Optimal configuration method for energy storage system of power distribution network

Abstract

The invention provides an optimal configuration method for an energy storage system of a power distribution network, which can be applied to the technical field of energy storage grid connection and operation control and comprises the following steps: acquiring initial distributed power supply data and related parameters; carrying out reduction clustering on the initial distributed power supply data to obtain photovoltaic data and probability of a typical day, and taking the photovoltaic data and probability as original data; the method comprises the steps that original data and related parameters are input into a double-layer model of a power distribution network energy storage system, current photovoltaic absorption and future photovoltaic bearing capacity are comprehensively considered, a target energy storage configuration scheme is obtained, and the double-layer model comprises an energy storage configuration layer on the outer layer and a light storage operation layer on the inner layer; the light storage operation layer comprises a first-stage optimization operation model and a second-stage optimization operation model, and the energy storage configuration layer comprehensively considers the power distribution network operation comprehensive score and the photovoltaic accessible number obtained by the first-stage optimization operation model and the second-stage optimization operation model to obtain a target energy storage configuration scheme. Through double-stage arrangement of the outer layer and the inner layer, photovoltaic absorption and bearing capacity improvement are both considered.

Description

Technical Field

[0001] The present invention relates to the technical field of energy storage grid connection and operation control, and in particular to a method for optimizing configuration of a distribution network energy storage system. Background Art

[0002] In the context of whole-county photovoltaics, distributed photovoltaics (DPV) are connected to the grid on a large scale. Their randomness and volatility pose huge challenges to the safe and economic operation of the active distribution network (ADN). At the same time, the proposal of the "dual carbon" goal will lead to a new round of explosive growth in new energy power generation such as distributed photovoltaics. How to efficiently absorb the connected distributed photovoltaics and further improve the carrying capacity of distributed photovoltaics under the safe operation conditions of the active distribution network has become an urgent problem to be solved. The energy storage system (ESS) has flexible charging and discharging characteristics, which can effectively alleviate the problem of timing mismatch between distributed photovoltaic output and load demand, and can effectively alleviate the "reverse load" of distributed photovoltaic output, improve the power supply quality of the power grid, and enhance the resilience of the power grid.

[0003] At present, the purpose of energy storage configuration is mostly to promote the consumption of renewable energy. Generally, it takes economy or power quality as the goal, comprehensively considers system operation and equipment constraints, builds a two-layer model, and completes the upper and lower layer solutions based on appropriate intelligent algorithms or solvers. Hu Wei et al. proposed a two-layer optimization configuration model for energy storage considering the vulnerability of active distribution networks (Hu Wei, Yang Shuo. A two-layer optimization configuration model for energy storage considering the vulnerability of active distribution networks [J]. Smart Electric Power, 2024, 52(06): 1-8.). The process of the two-layer model is as follows Figure 1 As shown by Figure 1 It can be seen that the process of the two-layer configuration model includes: passing the energy storage configuration plan from the upper layer to the lower layer; after the lower layer determines the configuration plan, it considers the comprehensive vulnerability index of the power grid, optimizes the energy storage operation strategy and passes it to the upper layer; the upper layer then solves the economic index through the energy storage output strategy and determines the final configuration plan. This model improves the existing power supply quality while ensuring economy, but it cannot alleviate the impact of reverse load on transformers and pressure on lines caused by the further access of distributed photovoltaics in the future; it cannot give full play to the full potential of the energy storage system; at the same time, the subjective weighting method has a large degree of subjective arbitrariness and cannot accurately reflect the importance of each indicator; the weighted single-objective solution may lose information about individual objectives and their relationships, and cannot obtain the optimal solution. Summary of the invention

[0004] 1. Technical issues to be resolved

[0005] In order to solve at least one of the above-mentioned technical problems arising from the optimization configuration of the distribution network energy storage system in the prior art, an embodiment of the present invention provides a method for optimizing the configuration of the distribution network energy storage system. Through the dual-stage setting of the outer layer and the inner layer, the method takes into account both photovoltaic absorption and carrying capacity improvement.

[0006] (II) Technical solution

[0007] In view of the above technical problems, an embodiment of the present invention proposes a method for optimizing configuration of a distribution network energy storage system.

