtained by improving the standard particle swarm algorithm by combining the Sigmoid function and the niche technology. The improved particle swarm algorithm is used to solve the multi-objective dynamic reconstruction model of the distribution network to obtain the distribution network flexible intelligent switch and branch interconnection strategy, including: After introducing the Sigmoid function to improve the position update formula of the standard particle swarm algorithm, the fitness of each individual in the group is adjusted through the niche technology, and the best particle is selected by the roulette wheel method to obtain the improved particle swarm algorithm. The charging optimization model is solved by the improved particle swarm algorithm to obtain multiple groups of solutions, and a fuzzy membership decision method is introduced to select a group of compromise solutions from the multiple groups of solutions as the final solution of the improved particle swarm algorithm, which serves as the flexible intelligent switch and branch interconnection strategy of the distribution network.

5. A distribution network reconstruction system, characterized in that: include: The first construction unit is used to construct a charging optimization model of a distributed Lagrangian relaxation algorithm by describing the charging behavior of the electric vehicle cluster and managing the charging behavior through a virtual electricity price; The second construction unit is configured to construct a multi-objective dynamic reconstruction model of the distribution network based on the operation economy of the distribution network, voltage quality, and life of the flexible intelligent switch under the guidance of the charging optimization model for the electric vehicle; The solving unit is used to improve the standard particle swarm algorithm by combining the Sigmoid function and the niche technology to obtain an improved particle swarm algorithm, and solve the multi-objective dynamic reconstruction model of the distribution network by the improved particle swarm algorithm to obtain the distribution network flexible intelligent switch and branch interconnection strategy.

6. The distribution network reconstruction system according to claim 5, characterized in that: The first building block is used to: By predicting the charging time distribution of each electric vehicle, a disorderly charging model of electric vehicles is constructed to determine the disorderly charging behavior of electric vehicles; Managing the disorderly charging behavior of electric vehicles by means of virtual electricity prices, and constructing a load demand model under the influence of virtual electricity prices; A centralized electric vehicle optimization scheduling model is constructed based on the load demand model, and the centralized electric vehicle optimization scheduling model is optimized by a distributed Lagrangian relaxation method to obtain a charging optimization model of a distributed Lagrangian relaxation algorithm.

7. The distribution network reconstruction system according to claim 5, characterized in that: The second building block is used to: Under the guidance of the charging optimization model for electric vehicles, a multi-objective dynamic reconstruction model of the distribution network is constructed by combining the distribution network operation economy, node voltage performance, and flexible intelligent switch operation and investment cost; The constraints of the multi-objective dynamic reconstruction model of the distribution network are set, including: system flow constraints, node voltage constraints, branch current constraints, network topology constraints, switch conversion times constraints in the distribution network, and operation constraints of flexible intelligent switches in the distribution system.

8. The distribution network reconstruction system according to claim 5, characterized in that: The solving unit is used to: After introducing the Sigmoid function to improve the position update formula of the standard particle swarm algorithm, the fitness of each individual in the group is adjusted through the niche technology, and the best particle is selected by the roulette wheel method to obtain the improved particle swarm algorithm. The charging optimization model is solved by the improved particle swarm algorithm to obtain multiple groups of solutions, and a fuzzy membership decision method is introduced to select a group of compromise solutions from the multiple groups of solutions as the final solution of the improved particle swarm algorithm, which serves as the flexible intelligent switch and branch interconnection strategy of the distribution network.

9. A distribution network reconstruction device, characterized in that: The device includes a processor and a memory: The memory is used to store program code and transmit the program code to the processor; The processor is used to execute the distribution network reconstruction method according to any one of claims 1 to 4 according to the instructions in the program code.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium is used to store program code, and the program code is used to execute the distribution network reconstruction method according to any one of claims 1 to 4.

Images