Distribution network reconstruction analysis method, device, storage medium and electronic equipment

Abstract

A distribution network reconstruction analysis method, device, storage medium and electronic device, when the distribution network is operating normally, the method obtains the current grid structure, load level and load growth information of the distribution network according to the obtained spatiotemporal load model of electric vehicles, and calculates the distribution network flow to obtain the current maximum power supply multiple K value: if the maximum power supply multiple K value is lower than the preset safety limit value, the distribution network is reconstructed with the goal of improving the power supply capacity of the distribution network; if the maximum power supply multiple K value is higher than the preset safety limit value, the distribution network is reconstructed with the goal of economy to reduce line losses, and an active distribution network optimization model is obtained; a hybrid calculation method of genetic algorithm and linear programming is used to solve the active distribution network optimization model to obtain the power supply index of the reconstructed distribution network. The present invention solves the problem of poor reliability caused by the space-time charging load brought about by the increase in the number of EVs in ADN.

Description

Technical Field

[0001] The present invention relates to a distribution network reconstruction analysis method, device, storage medium and electronic equipment, belonging to the technical field of distribution network analysis. Background Art

[0002] Electric Vehicle (EV), as a "zero-emission" new energy vehicle, can reduce urban environmental pollution and greenhouse gas emissions. It is one of the important ways to achieve low-carbon development of transportation and has received widespread attention. In the future, electric vehicle load will account for an increasingly important proportion as a new form of load, and the distribution network, as a link between the power generation and transmission system and end users, is an important link to ensure power supply reliability and improve the operation of the power grid. With the increasing penetration rate of flexible loads represented by EVs, their randomness makes the system's operating state more complicated, thus bringing new challenges to the safe and stable operation of the distribution network.

[0003] Active Distribution Network (ADN) has attracted more and more attention as the main carrier of distributed generation and EV. In the future, large-scale EVs with spatiotemporal charging loads are expected to grow rapidly, which will cause potential burdens on the distribution network. In this case, it may lead to violations of network security restrictions, such as thermal restrictions and voltage restrictions, which seriously limit the EV carrying capacity of ADN and affect the reliable operation of ADN.

[0004] As an important role in smart grids, ADN has a reconfigurable topology, which provides new opportunities for improving EV carrying capacity through network reconfiguration. Network reconstruction is the main control method for active distribution management systems to achieve various goals, such as active loss reduction, load balancing, service recovery, reliability improvement, etc. The existing ADN reconstruction strategy does not divide the distribution network into functions, and ignores the fact that EV is a load with time-space transfer characteristics, with dual uncertainties in time and space. Therefore, when optimizing and reconstructing the distribution network at different time periods, the transfer characteristics of electric vehicles should be taken into account. In addition, when the existing technology performs active ADN reconstruction, it often only considers a single goal, resulting in insufficient EV carrying capacity and reliability of ADN. Summary of the invention

[0005] To this end, the present invention provides a distribution network reconstruction analysis method, device, storage medium and electronic device to solve the problem of poor reliability caused by the space-time charging load brought about by the increase in the number of EVs (EVs) in ADN.

[0006] In order to achieve the above object, the present invention provides the following technical solutions: In a first aspect, a distribution network reconstruction analysis method is provided, comprising:

[0007] Establishing an electric vehicle database, performing capacity analysis on the electric vehicle using a probability density function based on the electric vehicle database, and determining the maximum driving radius of the electric vehicle using the generated battery capacity value;

[0008] According to the maximum driving radius of electric vehicles, the departure-destination matrix is ​​used to analyze and simulate the driving distance of electric vehicles, determine the start time of slow charging of electric vehicles, and obtain the charging load of electric vehicles when the electric vehicles start charging; then the charging loads of electric vehicles in the same time and area are summed to obtain the spatiotemporal load model of electric vehicles;

[0009] When the distribution network is operating normally, the current grid structure, load level and load growth information of the distribution network are obtained, and the spatiotemporal load model of the electric vehicle is substituted to calculate the distribution network flow to obtain the current maximum power supply multiple K value: if the maximum power supply multiple K value is lower than the preset safety limit value, the distribution network is reconstructed with the goal of improving the power supply capacity of the distribution network; if the maximum power supply multiple K value is higher than the preset safety limit value, the distribution network is reconstructed with the goal of reducing line losses to obtain an active distribution network optimization model;

[0010] The hybrid calculation method of genetic algorithm and linear programming is used to solve the active distribution network optimization model and obtain the power supply indicators of the reconstructed distribution network.

[0011] As a preferred solution of the distribution network reconstruction analysis method, the electric vehicle database stores information on battery types, battery capacities and maximum driving radius of electric vehicles.

[0012] As the optimal solution for the distribution network reconstruction analysis method, the normal distribution expression of the daily driving distance of electric vehicles is:

[0013]

[0014] Where d is the daily driving distance of the electric vehicle; μ od is the average daily driving distance of electric vehicles; σ od is the standard deviation of the daily driving distance of electric vehicles;

[0015] The origin-destination matrix expression used to analyze and simulate the driving distance of electric vehicles is:

[0016]

[0017] In the formula, c ij The physical meaning of is the probability of electric vehicles moving from area i to area j from time t to t+1; b ij is the origin-destination matrix The elements in .

[0018] As the optimal solution for the distribution network reconstruction analysis method, the expression of the maximum power supply capacity index PSCI of the distribution network power supply index is:

[0019]

[0020] Where S is the maximum load that the target area can supply; S ini-j is the initial load of region j, N is the total number of regions, S d-j is the load growth step of region j; Z is the set of all regions; k is the load growth coefficient.