[0008] According to a first aspect of the present invention, a method for optimizing the configuration of a distribution network energy storage system is provided, comprising: obtaining initial distributed power source data and related parameters; reducing and clustering the initial distributed power source data to obtain photovoltaic data and probability of a typical day as original data; inputting the original data and related parameters into a two-layer model of the distribution network energy storage system, comprehensively considering the current photovoltaic absorption and future photovoltaic carrying capacity, and obtaining a target energy storage configuration plan, wherein the two-layer model includes an outer energy storage configuration layer and an inner photovoltaic storage operation layer, the photovoltaic storage operation layer includes a first-stage optimization operation model and a second-stage optimization operation model, the energy storage configuration layer comprehensively considers the distribution network operation comprehensive score and the number of photovoltaic access points obtained by the first-stage optimization operation model and the second-stage optimization operation model, and obtains the target energy storage configuration plan; and related parameters include distribution network grid structure, energy storage configurable nodes and capacity, photovoltaic configurable node capacity, time-of-use electricity price, unit network loss cost, unit abandoned light penalty cost, unit energy storage construction cost, unit energy storage operation and maintenance cost, and unit photovoltaic operation and maintenance cost.

[0009] In some exemplary embodiments, the original data and related parameters are input into a two-layer model of the distribution network energy storage system, and the current photovoltaic absorption and future photovoltaic carrying capacity are comprehensively considered to obtain a target energy storage configuration plan, including: using the energy storage configuration layer to randomly generate an initial energy storage configuration plan; using the energy storage configuration layer to input the original data, related parameters and the initial energy storage configuration plan into the photovoltaic storage operation layer; based on the constraints of the photovoltaic storage operation layer, considering the optimization target of the first stage, constructing the first stage optimization operation model; solving the first stage optimization operation model to obtain the first stage operation strategy and its corresponding distribution network operation comprehensive score; based on the constraints of the photovoltaic storage operation layer, considering the optimization target of the second stage, constructing the second stage optimization operation model; solving the second stage optimization operation model to obtain the second stage operation strategy and its corresponding additional photovoltaic accessibility number; using the distribution network operation index and the additional photovoltaic accessibility number as the optimization target of the energy storage configuration layer, constructing the optimization configuration model of the energy storage configuration layer; and solving the optimization configuration model of the energy storage configuration layer to obtain the target energy storage configuration plan.

[0010] In some exemplary embodiments, clustering the initial distributed power data includes: using a K-means clustering algorithm to classify the initial distributed power data with similar characteristics within a research period, and deriving the probability of occurrence of each classification.

[0011] In some exemplary embodiments, the optimization objectives of the first stage include: lowest total cost, smallest voltage deviation and lowest net load fluctuation rate, wherein the total cost includes the average daily investment and construction cost, operation and maintenance cost, network loss cost, cost of purchasing electricity from the upper power grid and cost of abandoned light; and the total cost, the voltage deviation and the net load fluctuation rate are related to the photovoltaic data and probability of a typical day.

[0012] In some exemplary embodiments, the optimization objectives of the second stage include: the photovoltaic equivalent grid-connected capacity is the highest, wherein the photovoltaic equivalent grid-connected capacity is the difference between the grid-connected capacity of distributed photovoltaics and the network loss caused by distributed photovoltaics.

[0013] In some exemplary embodiments, the constraints of the photovoltaic storage operation layer include: power flow constraints, power balance constraints, node voltage constraints, distributed photovoltaic accessible node constraints, reverse load rate constraints, and energy storage operation constraints.

[0014] In some exemplary embodiments, solving the first-stage optimization operation model includes: using the NSGA-III algorithm to optimize the output power of the photovoltaic and energy storage systems in the first-stage optimization operation model to obtain an optimization target solution set, wherein the optimization target solution set is a set of non-dominated solutions; using a superior and inferior solution distance method to consider subjective and objective weights to select the optimal solution from the optimization target solution set, wherein the subjective and objective weights are obtained using a weighting method that combines a hierarchical analysis method with an entropy weight method.

[0015] In some exemplary embodiments, solving the second-stage optimization operation model includes: using a solver to calculate power flow to optimize the second-stage output strategy.

[0016] In some exemplary embodiments, under the first-stage optimization operation model, the optimization goal of the energy storage configuration layer is to maximize the comprehensive score of distribution network operation, wherein the comprehensive score of distribution network operation is negatively correlated with the distance between the optimization target solution set and the worst optimization target solution set in the distribution network operation index; and / or under the second-stage optimization operation model, the optimization goal of the energy storage configuration layer is to maximize the accessible capacity of distributed photovoltaics, wherein the accessible capacity of distributed photovoltaics is positively correlated with the photovoltaic equivalent grid-connected capacity.

