Joint planning method and system for distributed photovoltaic and energy storage systems in distribution networks

Abstract

The present application relates to a kind of joint planning method and system of distributed photovoltaic and energy storage system in power distribution network.The present application determines the number of power distribution network typical scenarios according to the node distributed photovoltaic and energy storage system data obtained using information entropy, then extracts typical scenarios using K-medoids clustering algorithm according to the number of power distribution network typical scenarios, then generates the joint planning model of distributed photovoltaic and energy storage system based on the extracted typical scenarios under the condition of meeting objective function, then solves the joint planning model of distributed photovoltaic and energy storage system using COBL-PSO algorithm to obtain the joint planning scheme of distributed photovoltaic and energy storage system, and realizes the planning control of distributed photovoltaic and energy storage system based on the obtained joint planning scheme, thereby being able to improve the stability of operating state under the premise of shortening the calculation time length of planning scheme.

Description

Technical Field

[0001] The present invention relates to the technical field of joint planning of distribution networks, and in particular to a joint planning method and system for distributed photovoltaic and energy storage systems in a distribution network. Background Art

[0002] With the rapid growth of wind and photovoltaic power generation in power system operations, the number of distributed generation sources connected to distribution networks has increased annually. The emergence of new loads such as electric vehicles and electric heating, along with the continued advancement of energy substitution, pose significant challenges to distribution network planning. As renewable energy generation and new loads continue to increase, their inherent randomness and volatility will impact the grid. Therefore, distribution networks must gradually improve their operational capabilities after renewable energy integration to effectively address the resulting diversification of operating scenarios, complexity of operating modes, and randomized operating states. Furthermore, the unpredictability of the scale and region of energy substitution will further complicate distribution network operations. To fully leverage the efficiency of all assets and equipment in the grid, meet the increasing integration of renewable energy and the growing load demand, and maximize grid operational efficiency, it is necessary to rationally integrate and regulate renewable energy sources to develop more efficient distribution network planning solutions.

[0003] With the gradual promotion of distributed power sources in the distribution network, the local absorption capacity of new energy power generation systems and the ability to ensure the safe and stable operation of the power system need to be further improved. Based on the bidirectional power characteristics and flexible adjustment capabilities of the energy storage system, the operation of the power grid can be guaranteed through effective planning and technology. In the existing technology, the power system planning method based on direct processing of a large amount of initial system data is widely used. By analyzing a single section (deterministic scenario), there will not only be too much unfiltered raw data, but also the planning scheme calculation takes too long. In addition, without considering the power system, it is very likely that there will be situations that affect the operating status. Therefore, there is an urgent need to provide a more efficient method to better deal with the new problems arising in the current power system. Summary of the Invention

[0004] In order to solve the above problems existing in the prior art, the present invention provides a method and system for joint planning of distributed photovoltaic and energy storage systems in a distribution network.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] A joint planning method for distributed photovoltaic and energy storage systems in a distribution network, comprising:

[0007] Obtain node distributed photovoltaic and energy storage system data;

[0008] Using information entropy to determine the number of typical distribution network scenarios based on the distributed photovoltaic and energy storage system data at the node;

[0009] The K-medoids clustering algorithm is used to extract typical scenarios based on the typical scenarios of the distribution network to obtain typical scenarios;

[0010] Based on the extracted typical scenarios and under the condition of meeting the objective function, a joint planning model for distributed photovoltaic and energy storage systems is generated;

[0011] The COBL-PSO algorithm is used to solve the joint planning model of the distributed photovoltaic and energy storage system to obtain a joint planning scheme of the distributed photovoltaic and energy storage system.

[0012] Preferably, the method of using information entropy to determine the number of typical distribution network scenarios based on the node distributed photovoltaic and energy storage system data specifically includes:

[0013] Obtaining a power distribution density function of a scene to be planned, and using the scene power distribution density function as an initial root node;

[0014] Determining the information entropy of the initial root node;

[0015] Determining the split position of the initial root node with the goal of maximizing the reduction of the information entropy;

[0016] Splitting the root node based on the split position to obtain a split node;

[0017] Determining the information entropy of the segmentation node;

[0018] Determine whether the information entropy of the split node meets the stopping condition;

[0019] When the information entropy of the split node meets the stopping condition, the split node is used as a leaf node;

[0020] When the information entropy of the split node does not meet the stopping condition, the current split node is used as the new initial root node and the process returns to "determining the information entropy of the initial root node" until the information entropy of all split nodes meets the stopping condition;

[0021] The number of leaf nodes is determined, and the number of leaf nodes is used as the number of typical scenarios of the distribution network.

[0022] Preferably, the stopping condition is: the information entropy of the split node is less than a critical value.

[0023] Preferably, the method of extracting typical scenarios based on the typical scenarios of the distribution network using the K-medoids clustering algorithm specifically includes:

[0024] Take a leaf node as a typical scene set to generate dataset D;

[0025] The K-medoids clustering algorithm is used to cluster the dataset D to obtain typical scenarios.

[0026] Preferably, the objective function is:

[0027] min C = min(C1, C2, C3, C4);

[0028] Among them, C is the objective function value, C1 is the economic target, C2 is the reliability target, C3 is the environmental target, and C4 is the coordination target.

[0029] Preferably, the reliability target C2 is determined as follows:

[0030] Determine node voltage stability and load stability;

[0031] Generate a reliability target C2 based on the node voltage smoothness and the load smoothness:

[0032] C2=f(SSD U ,SSD l );

[0033] Among them, f(*) objective function, SSD U is the node voltage stability, SSD l For load stability.