[0021] As the optimal solution for the distribution network reconstruction analysis method, the objective function of the active distribution network optimization model is:

[0022] Max PSCI

[0023] Min Loss

[0024] In the formula, Loss represents the line loss of the entire distribution network at a specified time;

[0025] During the distribution network reconstruction process, a switch is selected in each loop of the distribution network to be turned on, and the switch combination that achieves the optimal goal is found while satisfying the radiation operation conditions of the distribution network. Among them, the optimization variables of the active distribution network optimization model correspond to the state set of all switches in the entire distribution network. Described as:

[0026]

[0027] Where N RCS is the total number of switches, represents the state of the first switch at time t, Represents the state of the second switch at time t; t represents the time, and the state of the switch includes 0 and 1, 0 represents the switch is open, and 1 represents the switch is closed;

[0028] The constraints of the active distribution network optimization model include network flow constraints, branch capacity constraints, node voltage constraints, and network topology constraints.

[0029] As a preferred solution for the distribution network reconstruction analysis method, a hybrid calculation method of genetic algorithm and linear programming is used to solve the active distribution network optimization model. The steps include:

[0030] 1) System initialization: Initialize the switch state in the distribution network topology of the active distribution network optimization model and input the original data;

[0031] 2) Separate the continuous and discrete control variables in the active distribution network optimization model, generate the initial population, and set the number of iterations iter = 1;

[0032] 3) Call the CPLEX commercial solver to solve the optimal combination of switch states corresponding to the continuous optimization problem, and obtain the maximum power supply capacity PSCI and minimum network loss of the distribution network under the current distribution network topology;

[0033] 4) Calculate the fitness of the individual. If the convergence condition is met, proceed to step 6); otherwise, set iter = iter + 1 and execute step 5);

[0034] 5) Solve the discrete optimization problem of the switch state, generate offspring based on genetic operations, and execute steps 3) and 4);

[0035] 6) Obtain the maximum power supply capacity PSCI and minimum network loss of the distribution network that meets the constraints, and output the value of the control variable.

[0036] In a second aspect, a distribution network reconstruction analysis device is provided, comprising:

[0037] An electric vehicle database building module, used to build an electric vehicle database;

[0038] A driving radius analysis module, used to perform capacity analysis on the electric vehicle using a probability density function according to the electric vehicle database, and determine the maximum driving radius of the electric vehicle using the generated battery capacity value;

[0039] The spatiotemporal load model building module is used to simulate the driving distance of electric vehicles based on the maximum driving radius of electric vehicles by using the departure-destination matrix analysis to determine the start time of slow charging of electric vehicles. When the electric vehicles start charging, the charging load of the electric vehicles is obtained; then the charging loads of electric vehicles in the same time and area are summed to obtain the spatiotemporal load model of electric vehicles;

[0040] The optimization model building module is used to obtain the current grid structure, load level and load growth information of the distribution network when the distribution network is operating normally, and substitute the spatiotemporal load model of the electric vehicle to calculate the distribution network flow to obtain the current maximum power supply multiple K value: if the maximum power supply multiple K value is lower than the preset safety limit value, the distribution network is reconstructed with the goal of improving the power supply capacity of the distribution network; if the maximum power supply multiple K value is higher than the preset safety limit value, the distribution network is reconstructed with the goal of reducing line losses to obtain an active distribution network optimization model;

[0041] The distribution network reconstruction solution module is used to solve the active distribution network optimization model by adopting a hybrid calculation method of genetic algorithm and linear programming to obtain the power supply indicators of the reconstructed distribution network.

[0042] As a preferred solution of the distribution network reconstruction analysis device, the electric vehicle database of the electric vehicle database construction module stores information on battery types, battery capacities and maximum driving radius of electric vehicles.

[0043] As a preferred solution of the distribution network reconstruction analysis device, in the spatiotemporal load model construction module, the normal distribution expression obeyed by the daily driving distance of electric vehicles is:

[0044]

[0045] Where d is the daily driving distance of the electric vehicle; μ od is the average daily driving distance of electric vehicles; σ od is the standard deviation of the daily driving distance of electric vehicles;

[0046] The origin-destination matrix expression used to analyze and simulate the driving distance of electric vehicles is:

[0047]

[0048] In the formula, c ij The physical meaning of is the probability of electric vehicles moving from area i to area j from time t to t+1; b ij is the origin-destination matrix The elements in .

[0049] As a preferred solution of the distribution network reconstruction analysis device, in the optimization model construction module, the expression of the maximum power supply capacity index PSCI of the distribution network power supply index is:

[0050]

[0051] Where S is the maximum load that the target area can supply; S ini-j is the actual load of area j, N is the total number of areas, S d-j is the load growth step of region j; Z is the set of all regions; k is the load growth coefficient;

[0052] In the optimization model construction module, the objective function of the active distribution network optimization model is:

[0053] Max PSCI

[0054] Min Loss

[0055] In the formula, Loss represents the line loss of the entire distribution network at a specified time;

[0056] In the optimization model construction module, during the distribution network reconstruction process, a switch is selected in each loop of the distribution network to be turned on, and the switch combination that achieves the optimal goal is found while satisfying the distribution network radiation operation conditions. Among them, the state set of all switches in the entire distribution network corresponding to the optimization variables of the active distribution network optimization model is Described as:

[0057]

[0058] Where N RCS is the total number of switches, represents the state of the first switch at time t, Represents the state of the second switch at time t; t represents the time, and the state of the switch includes 0 and 1, 0 represents the switch is open, and 1 represents the switch is closed;

[0059] The constraints of the active distribution network optimization model include network flow constraints, branch capacity constraints, node voltage constraints, and network topology constraints.

[0060] As a preferred solution for the distribution network reconstruction analysis device, in the distribution network reconstruction solution module, the switch state in the distribution network topology structure of the active distribution network optimization model is initialized and the original data is input; the continuous and discrete control variables in the dynamic distribution network optimization model are separated to generate an initial population, and the number of iterations iter=1; the CPLEX commercial solver is called to solve the optimal combination of switch states corresponding to the continuous optimization problem to obtain the maximum power supply capacity PSCI and the minimum network loss of the distribution network under the current distribution network topology structure; the fitness of the individual is calculated, and if the convergence conditions are met, the maximum power supply capacity PSCI and the minimum network loss of the distribution network that meet the constraints are obtained, and the value of the control variable is output; otherwise, iter=iter+1 is set to solve the discrete optimization problem of the switch state, and offspring are generated based on genetic operations.