[0017] In some exemplary embodiments, solving the optimal configuration model of the energy storage configuration layer includes: using a multi-objective evolutionary algorithm based on enhanced reverse generation distance metric to design the optimal configuration model of the energy storage configuration layer.

[0018] (III) Beneficial effects

[0019] It can be seen from the above technical solutions that the method for optimizing the configuration of a distribution network energy storage system provided by the embodiment of the present invention has at least the following beneficial effects:

[0020] (1) By adopting a two-layer two-stage energy storage system configuration method, it is possible to take into account both photovoltaic consumption and carrying capacity improvement. This method comprehensively considers the operational stability, reliability, economy, greenness and further photovoltaic carrying capacity of the active distribution network. The solution strategy is efficient and can be used to achieve the site selection and capacity determination of the energy storage system. It has important guiding significance for ensuring the safe operation of the active distribution network and the realization of the "dual carbon" goals.

[0021] (2) The present invention applies a multi-objective evolutionary algorithm based on enhanced reverse generation distance metric (MOEA / IGD-NS) to solve the energy storage configuration model. The algorithm is based on biological evolution and population genetic mechanisms in nature, has a simple structure, few control parameters and significant optimization performance, optimizes the solution process and improves the solution efficiency.

[0022] (3) The first stage of the inner layer of the present invention adopts the reference point-based non-dominated genetic algorithm NSGA-III (NSGA-III), which can handle multi-objective optimization problems with four or more objectives. More optimization objectives can be added for actual operation to determine the actual photovoltaic storage operation strategy.

[0023] (4) The optimal solution selection method of the non-dominated solution set adopted in the present invention is the improved TOPSIS method that considers subjective and objective weights, and the weighting adopts the combined weighting method of AHP-anti-entropy weight method. Taking into account both expert experience and objectivity, reasonable weighting ensures the credibility of the evaluation results. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The above contents and other objects, features and advantages of the present invention will become more apparent through the following description of the embodiments of the present invention with reference to the accompanying drawings, in which:

[0025] Figure 1 The structural block diagram of the traditional energy storage double-layer optimization configuration model considering the vulnerability of the active distribution network is schematically shown;

[0026] Figure 2 A schematic diagram of a flow chart of a method for optimizing configuration of a distribution network energy storage system according to an embodiment of the present invention is shown;

[0027] Figure 3 A framework diagram of a double-layer model of a distribution network energy storage system according to an embodiment of the present invention is schematically shown;

[0028] Figure 4 A schematic diagram of a process for obtaining a target energy storage configuration solution by comprehensively considering current photovoltaic consumption and future photovoltaic carrying capacity according to an embodiment of the present invention is shown;

[0029] Figure 5 A schematic diagram of an architecture algorithm flow chart of a distribution network energy storage system optimization configuration model according to an embodiment of the present invention is shown;

[0030] Figure 6 Schematically shows a 4IEEE 33-node topology diagram according to an embodiment of the present invention; and

[0031] Figure 7 A schematic diagram of a photovoltaic daily curve according to an embodiment of the present invention is schematically shown. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical scheme and advantages of the present invention more clearly understood, the present invention is further described in detail below in combination with specific embodiments and with reference to the accompanying drawings. 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.

[0033] Figure 2 A schematic diagram of a flow chart of a method for optimizing configuration of a distribution network energy storage system according to an embodiment of the present invention is schematically shown.

[0034] Depend on Figure 2 It can be seen that a method for optimizing configuration of a distribution network energy storage system according to an embodiment of the present invention includes steps S110-S120.

[0035] In step S110, the initial distributed power source data and related parameters are obtained, wherein the related parameters include the distribution network structure, energy storage configurable nodes and capacity, photovoltaic configurable node capacity, time-of-use electricity price, unit network loss cost, unit abandoned light penalty cost, unit energy storage construction cost, unit energy storage operation and maintenance cost, and unit photovoltaic operation and maintenance cost.

[0036] In step S120, the initial distributed power source data is reduced and clustered to obtain photovoltaic data and probability of a typical day as original data.

[0037] In some exemplary embodiments, clustering the initial distributed power data includes: using a K-means clustering algorithm to classify the initial distributed power data with similar characteristics within a research period, and deriving the probability of occurrence of each classification.

[0038] For example, the K-means clustering algorithm is used to classify photovoltaic data with similar characteristics during the study period, and the data are clustered into four typical days to simulate the four seasons of spring, summer, autumn and winter in a year, and the probability of occurrence of each scene is obtained.