[0061] In a third aspect, a non-transitory computer-readable storage medium is provided, wherein the computer-readable storage medium stores a program code of a distribution network reconstruction analysis method, wherein the program code includes instructions for executing the distribution network reconstruction analysis method of the first aspect or any possible implementation thereof.

[0062] In a fourth aspect, an electronic device is provided, comprising: a memory and a processor; the processor and the memory communicate with each other via a bus; the memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute the distribution network reconstruction analysis method of the first aspect or any possible implementation thereof.

[0063] The present invention has the following advantages: by establishing an electric vehicle database, a probability density function is used to perform capacity analysis on the electric vehicle according to the electric vehicle database, and the maximum driving radius of the electric vehicle is determined by using the generated battery capacity value; according to the maximum driving radius of the electric vehicle, a departure-destination matrix is ​​used to analyze and simulate the driving distance of the electric vehicle, and the start time of slow charging of the electric vehicle is determined. When the electric vehicle starts to charge, the charging load of the electric vehicle is obtained; then the charging loads of the electric vehicles in the same time and area are summed to obtain the spatiotemporal load model of the electric vehicle; when the distribution network operates normally, the current grid structure, load level and load growth information of the distribution network are obtained, and the spatiotemporal load model of the electric vehicle is substituted to calculate the distribution network flow to obtain the current maximum power supply multiple K value: if the maximum power supply multiple K value is lower than the preset safety limit value, the distribution network is reconstructed with the goal of improving the power supply capacity of the distribution network; if the maximum power supply multiple K value is higher than the preset safety limit value, the distribution network is reconstructed with the goal of reducing line loss to obtain an active distribution network optimization model; a hybrid calculation method of genetic algorithm and linear programming is used to solve the active distribution network optimization model to obtain the power supply index of the reconstructed distribution network. The present invention takes into account the temporal and spatial characteristics of electric vehicle loads, divides the basic load and the electric vehicle load according to time and space, and can obtain the trend changes of loads at different times in distribution networks with different functions; the proposed active distribution network reconstruction strategy meets the control requirements of the active distribution network and can select an economical and reliable operation mode according to different objectives. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the implementation methods of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the implementation methods or the description of the prior art. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other implementation drawings can be derived from the provided drawings without creative work.

[0065] Figure 1 A schematic diagram of a flow chart of a power distribution network reconstruction analysis method provided in an embodiment of the present invention;

[0066] Figure 2 It is an initial state topology diagram of an IEEE 33-node distribution network in the distribution network reconstruction analysis method provided in an embodiment of the present invention;

[0067] Figure 3 The load curves of three distribution networks for 24 hours a day in the distribution network reconstruction analysis method provided in the embodiment of the present invention;

[0068] Figure 4A schematic diagram of comparing K values ​​of three distribution networks including EVs and not including EVs in the distribution network reconstruction analysis method provided in an embodiment of the present invention;

[0069] Figure 5 It is a schematic diagram for comparing K values ​​and PCSI values ​​before and after three distribution network reconstructions in the distribution network reconstruction analysis method provided in an embodiment of the present invention;

[0070] Figure 6 This is a diagram of the architecture of a power distribution network reconstruction analysis device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0071] The following is a description of the implementation of the present invention by specific embodiments. People familiar with the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. Obviously, the described embodiments are 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 creative work are within the scope of protection of the present invention.

[0072] Example 1

[0073] See also Figure 1 Embodiment 1 of the present invention provides a distribution network reconstruction analysis method, comprising the following steps:

[0074] S1. Divide the charging load area of ​​electric vehicles into commercial areas, residential areas and industrial areas, and represent the charging loads of commercial areas, residential areas and industrial areas through corresponding distribution networks respectively;

[0075] S2. Establish an electric vehicle database, wherein the electric vehicle database stores information on battery types, battery capacities, and maximum driving radius of the electric vehicle; perform capacity analysis on the electric vehicle using a probability density function based on the electric vehicle database, and determine the maximum driving radius of the electric vehicle using the generated battery capacity value;

[0076] S3. According to the maximum driving radius of the electric vehicle, the departure-destination matrix is ​​used to analyze and simulate the driving distance of the electric vehicle, and the start time of slow charging of the electric vehicle is determined. When the electric vehicle starts charging, the charging load of the electric vehicle is obtained, and then the charging loads of the electric vehicles in the same time and area are summed to obtain the spatiotemporal load model of the electric vehicle;

[0077] S4. When the distribution network is operating normally, the current grid structure, load level and load growth information of the distribution network are obtained, and the time-space load model of the electric vehicle is substituted to calculate the distribution network flow to obtain the current maximum power supply multiple K value: if the maximum power supply multiple K value is lower than the preset safety limit value, the distribution network is reconstructed with the goal of improving the power supply capacity of the distribution network; if the maximum power supply multiple K value is higher than the preset safety limit value, the distribution network is reconstructed with the goal of reducing line losses to obtain an active distribution network optimization model;

[0078] S5. A hybrid calculation method of genetic algorithm and linear programming is used to solve the active distribution network optimization model and obtain the power supply index of the reconstructed distribution network.

[0079] In this embodiment, in step S1, the charging load area is divided into commercial area, residential area and industrial area according to different functions, and the charging load of each area is represented by a distribution network.

[0080] In step S2, based on the electric vehicle database formed by market research, the electric vehicle database provides: battery type, battery capacity and maximum driving radius of electric vehicles. Based on the information provided by the electric vehicle database, the probability density function is used to describe the diversity of electric vehicles. The movement of electric vehicles in a day can be determined by the hourly departure-destination (OD) matrix. Combining the given electric vehicle characteristics and OD analysis methods, the charging load of electric vehicles in each area is estimated through Monte Carlo simulation.

[0081] In a possible embodiment, electric vehicles can be divided into the following three categories according to different user objects: private vehicles for home work (Home-Based-Work, HBW); private vehicles for non-staff travel (Home-Based-Other, HBO); and official vehicles for enterprises (Non-Home-Based, NHB).

[0082] According to statistics, the proportion of the above three types of electric vehicles is 61%, 30% and 9% respectively. This classification of electric vehicles is to obtain the driving mode of electric vehicles. Electric vehicles can be divided into:

[0083] 1) M1 passenger cars with no more than 8 seats; 2) L7e small four-wheel trucks with a maximum load of 550kg; 3) N1 trucks with a maximum load of 3500kg; 4) N2 trucks with a maximum load of 3500-12000kg. These four different types of electric vehicles account for 87.5%, 1.5%, 10% and 1% of the market respectively. This classification is for the convenience of analyzing battery capacity and type.

[0084] Since lead-acid batteries and lithium batteries account for the majority of actual electric vehicle battery use, in order to obtain the charging load and battery state of charge (SOC) of electric vehicles, it is assumed that the capacity of electric vehicles using lead-acid batteries accounts for 60% and the capacity of electric vehicles using lithium batteries accounts for 40%.

[0085] Specifically, the probability distribution of the battery capacity of electric vehicles is obtained according to the type of electric vehicles, and the Γ distribution is used for M1 and L7e electric vehicles:

[0086]

[0087] For N1 and N2 electric vehicles, a normal distribution is used:

[0088]

[0089] In the formula, Cap r is the battery capacity, α D is the shape parameter, β D is the scale parameter, σ D is the standard deviation, μ D is the mean. Among them, the probability distribution parameter values ​​of electric vehicle battery capacity are shown in Table 1:

[0090] Table 1 Probability distribution parameters of electric vehicle battery capacity

[0091]

[0092] In a possible embodiment, since the areas where electric vehicles travel in a day can be divided into industrial areas, residential areas, and commercial areas according to their functions, the normal distribution expression of the distance traveled by electric vehicles per day is as follows:

[0093]

[0094] Where d is the distance traveled by the electric vehicle every day; μ od is the average daily driving distance of electric vehicles; σ od is the standard deviation of the daily driving distance of electric vehicles.

[0095] Among them, HBW type electric vehicles, which are private vehicles used for home work, generally have only two trips in a day, while HBO and NHB type electric vehicles often have multiple trips. In a fixed area, residents' daily travel follows a fixed pattern with a clustering effect. In the spatiotemporal model, each electric vehicle is assigned an initial position. In step S3, the OD matrix is ​​used to describe the mobility of electric vehicles from departure to the end of the day. The departure-destination matrix expression used to analyze and simulate the mobility of electric vehicles is:

[0096]

[0097] In the formula, c ij The physical meaning of is the probability of electric vehicles moving from area i to area j from time t to t+1; b ij is the origin-destination matrix The elements in .

[0098] Thus, through steps S1, S2 and S3, the Monte Carlo method and the related density function are used to generate the battery capacity value of each electric vehicle, the maximum driving radius is determined based on the generated battery capacity value, the mobility of the electric vehicle is simulated through OD analysis, and then the start time tsc of slow charging is determined. When the electric vehicle starts charging, the SOC of the electric vehicle is determined, and finally the power demand of the electric vehicle is obtained. The load of the electric vehicles in the same time and area is summed to obtain the spatiotemporal load model of the electric vehicle.

[0099] In a possible embodiment, the maximum power supply margin factors include: the topology of the distribution network, the load level and the load growth level. The mathematical expression of the distribution network maximum power supply capability index (PSCI) defined in the present invention is:

[0100]

[0101] Where S is the maximum load that the target area can supply; S ini-j is the actual load of area j, N is the total number of areas, S d-j is the load growth step of region j; Z is the set of all regions; k is the load growth coefficient.

[0102] Among them, let S d-j =S ini-j , and introduce the maximum power supply multiple K as the inspection index of the power supply capacity of the distribution network. K=1+k max ;k max is the corresponding k value when S reaches its maximum value.

[0103] In a possible embodiment, according to the purpose of upper distribution network reconstruction, the objective function of the active distribution network optimization model is multi-objective, and the objective function of the active distribution network optimization model is:

[0104] Max PSCI(6)

[0105] Min Loss(7)

[0106] Where Loss represents the line loss of the entire distribution network at a specified time.

[0107] In the process of distribution network reconstruction, a switch is selected in each loop of the distribution network to be turned on, and the optimal switch combination is found while satisfying the radiation operation conditions of the distribution network. The state set of all switches in the entire distribution network corresponding to the optimization variables of the active distribution network optimization model is Described as:

[0108]

[0109] Where N RCS is the total number of switches, represents the state of the first switch at time t, Represents the state of the second switch at time t; t represents the time, and the states of the switch include 0 and 1, 0 represents that the switch is open, and 1 represents that the switch is closed.

[0110] Among them, in solving the optimal problem, the constraints of the active distribution network optimization model include network flow constraints, branch capacity constraints, node voltage constraints, and network topology constraints. The specific representation of the constraints is as follows:

[0111]

[0112] V i min ≤V i t ≤V i max (10)

[0113]

[0114]

[0115] N loop =N branch -N bus +1 (13)

[0116] In the above formula, represents the active power of node g at time t, represents the active power of node d at time t; V i t represents the voltage of node i at time t, represents the voltage of node j at time t, Y ij is the admittance from node i to node j, θ ij is the line admittance angle between node i and node j, represents the phase angle of node i at time t, represents the phase angle of node j at time t; is the reactive power of node g at time t; is the reactive power of node d at time t; Equation (9) represents the network power flow constraint. i min and V i max is the upper and lower limits of voltage, V i t is the voltage of node i at time t; Equation (10) represents the network voltage constraint; is the upper limit of current, is the upper capacity limit, is the line current at time t; is the complex power of the line flow at time t; Equations (11) and (12) are the transmission current and power capacity constraints. Equation (13) is the distribution network topology radiation operation constraint, N loop N is the number of main loops in the distribution network; branch and N bus are the number of branches and nodes in the distribution network respectively.