[0039] In step S130, the original data and related parameters are input into the double-layer model of the distribution network energy storage system, and the current photovoltaic absorption and future photovoltaic carrying capacity are comprehensively considered to obtain the target energy storage configuration plan, wherein the double-layer model includes an outer energy storage configuration layer and an inner photovoltaic storage operation layer, and the photovoltaic storage operation layer includes the first-stage optimization operation model and the second-stage optimization operation model. The energy storage configuration layer comprehensively considers the comprehensive distribution network operation score and the number of photovoltaic access points obtained by the first-stage optimization operation model and the second-stage optimization operation model to obtain the target energy storage configuration plan. Figure 3 A framework diagram of a double-layer model of a distribution network energy storage system according to an embodiment of the present invention is schematically shown.

[0040] Depend on Figure 3 It can be seen that in order to meet the constraints of safe and stable operation of the distribution network energy storage system, improve the active distribution network's absorption of distributed photovoltaics and the maximum distributed photovoltaic grid-connected capacity that the active distribution network can bear, the distribution network energy storage system model is set as a two-layer two-stage. Among them, the outer layer is the active distribution network energy storage configuration layer. The typical daily rated power of the photovoltaic unit is obtained by clustering reduction, and the energy storage configuration plan (including configuration position and capacity transfer) is proposed, and the configuration position and capacity are transferred to the inner layer; the inner layer is the photovoltaic storage operation layer, which considers the optimization objectives of the first and second stages respectively, and solves the operation strategy. The optimization objectives of the first stage are to minimize the total cost, voltage deviation, and load fluctuation rate, and the optimization objectives of the second stage are to maximize the equivalent grid-connected capacity of distributed photovoltaics, and obtain the photovoltaic storage output under the two operating modes. And the distribution network operation indicators of the first stage and the number of additional photovoltaic access in the second stage are fed back to the outer layer as decision results, and the planning scheme is determined by comprehensive analysis.

[0041] In some exemplary embodiments, step S130 includes steps S131-S138, see Figure 4 .

[0042] In step S131, an initial energy storage configuration scheme is randomly generated using the energy storage configuration layer.

[0043] In step S132, the energy storage configuration layer is used to input the original data, related parameters and the initial energy storage configuration plan into the photovoltaic storage operation layer.

[0044] In step S133, based on the constraints of the photovoltaic storage operation layer and taking into account the optimization target of the first stage, a first stage optimization operation model is constructed.

[0045] In some exemplary embodiments, the optimization objectives of the first stage include: minimum total cost, minimum voltage deviation and minimum net load fluctuation rate. The total cost includes the average daily investment and construction cost, operation and maintenance cost, network loss cost, power purchase cost from the upper power grid and abandoned light cost; and the total cost, voltage deviation and net load fluctuation rate are related to the photovoltaic data and probability of a typical day.

[0046] For example, the inner first-stage optimization operation model for photovoltaic consumption is as follows.

[0047] The objective function with the lowest total cost is:

[0048] (1)

[0049] In the formula, K is the number of clustered scenes; P(k) is the probability of scene k appearing; C plan is the total cost of active distribution network; C inv is the average daily investment and construction cost; C main C is the operation and maintenance cost; ab is the network loss cost; C buy is the cost of purchasing electricity from the upper power grid; C q The investment and construction costs are matched with the operation optimization cycle and converted into the average daily investment and construction costs.

[0050] (2)

[0051] In the formula, r ESS is the discount rate of energy storage; ESS The service life of the energy storage; The unit capacity investment cost of energy storage; is the investment cost per unit power; Grid-connected storage capacity for energy storage; is the power capacity, where:

[0052] (3)

[0053] (4)

[0054] Where B is the number of energy storage nodes; M1 is the initial number of distributed photovoltaic nodes in the active distribution network; C ESS is the unit photovoltaic power; C PV The operation and maintenance cost of energy storage charging and discharging power; is the unit network loss price; is the line loss of the system during period t; T is the operation optimization period, where:

[0055] (5)

[0056] In the formula, C buy,t is the time-of-use electricity price; P buy,t is the interactive power between the active distribution network and the upper power grid during period t, where:

[0057] (6)

[0058] In the formula, The penalty price for abandoned light per unit power; is the maximum power tracking value of distributed photovoltaics during period t; and is the actual photovoltaic power during period t.

[0059] The objective function for minimizing voltage deviation is:

[0060] (7)

[0061] Where: is the average voltage of n nodes in the scheduling period. n,t is the voltage value of the n-node in period t; and N is the total number of nodes.