[0117] In a possible embodiment, in step S5, a multi-objective optimization method for a distribution network with active network reconstruction is considered, and it is necessary to optimize the switch state of each branch of the distribution network. Therefore, the active distribution network optimization model in the present invention is a mixed integer nonlinear optimization problem including continuous variables and a large number of discrete 0 and 1 integer variables. In order to efficiently solve the optimization problem, a hybrid calculation method combining genetic algorithm and linear programming (which can be efficiently solved by calling the CPLEX commercial solver) is adopted. The model to be solved includes continuous variables (that is, the maximum power supply capacity PSCI and minimum network loss of the distribution network) and discrete variables (that is, the state SW of each switch of the distribution network). r ), when the CPLEX commercial solver is called to solve the continuous part, the control variables of the discrete link remain fixed; while when the genetic algorithm is used to solve the discrete part, the control variables of the continuous part remain unchanged.

[0118] Among them, the hybrid calculation method of genetic algorithm and linear programming is used to solve the active distribution network optimization model, and the steps include:

[0119] 1) System initialization: Initialize the switch state in the distribution network topology of the active distribution network optimization model and input the original data;

[0120] 2) Separate the continuous and discrete control variables of the active distribution network optimization model, generate the initial population, and set the number of iterations iter = 1;

[0121] 3) Call the CPLEX commercial solver to solve the optimal combination of switch states corresponding to the continuous optimization problem, and obtain the maximum power supply capacity PSCI and minimum network loss of the distribution network under the current distribution network topology;

[0122] 4) Calculate the fitness of the individual. If the convergence condition is met, proceed to step 6); otherwise, set iter = iter + 1 and execute step 5);

[0123] 5) Solve the discrete optimization problem of the switch state, generate offspring based on genetic operations, and execute steps 3) and 4);

[0124] 6) Obtain the maximum power supply capacity PSCI and minimum network loss of the distribution network that meets the constraints, and output the value of the control variable.

[0125] The convergence condition of the hybrid calculation method using genetic algorithm and linear programming is: the number of optimization iterations reaches the set maximum number of iterations (set to 150 times in this embodiment), or the optimal individual remains unchanged in 10 consecutive iterations.

[0126] In this embodiment, IEEE33 nodes are used to perform case verification and analyze the impact of introducing electric vehicle loads on the reconstruction of distribution networks with different functions, such as Figure 2 As shown. Three 33-node distribution network systems are used to represent the industrial area distribution network, residential area distribution network and commercial area distribution network respectively. Each distribution network contains 5 interconnecting switches and 32 segment switches, and the system base voltage is 12.66kV. Due to their different functions, the basic load change trends of different distribution networks in different time periods are different. Assuming that the total number of electric vehicles in the three distribution networks is 1500, and they move in the three distribution networks at different times of the day, the lowest power supply multiple k is selected cir =1.6. In a distribution network, the number of electric vehicles connected to different nodes is proportional to the power value of the initial load. The topology of the initial network is as follows: Figure 2 shown.

[0127] As you can see, Figure 2 The IEEE 33-node example distribution network used has 32 section switches (numbered S1 to S32) and 5 tie switch branches (numbered S33, S34, S35, S36, S37). By changing the opening and closing states of the distribution network tie switches and section switches, the topology of the distribution network is changed, thereby improving the maximum power supply capacity of the system.

[0128] Among them, in order to reflect the active reconstruction strategy for distribution networks with different functions, the case simulation is divided into three steps:

[0129] First, based on the spatiotemporal distribution load model of electric vehicles, the PCSI value and K value of each distribution network under the initial network topology are calculated.

[0130] Second, select the moment when the K value requirement is not met, actively reconstruct the distribution network, and verify whether the K value after reconstruction meets the operation requirements.

[0131] Third, when the K value requirement is met, from an economic perspective, active reconstruction is used to reduce the network loss of the distribution network.

[0132] Among them, based on the time-space distribution load model of electric vehicles and the description of the load change trend of residential, industrial and commercial distribution networks, the load curves of the three distribution networks for 24 hours a day are obtained as follows: Figure 3 As shown. Figure 3 By comparison, it can be seen that due to the load transfer characteristics of electric vehicles, different time load trends will be introduced into distribution networks with different functions. Therefore, when actively reconstructing the distribution network to obtain better operating indicators, the time characteristics of electric vehicles and their spatial transfer characteristics should be considered in detail.

[0133] Among them, based on the constructed electric vehicle spatiotemporal distribution load model, the maximum power supply capacity indicators of the three functional distribution networks throughout the 24 hours are evaluated, and the maximum power supply multiples of the three different distribution network areas with and without electric vehicle load are obtained. Figure 4 As shown, in Figure 4 It can be seen that due to the addition of electric vehicle load, the power supply multiple has decreased in different proportions compared with the initial load, especially in the period from 18:00 to 21:00, the power supply multiple has decreased significantly. In particular, at 19:00 in the residential distribution network, the power supply multiple has dropped to 1.42 times. And at 19:00 in the industrial distribution network and 18:00 in the commercial distribution network, the power supply multiple has dropped to 1.52 and 1.58, both less than k cir In this case, the distribution network should be actively reconstructed to ensure the power supply capacity of the distribution network at this moment. See Table 2 for the changes in the maximum power supply capacity and maximum power supply multiples before and after the reconstruction of the three distribution networks.

[0134] Table 2 Comparison of PCSI values ​​before and after active reconstruction

[0135]

[0136] From the data in Table 2, we can see that after the distribution network was reconstructed, the maximum power supply capacity of the three distribution networks at the time of reconstruction increased from 7604, 7478, and 7470 to 8621, 8540, and 8589, respectively, with an increase of 13.4%, 14.2%, and 15.0%. Under the reconstructed network topology, the power supply capacity and power supply multiple changes at subsequent times are as follows: Figure 5 As shown. Figure 5 It can be seen that after the reconstruction of the residential distribution network at 19:00, the industrial distribution network at 19:00, and the industrial distribution network at 18:00, their power supply capacities have also been improved, meeting the requirements for stable operation of the distribution network.