[0062] The objective function for minimizing net load fluctuation is:

[0063] (8)

[0064] Among them, net load is defined as the remaining system load not served by renewable energy generation.

[0065] In the formula, is the net load of the system during period t, is the average net load value.

[0066] In some exemplary embodiments, the constraints of the photovoltaic storage operation layer include: power flow constraints, power balance constraints, node voltage constraints, distributed photovoltaic accessible node constraints, reverse load rate constraints, and energy storage operation constraints.

[0067] For example, the function of the power flow constraint is:

[0068] (9)

[0069] Where: is the active power of node a in period t; is the reactive power of node a during period t; is the voltage of node a during period t, is the voltage of node b during period t; N is the total number of nodes; is the conductance of the line between node a and node b; is the susceptance of the line between node a and node b; and is the phase angle between node a and node b.

[0070] The function of the power balance constraint is:

[0071] (10)

[0072] In the formula, is the total system load at time t.

[0073] Node voltage U n,t The constraint function is:

[0074] (11)

[0075] In the formula, and are the lower and upper limits of the node voltage.

[0076] The function of distributed photovoltaic access node constraints is:

[0077] (12)

[0078] In the formula, is the number of nodes that can be connected to the actual distributed photovoltaic system; is the maximum number of access locations for distributed photovoltaics.

[0079] The function of the reverse load rate constraint is:

[0080] According to the Guidelines for Assessment of the Carrying Capacity of Distributed Generation Connected to the Grid (DL / T 2041), if the reverse load rate is greater than 80%, the access of new distributed generation projects will be suspended until the carrying capacity of the grid is effectively improved. The calculation formula is:

[0081] (13)

[0082] Where: is the reverse load rate; Provide distributed power generation; It is the equivalent power load in the same period, that is, the load minus the output of other power sources except distributed power sources; This is the actual operating limit of the transformer or line.

[0083] The energy storage operation constraint function is:

[0084] The model of the battery energy storage system mainly considers several aspects such as charging and discharging power limits, working state constraints, and capacity update restrictions. The mathematical model is shown in equations (14) to (18):

[0085] (14)

[0086] (15)

[0087] (16)

[0088] (17)

[0089] (18)

[0090] Where: B is the number of energy storage system nodes; is the maximum charge and discharge power of the i-th energy storage system node in period t; is the minimum capacity limit of the i-th energy storage system node; is the maximum capacity limit of the i-th energy storage system node; is the remaining power of the energy storage system of the i-th energy storage system node in period t; is the charging and discharging efficiency of the i-th energy storage system node in period t; Charging efficiency for energy storage systems; is the discharge efficiency of the energy storage system; is the remaining power at the end of a scheduling cycle; Formula (18) indicates that the remaining power at the end of a scheduling cycle should be equal to the initial power at the beginning of the next scheduling cycle, so as to keep the initial value consistent.

[0091] In step S134, the first-stage optimization operation model is solved to obtain the first-stage operation strategy and its corresponding distribution network operation comprehensive score.

[0092] In some exemplary embodiments, solving the first-stage optimization operation model includes: using the NSGA-III (Non-dominated sorting genetic algorithm III, a non-dominated genetic algorithm based on a reference point) algorithm to optimize the output power of the photovoltaic and energy storage systems in the first-stage optimization operation model to obtain an optimization target solution set, wherein the optimization target solution set is a set of non-dominated solutions; using the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) method to consider subjective and objective weights to select the optimal solution from the optimization target solution set, wherein the subjective and objective weights are obtained by a weighting method that combines a hierarchical analysis method with an entropy weight method.

[0093] For example, the solution strategy of the first-stage model includes the first-stage optimization model and the improved TOPSIS method for multi-attribute decision-making. The NSGA-III algorithm is used to optimize the output power of the inner layer DPV and the energy storage system. The final optimization result of the NSGA-III method is a set of non-dominated solutions. In order to reflect the improvement of the configuration scheme for the photovoltaic absorption capacity of the active distribution network, the improved TOPSIS method considering subjective and objective weights is used to select the optimal solution. The analytic hierarchy process (AHP) is used subjectively. In order to take into account expert experience and objectivity, a combined weighting method of AHP-entropy weight method is proposed. The NSGA-III algorithm can handle multi-objective optimization problems with four or more objectives, and more optimization objectives can be added for actual operation to determine the actual photovoltaic storage operation strategy.

[0094] In step S135, based on the constraints of the photovoltaic storage operation layer and taking into account the optimization objectives of the second stage, a second stage optimization operation model is constructed.