[0137] Among them, before the distribution network is reconfigured, since the maximum power supply multiple is met, there is no need to reconfigure the distribution network. Therefore, from the perspective of economy, the load power demand can be met at this moment while reducing line losses. For this purpose, the objective function of the optimization problem is transformed into equation (7). The constraints remain unchanged, and the reconstruction schemes at different times are shown in Table 3 (a), (b), and (c):

[0138] Table 3 Active reconstruction scheme to reduce line loss (a) Industrial distribution network

[0139]

[0140] (b) Residential distribution network

[0141]

[0142] (C) Commercial distribution network

[0143]

[0144]

[0145] It can be seen from Table 3 above that when the power supply capacity is sufficient, the network loss of the distribution network can be reduced through reconstruction, which can increase the economy of the distribution network operation.

[0146] In summary, the embodiment of the present invention divides the charging load area of ​​electric vehicles into commercial areas, residential areas and industrial areas, and represents the charging loads of commercial areas, residential areas and industrial areas through corresponding distribution networks respectively; establishes an electric vehicle database, and stores battery types, battery capacities and maximum driving radius information of electric vehicles in the electric vehicle database; performs capacity analysis on electric vehicles using probability density functions according to the electric vehicle database, and determines the maximum driving radius of electric vehicles using the generated battery capacity values; according to the maximum driving radius of electric vehicles, uses the departure-destination matrix analysis to simulate the mobility of electric vehicles, determines the start time of slow charging of electric vehicles, and obtains the charging load, battery charging status and Power demand, sum the loads of electric vehicles in the same time and area to obtain the spatiotemporal load model of electric vehicles; when the distribution network operates normally, according to the obtained spatiotemporal load model of electric vehicles, obtain the current grid structure, load level and load growth information of the distribution network, calculate the distribution network flow to obtain the current maximum power supply multiple K value: if the maximum power supply multiple K value is lower than the preset safety limit value, the distribution network is reconstructed with the goal of improving the power supply capacity of the distribution network; if the maximum power supply multiple K value is higher than the preset safety limit value, the distribution network is reconstructed with the goal of economy to reduce line loss, and the active distribution network optimization model is obtained; the hybrid calculation method of genetic algorithm and linear programming is used to solve the active distribution network optimization model to obtain the power supply index of the reconstructed distribution network. The present invention takes into account the spatiotemporal characteristics of electric vehicle loads, divides the basic load and electric vehicle load according to time and space, and can obtain the trend changes of loads at different times in distribution networks with different functions; the proposed active distribution network reconstruction strategy meets the control requirements of the active distribution network, and can select an economical and reliable operation mode according to different goals.

[0147] It should be noted that the method of the embodiment of the present disclosure can be performed by a single device, such as a computer or a server. The method of the present embodiment can also be applied in a distributed scenario and completed by multiple devices cooperating with each other. In the case of such a distributed scenario, one of the multiple devices can only perform one or more steps in the method of the embodiment of the present disclosure, and the multiple devices will interact with each other to complete the described method.

[0148] It should be noted that the above describes some embodiments of the present disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the above embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0149] Example 2

[0150] See also Figure 6 Embodiment 2 of the present invention provides a distribution network reconstruction analysis device, comprising:

[0151] The electric vehicle database construction module 001 is used to establish an electric vehicle database, wherein the electric vehicle database stores information on battery types, battery capacities and maximum driving radius of the electric vehicle;

[0152] A driving radius analysis module 002 is used to perform capacity analysis on the electric vehicle using a probability density function according to the electric vehicle database, and determine the maximum driving radius of the electric vehicle using the generated battery capacity value;

[0153] The spatiotemporal load model construction module 003 is used to simulate the mobility of electric vehicles based on the maximum driving radius of electric vehicles and the departure-destination matrix analysis, determine the start time of slow charging of electric vehicles, obtain the charging load, battery charging state and power demand of electric vehicles when the electric vehicles start charging, and sum the loads of electric vehicles in the same time and area to obtain the spatiotemporal load model of electric vehicles;

[0154] The optimization model construction module 004 is used to obtain the current grid structure, load level and load growth information of the distribution network according to the obtained spatiotemporal load model of electric vehicles when the distribution network is operating normally, and calculate the distribution network flow to obtain the current maximum power supply multiple K value: if the maximum power supply multiple K value is lower than the preset safety limit value, the distribution network is reconstructed with the goal of improving the power supply capacity of the distribution network; if the maximum power supply multiple K value is higher than the preset safety limit value, the distribution network is reconstructed with the goal of economy to reduce line losses, and an active distribution network optimization model is obtained;

[0155] The distribution network reconstruction solution module 005 is used to solve the active distribution network optimization model by adopting a hybrid calculation method of genetic algorithm and linear programming to obtain the power supply index of the reconstructed distribution network.

[0156] In this embodiment, the electric vehicle database of the electric vehicle database construction module 001 stores information on battery types, battery capacities and maximum driving radius of the electric vehicle.

[0157] In this embodiment, in the spatiotemporal load model construction module 003, the normal distribution expression obeyed by the daily driving distance of the electric vehicle is:

[0158]

[0159] Where d is the daily driving distance of the electric vehicle; μ od is the average daily driving distance of electric vehicles; σod is the standard deviation of the daily driving distance of electric vehicles;

[0160] The origin-destination matrix expression used to analyze and simulate the driving distance of electric vehicles is:

[0161]

[0162] In the formula, c ij The physical meaning of is the probability of electric vehicles moving from area i to area j from time t to t+1; b ij is the origin-destination matrix The elements in .