[0095] In some exemplary embodiments, the optimization objectives of the second stage include: the photovoltaic equivalent grid-connected capacity is the highest, wherein the photovoltaic equivalent grid-connected capacity is the difference between the grid-connected capacity of distributed photovoltaics and the network loss caused by distributed photovoltaics.

[0096] For example, the photovoltaic equivalent grid-connected capacity is the difference between the grid-connected capacity of distributed photovoltaics and the network loss caused by distributed photovoltaics, which is used as the objective function:

[0097] (19)

[0098] In the formula, is the sum of the power of PV node m during its operation cycle, is the total line loss in one day.

[0099] In step S136, the second-stage optimization operation model is solved to obtain the second-stage operation strategy and its corresponding additional connectable photovoltaic quantity.

[0100] In some exemplary embodiments, solving the second-stage optimization operation model includes: using a solver to calculate power flow to optimize the second-stage output strategy.

[0101] For example, in the second stage of the inner layer, for single-objective optimization, the solver is called to calculate the power flow and optimize the output strategy.

[0102] In step S137, the distribution network operation index and the number of additional photovoltaic cells that can be connected are used as optimization targets of the energy storage configuration layer, and an optimization configuration model of the energy storage configuration layer is constructed.

[0103] In some exemplary embodiments, under the first-stage optimization operation model, the optimization target of the energy storage configuration layer is to maximize the comprehensive score of distribution network operation, wherein the comprehensive score of distribution network operation is negatively correlated with the distance between the optimization target solution set and the worst optimization target solution set in the distribution network operation index.

[0104] For example, the objective function is to maximize the comprehensive index score of the distribution network operation under the first stage operation mode, and its value is the Euclidean distance between the final compromise solution set selected by the improved TOPSIS method and the worst solution set.

[0105] (20)

[0106] In the formula, It is the comprehensive index score of photovoltaic absorption capacity of active distribution network;

[0107] is the optimal solution selected by the improved TOPSIS method; f h is the inner one-stage optimization target solution set corresponding to the optimal solution; H is the number of inner one-stage optimization targets; f h- is the set of solutions for the worst optimization objective.

[0108] In the embodiment of the present invention, the optimal solution selection method of the non-dominated solution set is the improved TOPSIS method considering subjective and objective weights, and the weighting adopts the combined weighting method of AHP-anti-entropy weight method. Taking into account both expert experience and objectivity, reasonable weighting ensures the credibility of the evaluation results. Among them, the equivalent grid-connected capacity is the objective function of the inner two-stage, and the additional photovoltaic access number is the second objective function of the outer layer. The two-stage operation is based on the maximum equivalent grid-connected capacity, and then the corresponding additional photovoltaic access number is passed to the outer layer.

[0109] In some exemplary embodiments, under the second-stage optimization operation model, the optimization goal of the energy storage configuration layer is to maximize the accessible capacity of distributed photovoltaics, wherein the accessible capacity of distributed photovoltaics is positively correlated with the photovoltaic equivalent grid-connected capacity.

[0110] For example, the objective function is to maximize the distributed photovoltaic access capacity in the second stage operation mode.

[0111] (twenty one)

[0112] Where: is the distributed photovoltaic access capacity of node m, is the number of nodes that can be connected to distributed photovoltaics.

[0113] In step S138, the optimization configuration model of the energy storage configuration layer is solved to obtain the target energy storage configuration solution.

[0114] In some exemplary embodiments, solving the optimal configuration model of the energy storage configuration layer includes: using a multi-objective evolutionary algorithm based on an enhanced inverted generational distance metric (MOEA / IGD-NS, Multi-objective evolutionary algorithm based on an enhanced inverted generational distance metric) to design the optimal configuration model of the energy storage configuration layer to solve the energy storage planning problem that comprehensively considers the two stages.

[0115] The multi-objective evolutionary algorithm based on enhanced reverse generation distance metric (MOEA / IGD-NS) is based on the biological evolution and population genetic mechanism in nature. It has a simple structure, few control parameters and significant optimization performance, which optimizes the solution process and improves the solution efficiency.

[0116] The inner and outer layer algorithms in the embodiment of the present invention may select more common multi-objective algorithms, including multi-objective genetic algorithms, multi-objective particle swarm algorithms, etc.

[0117] Figure 5 The flowchart schematically shows an architecture algorithm of a distribution network energy storage system optimization configuration model according to an embodiment of the present invention.

[0118] Depend on Figure 5 It can be seen that according to an architecture algorithm of a distribution network energy storage system optimization configuration model according to an embodiment of the present invention, by nesting the inner layer optimization between the outer layer optimization population acquisition and population evaluation, the population configuration plan is passed to the inner layer to obtain the corresponding objective function value of the population, which is used to evaluate the quality of the population.

[0119] The distribution network energy storage system optimization configuration model according to the embodiment of the present invention takes into account both photovoltaic consumption and carrying capacity improvement. The method comprehensively considers the operational stability, reliability, economy, greenness and further photovoltaic carrying capacity of the active distribution network, and the solution strategy is efficient, which realizes the site selection and capacity determination of the energy storage system, and has important guiding significance for ensuring the safe operation of the active distribution network and the realization of the "dual carbon" goals. The energy storage system can adopt other energy storage technologies besides electrochemical energy storage, such as hydrogen energy storage, pumped storage, compressed air, etc.

[0120] Case analysis:

[0121] In order to verify the feasibility of the energy storage dual-layer configuration model proposed in the embodiment of the present invention, an improved IEEE 33 node model active distribution network system is used to configure two energy storage devices to access the power system as an example. Figure 6 .

[0122] Scenario generation results and simulation parameters:

[0123] The IEEE 33-node active distribution network system has a base voltage of 12.66 kV, a base capacity of 1 MVA, a line power limit of 6 MW, and a node voltage amplitude upper and lower limits set to 1.07 and 0.93 pu, respectively. The total system load is 3715 kW + j2300 kvar. A single photovoltaic unit with a capacity of 50 kW and a single 2h system energy storage unit with a capacity of 10 kW h / 5 kW are selected. The maximum number of connected units of a single node energy storage system is set to 150, and the initial state of charge is 50%. The subjective weights of the indicators are shown in Table 1. The four typical daily unit photovoltaic power curves obtained by K-means clustering are shown in Figure 7 The parameter settings of the model built in this example are shown in Table 2 and Table 3.

[0124] Table 1 Subjective weights of indicators

[0125]

[0126] Table 2 Time-of-use electricity prices

[0127]

[0128] Table 3 Related parameters

[0129]

[0130] Analysis of energy storage configuration results:

[0131] The following four scenarios are set to analyze the changes in operating parameters before and after connecting to energy storage, as well as the impact of optimization targets on energy storage planning results by considering photovoltaic consumption and improving carrying capacity: Scenario 1, without considering energy storage system connection; Scenario 2, considering energy storage system connection, and the optimization target does not consider improving carrying capacity; Scenario 3, considering energy storage system connection, and the optimization target does not consider photovoltaic consumption; and Scenario 4, considering energy storage system connection, using the method of the embodiment of the present invention, the optimization target considers photovoltaic consumption and carrying capacity improvement. Taking the penetration rate of 25% as an example, the energy storage capacity configuration and optimization results under each scenario are summarized in Table 4.

[0132] Table 4 Results after configuring energy storage

[0133]

[0134] Comparing scenes 1 to 4 in Table 4, it can be found that in scene 4, the method of the embodiment of the present invention is adopted, and the photovoltaic absorption and photovoltaic carrying capacity improvement are comprehensively considered during planning. Although scene 4 is slightly higher in cost than scene 2, it also shows better control in voltage deviation and net load fluctuation rate, and further improves the photovoltaic access capacity, realizing the comprehensive optimization of energy storage configuration. In addition, the energy storage configuration capacity of scene 4 is relatively small, which provides greater flexibility for the future expansion and upgrading of the active distribution network. With the advancement of technology and the reduction of costs, the active distribution network can more easily increase the energy storage capacity to meet future needs. Combining the two perspectives of the present and the future, it can be seen that the scheme proposed by the method of the embodiment of the present invention is the optimal configuration scheme, which verifies the effectiveness of the present invention.

[0135] The above specific embodiments further illustrate the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above are only specific embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for optimizing the configuration of a distribution network energy storage system. It is characterized in that include: Obtain initial distributed power data and related parameters; The initial distributed power data is reduced and clustered to obtain photovoltaic data and probability of a typical day as original data; The original data and the related parameters are input into the double-layer model of the distribution network energy storage system, and the current photovoltaic consumption and future photovoltaic carrying capacity are comprehensively considered to obtain the target energy storage configuration plan. The double-layer model includes an outer energy storage configuration layer and an inner photovoltaic storage operation layer. The photovoltaic storage operation layer includes a first-stage optimization operation model and a second-stage optimization operation model. The energy storage configuration layer comprehensively considers the distribution network operation comprehensive score and the number of photovoltaic accessibility obtained by the first-stage optimization operation model and the second-stage optimization operation model to obtain a target energy storage configuration plan. as well as The relevant parameters include distribution network structure, energy storage configurable nodes and capacity, photovoltaic configurable node capacity, time-of-use electricity price, unit network loss cost, unit abandoned light penalty cost, unit energy storage construction cost, unit energy storage operation and maintenance cost, and unit photovoltaic operation and maintenance cost.