[0163] In this embodiment, in the optimization model building module 004, the expression of the maximum power supply capacity index PSCI of the power supply index of the distribution network is:

[0164]

[0165] Where S is the maximum load that the target area can supply; S ini-j is the actual load of area j, N is the total number of areas, S d-j is the load growth step of region j; Z is the set of all regions; k is the load growth coefficient;

[0166] In the optimization model construction module 004, the objective function of the active distribution network optimization model is:

[0167] Max PSCI

[0168] Min Loss

[0169] In the formula, Loss represents the line loss of the entire distribution network at a specified time;

[0170] In the optimization model construction module 004, during the distribution network reconstruction process, a switch is selected in each loop of the distribution network to be turned on, and the switch combination that achieves the optimal goal is found while satisfying the distribution network radiation operation conditions. Among them, the state set of all switches in the entire distribution network corresponding to the optimization variables of the active distribution network optimization model Described as:

[0171]

[0172] Where N RCS is the total number of switches, represents the state of the first switch at time t, Represents the state of the second switch at time t; t represents the time, and the state of the switch includes 0 and 1, 0 represents the switch is open, and 1 represents the switch is closed;

[0173] The constraints of the active distribution network optimization model include network flow constraints, branch capacity constraints, node voltage constraints, and network topology constraints.

[0174] In this embodiment, in the distribution network reconstruction solution module 005, the switch state in the distribution network topology of the active distribution network optimization model is initialized and the original data is input; the continuous and discrete control variables in the dynamic distribution network optimization model are separated to generate an initial population, and the number of iterations iter=1; the CPLEX commercial solver is called to solve the optimal combination of switch states corresponding to the continuous optimization problem to obtain the maximum power supply capacity PSCI and the minimum network loss of the distribution network under the current distribution network topology; the fitness of the individual is calculated, and if the convergence conditions are met, the maximum power supply capacity PSCI and the minimum network loss of the distribution network that meet the constraints are obtained, and the value of the control variable is output; otherwise, iter=iter+1 is set to solve the discrete optimization problem of the switch state, and offspring are generated based on genetic operations.

[0175] It should be noted that the information interaction, execution process and other contents between the modules of the above-mentioned device are based on the same concept as the method embodiment in Example 1 of the present application, and the technical effects they bring are the same as those of the method embodiment of the present application. For specific contents, please refer to the description in the method embodiment shown in the previous part of the present application, and will not be repeated here.

[0176] Example 3

[0177] Embodiment 3 of the present invention provides a non-transitory computer-readable storage medium, in which a program code of a distribution network reconstruction analysis method is stored, and the program code includes instructions for executing the distribution network reconstruction analysis method of embodiment 1 or any possible implementation thereof.

[0178] The computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that includes one or more available media. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0179] Example 4

[0180] Embodiment 4 of the present invention provides an electronic device, including: a memory and a processor;

[0181] The processor and the memory communicate with each other via a bus; the memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute the distribution network reconstruction analysis method of Example 1 or any possible implementation thereof.

[0182] Specifically, the processor can be implemented by hardware or by software. When implemented by hardware, the processor can be a logic circuit, an integrated circuit, etc.; when implemented by software, the processor can be a general-purpose processor implemented by reading software codes stored in a memory. The memory can be integrated in the processor or can be located outside the processor and exist independently.

[0183] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center.

[0184] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, and optionally, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0185] Although the present invention has been described in detail above by general description and specific embodiments, it is obvious to those skilled in the art that some modifications or improvements can be made to the present invention. Therefore, these modifications or improvements made without departing from the spirit of the present invention all belong to the scope of protection claimed by the present invention.

Claims

1. A distribution network reconstruction analysis method, characterized in that: include: Establishing an electric vehicle database, performing capacity analysis on the electric vehicle using a probability density function based on the electric vehicle database, and determining the maximum driving radius of the electric vehicle using the generated battery capacity value; According to the maximum driving radius of electric vehicles, the departure-destination matrix is ​​used to analyze and simulate the driving distance of electric vehicles, determine the start time of slow charging of electric vehicles, and obtain the charging load of electric vehicles when the electric vehicles start charging; then the charging loads of electric vehicles in the same time and area are summed to obtain the spatiotemporal load model of electric vehicles; When the distribution network is operating normally, the current grid structure, load level and load growth information of the distribution network are obtained, and the time-space load model of the electric vehicle is substituted to calculate the distribution network flow to obtain the current maximum power supply multiple K value: if the maximum power supply multiple K value is lower than the preset safety limit value, the distribution network is reconstructed with the goal of improving the power supply capacity of the distribution network; If the maximum power supply multiple K value is higher than the preset safety limit value, the distribution network is reconstructed with the goal of reducing line losses, and an active distribution network optimization model is obtained; The hybrid calculation method of genetic algorithm and linear programming is used to solve the active distribution network optimization model and obtain the power supply index of the reconstructed distribution network. The hybrid calculation method of genetic algorithm and linear programming is used to solve the active distribution network optimization model. The steps include: 1) System initialization: Initialize the switch state in the distribution network topology of the active distribution network optimization model and input the original data; 2) Separate the continuous and discrete control variables in the active distribution network optimization model, generate the initial population, and set the number of iterations iter = 1; 3) Call the CPLEX commercial solver to solve the optimal combination of switch states corresponding to the continuous optimization problem, and obtain the maximum power supply capacity PSCI and minimum network loss of the distribution network under the current distribution network topology; 4) Calculate the fitness of the individual. If the convergence condition is met, proceed to step 6); otherwise, set iter = iter + 1 and execute step 5); 5) Solve the discrete optimization problem of the switch state, generate offspring based on genetic operations, and execute steps 3) and 4); 6) Obtain the maximum power supply capacity PSCI and minimum network loss of the distribution network that meets the constraints, and output the value of the control variable.

2. The power distribution network reconstruction analysis method according to claim 1, characterized in that: The electric vehicle database stores information on battery types, battery capacities and the maximum driving radius of the electric vehicle.

3. The power distribution network reconstruction analysis method according to claim 2, characterized in that: The normal distribution expression of the daily driving distance of electric vehicles is: Where d is the daily driving distance of the electric vehicle; μ od is the average daily driving distance of electric vehicles; σ od is the standard deviation of the daily driving distance of electric vehicles; The origin-destination matrix expression used to analyze and simulate the driving distance of electric vehicles is: In the formula, c ij The physical meaning of is the probability of electric vehicles moving from area i to area j from time t to t+1; b ij is the origin-destination matrix The elements in .