2. The method according to claim 1, It is characterized in that The raw data and the related parameters are input into the double-layer model of the distribution network energy storage system, and the current photovoltaic consumption and future photovoltaic carrying capacity are comprehensively considered to obtain the target energy storage configuration scheme, including: Using the energy storage configuration layer, randomly generating an initial energy storage configuration plan; Using the energy storage configuration layer, the raw data, the relevant parameters and the initial energy storage configuration scheme are input into the photovoltaic storage operation layer; Based on the constraints of the photovoltaic storage operation layer, considering the optimization target of the first stage, constructing the first stage optimization operation model; Solve the first-stage optimization operation model to obtain the first-stage operation strategy and its corresponding distribution network operation comprehensive score; Based on the constraints of the photovoltaic storage operation layer, considering the optimization target of the second stage, constructing the second stage optimization operation model; Solve the optimization operation model of the second stage to obtain the operation strategy of the second stage and the corresponding number of additional photovoltaic access points; Taking the distribution network operation index and the additional photovoltaic access quantity as the optimization target of the energy storage configuration layer, constructing an optimization configuration model of the energy storage configuration layer; and Solve the optimization configuration model of the energy storage configuration layer to obtain the target energy storage configuration solution.

3. The method according to claim 2, It is characterized in that The reducing and clustering of the initial distributed power data includes: using a K-means clustering algorithm to classify the initial distributed power data with similar characteristics within a research period, and deriving the probability of occurrence of each classification.

4. The method according to claim 2, It is characterized in that The optimization objectives of the first stage include: minimum total cost, minimum voltage deviation and minimum net load fluctuation rate. The total cost includes the average daily investment and construction cost, operation and maintenance cost, network loss cost, power purchase cost from the upper power grid and abandoned light cost; and The total cost, the voltage deviation, and the net load fluctuation rate are related to photovoltaic data and probabilities on a typical day.

5. The method according to claim 2, It is characterized in that The optimization objectives of the second phase include: the highest photovoltaic equivalent grid-connected capacity, The photovoltaic equivalent grid-connected capacity is the difference between the grid-connected capacity of distributed photovoltaics and the network loss caused by distributed photovoltaics.

6. The method according to claim 2, It is characterized in that The constraints of the photovoltaic storage operation layer include: power flow constraints, power balance constraints, node voltage constraints, distributed photovoltaic accessible node constraints, reverse load rate constraints and energy storage operation constraints.

7. The method according to claim 2, It is characterized in that The solution of the first stage optimization operation model includes: The NSGA-III algorithm is used to optimize the output power of the photovoltaic and energy storage systems in the first-stage optimization operation model to obtain an optimization target solution set, wherein the optimization target solution set is a set of non-dominated solutions; The best solution is selected from the optimization target solution set by taking into account the subjective and objective weights using the superior and inferior solution distance method. The subjective and objective weights are obtained by a weighting method that combines the analytic hierarchy process with the entropy weight method.

8. The method according to claim 2, It is characterized in that The solving of the second stage optimization operation model includes: using a solver to calculate the power flow to optimize the second stage output strategy.

9. The method according to claim 7, It is characterized in that Under the first-stage optimization operation model, the optimization target of the energy storage configuration layer is to maximize the comprehensive score of distribution network operation, wherein the comprehensive score of distribution network operation is negatively correlated with the distance between the optimization target solution set and the worst optimization target solution set in the distribution network operation index; and / or Under the second-stage optimization operation model, the optimization goal of the energy storage configuration layer is to maximize the distributed photovoltaic accessible capacity, wherein the distributed photovoltaic accessible capacity is positively correlated with the photovoltaic equivalent grid-connected capacity.

10. The method according to claim 2, It is characterized in that The method of solving the optimal configuration model of the energy storage configuration layer includes: using a multi-objective evolutionary algorithm based on enhanced reverse generation distance metric to design the optimal configuration model of the energy storage configuration layer.

Images