4. The power distribution network reconstruction analysis method according to claim 3, characterized in that: The expression of the maximum power supply capacity index PSCI of the distribution network power supply index is: Where S is the maximum load that the target area can supply; S ini-j is the initial load of region j, N is the total number of regions, S d-j is the load growth step of region j; Z is the set of all regions; k is the load growth coefficient.

5. The power distribution network reconstruction analysis method according to claim 4, characterized in that: The objective function of the active distribution network optimization model is: Max PSCI Min Loss In the formula, Loss represents the line loss of the entire distribution network at a specified time; During the distribution network reconstruction process, a switch is selected in each loop of the distribution network to be turned on, and the switch combination that achieves the optimal goal is found while satisfying the radiation operation conditions of the distribution network. Among them, the optimization variables of the active distribution network optimization model correspond to the state set of all switches in the entire distribution network. Described as: Where N RCS is the total number of switches, represents the state of the first switch at time t, Represents the state of the second switch at time t; t represents the time, and the state of the switch includes 0 and 1, 0 represents the switch is open, and 1 represents the switch is closed; The constraints of the active distribution network optimization model include network flow constraints, branch capacity constraints, node voltage constraints, and network topology constraints.

6. A power distribution network reconstruction analysis device, characterized in that: include: An electric vehicle database building module, used to build an electric vehicle database; A driving radius analysis module, used to perform capacity analysis on the electric vehicle using a probability density function according to the electric vehicle database, and determine the maximum driving radius of the electric vehicle using the generated battery capacity value; The spatiotemporal load model building module is used to simulate the driving distance of electric vehicles based on the maximum driving radius of electric vehicles by using the departure-destination matrix analysis to determine the start time of slow charging of electric vehicles. When the electric vehicles start charging, the charging load of the electric vehicles is obtained; then the charging loads of electric vehicles in the same time and area are summed to obtain the spatiotemporal load model of electric vehicles; The optimization model building module is used to obtain the current grid structure, load level and load growth information of the distribution network when the distribution network is operating normally, and substitute the spatiotemporal load model of electric vehicles to calculate the distribution network flow to obtain the current maximum power supply multiple K value: if the maximum power supply multiple K value is lower than the preset safety limit value, the distribution network is reconstructed with the goal of improving the power supply capacity of the distribution network; If the maximum power supply multiple K value is higher than the preset safety limit value, the distribution network is reconstructed with the goal of reducing line losses, and an active distribution network optimization model is obtained; The distribution network reconstruction solution module is used to solve the active distribution network optimization model by using a hybrid calculation method of genetic algorithm and linear programming to obtain the power supply index of the reconstructed distribution network; In the distribution network reconstruction solution module, the switch state in the distribution network topology structure of the active distribution network optimization model is initialized and the original data is input; the continuous and discrete control variables in the active distribution network optimization model are separated to generate an initial population, and the number of iterations iter=1; The CPLEX commercial solver is called to solve the optimal combination of switch states corresponding to the continuous optimization problem, and the maximum power supply capacity PSCI and the minimum network loss of the distribution network under the current distribution network topology are obtained; the fitness of the individual is calculated. If the convergence conditions are met, the maximum power supply capacity PSCI and the minimum network loss of the distribution network that meet the constraints are obtained, and the value of the control variable is output; otherwise, iter=iter+1 is set to solve the discrete optimization problem of the switch state, and offspring are generated based on genetic operations.

7. The power distribution network reconstruction analysis device according to claim 6, characterized in that: The electric vehicle database of the electric vehicle database construction module stores information on battery types, battery capacities and maximum driving radius of the electric vehicle.

8. The power distribution network reconstruction analysis device according to claim 7, characterized in that: In the spatiotemporal load model construction module, the normal distribution expression of the daily driving distance of electric vehicles is: Where d is the daily driving distance of the electric vehicle; μ od is the average daily driving distance of electric vehicles; σ od is the standard deviation of the daily driving distance of electric vehicles; The origin-destination matrix expression used to analyze and simulate the driving distance of electric vehicles is: In the formula, c ij The physical meaning of is the probability of electric vehicles moving from area i to area j from time t to t+1; b ij is the origin-destination matrix The elements in .

9. The power distribution network reconstruction analysis device according to claim 8, characterized in that: In the optimization model construction module, the expression of the maximum power supply capacity index PSCI of the power supply index of the distribution network is: Where S is the maximum load that the target area can supply; S ini-j is the actual load of area j, N is the total number of areas, S d-j is the load growth step of region j; Z is the set of all regions; k is the load growth coefficient; In the optimization model construction module, the objective function of the active distribution network optimization model is: Max PSCI Min Loss In the formula, Loss represents the line loss of the entire distribution network at a specified time; In the optimization model construction module, during the distribution network reconstruction process, a switch is selected in each loop of the distribution network to be turned on, and the switch combination that achieves the optimal goal is found while satisfying the distribution network radiation operation conditions. Among them, the state set of all switches in the entire distribution network corresponding to the optimization variables of the active distribution network optimization model is Described as: Where N RCS is the total number of switches, represents the state of the first switch at time t, Represents the state of the second switch at time t; t represents the time, and the state of the switch includes 0 and 1, 0 represents the switch is open, and 1 represents the switch is closed; The constraints of the active distribution network optimization model include network flow constraints, branch capacity constraints, node voltage constraints, and network topology constraints.

10. A non-transitory computer-readable storage medium, wherein the computer-readable storage medium stores a program code of a distribution network reconstruction analysis method, characterized in that: The program code includes instructions for executing the distribution network reconstruction analysis method according to any one of claims 1 to 5.

11. An electronic device, comprising: Memory and processor; The processor and the memory communicate with each other via a bus; The memory stores program instructions executable by the processor, wherein the processor calls the program instructions to execute the distribution network reconstruction analysis method according to any one of claims 1 to 5.

Images