[0034] Preferably, the COBL-PSO algorithm is used to solve the joint planning model of the distributed photovoltaic and energy storage system based on constraints to obtain a joint planning scheme of the distributed photovoltaic and energy storage system; the constraints include: power flow constraints, line transmission power constraints, node voltage constraints, single-node distributed photovoltaic grid-connected capacity constraints, distribution network distributed photovoltaic grid-connected total amount constraints and energy storage system-related constraints.

[0035] Preferably, the COBL-PSO algorithm is used to solve the joint planning model of the distributed photovoltaic and energy storage system based on the constraints to obtain a joint planning scheme of the distributed photovoltaic and energy storage system, specifically including:

[0036] Use initialization parameters to initialize the particle swarm optimization algorithm;

[0037] Determining calculation parameters of the joint planning model of the distributed photovoltaic and energy storage system, and setting a period corresponding to the joint planning model of the distributed photovoltaic and energy storage system;

[0038] Configure the initial particle swarm parameters in the particle swarm optimization algorithm, update the particle speed and particle position, and record the updated particle speed and position;

[0039] Using the objective function in the joint planning model of the distributed photovoltaic and energy storage system as the fitness function of the particle swarm optimization algorithm to determine the fitness of particle generation;

[0040] Select the particle with the best fitness to update and obtain the updated particle, and record the fitness of the updated particle and the individual optimal position of the updated particle;

[0041] Determining a global optimal position based on the fitness of the updated particle and the individual optimal position of the updated particle, wherein the global optimal position is the output power of the distributed photovoltaic and energy storage system in a typical scenario that minimizes the objective function value;

[0042] Determine whether the output power of the distributed photovoltaic and energy storage system meets the calculation limit or whether the current number of iterations exceeds the maximum number of iterations;

[0043] If the output power of the distributed photovoltaic and energy storage system meets the calculation limit or the current number of iterations does not exceed the maximum number of iterations, the output power of the distributed photovoltaic and energy storage system is output as the corresponding output value of the distributed photovoltaic and energy storage system;

[0044] If the output power of the distributed photovoltaic and energy storage system does not meet the calculation limit or the current number of iterations exceeds the maximum number of iterations, the process returns to executing "update particle speed and particle position".

[0045] Prior to this, the process of updating the particle's velocity and position is:

[0046] If the b-th velocity component of particle a at the t-th iteration is greater than the preset velocity value, then the b-th velocity component of particle a at the t-th iteration is set to be equal to the preset velocity value; if the b-th velocity component of particle a at the t-th iteration is less than the opposite of the preset velocity value, then the b-th velocity component of particle a at the t-th iteration is set to be equal to the opposite of the preset velocity value, and the iteration is repeated;

[0047] The particle swarm center of gravity is determined, and the updated direction solution of each particle is determined based on the particle swarm center of gravity.

[0048] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0049] The present invention provides a joint planning method for distributed photovoltaic and energy storage systems in a distribution network. After using information entropy to determine the number of typical distribution network scenarios based on the acquired node distributed photovoltaic and energy storage system data, a K-medoids clustering algorithm is used to extract scenarios based on the typical distribution network scenarios to obtain typical scenarios. Then, based on the extracted typical scenarios, a joint planning model of the distributed photovoltaic and energy storage system is generated under the condition that the objective function is satisfied. Then, the COBL-PSO algorithm is used to solve the joint planning model of the distributed photovoltaic and energy storage system to obtain a joint planning scheme of the distributed photovoltaic and energy storage system. Based on the obtained joint planning scheme, planning and control of the distributed photovoltaic and energy storage system are realized, thereby improving the stability of the operating state while shortening the calculation time of the planning scheme.

[0050] In addition, the present invention also provides a joint planning system for distributed photovoltaic and energy storage systems in a distribution network, the system comprising:

[0051] Data acquisition module, used to obtain node distributed photovoltaic and energy storage system data;

[0052] A number determination module is used to determine the number of typical distribution network scenarios based on the node distributed photovoltaic and energy storage system data using information entropy;

[0053] A scenario extraction module is used to extract typical scenarios based on the typical scenarios of the distribution network using a K-medoids clustering algorithm;

[0054] The model generation module is used to generate a joint planning model for distributed photovoltaic and energy storage systems based on the extracted typical scenarios while satisfying the objective function;

[0055] The scheme generation module is used to solve the joint planning model of the distributed photovoltaic and energy storage system by using the COBL-PSO algorithm to obtain a joint planning scheme of the distributed photovoltaic and energy storage system.

[0056] Since the technical effects achieved by the joint planning system of distributed photovoltaic and energy storage systems in the distribution network provided by the present invention are the same as the technical effects achieved by the joint planning method of distributed photovoltaic and energy storage systems in the distribution network provided above, they will not be described in detail here. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0058] Figure 1Flowchart of the joint planning method of distributed photovoltaic and energy storage systems in the distribution network provided by the present invention;

[0059] Figure 2 A flowchart of using information entropy to determine the number of typical scenarios provided by an embodiment of the present invention;

[0060] Figure 3 A flowchart of extracting typical scenarios using the K-medoids clustering algorithm provided in an embodiment of the present invention;

[0061] Figure 4 This is a calculation flow chart of the COBL-PSO algorithm provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0062] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0063] The purpose of the present invention is to provide a method and system for joint planning of distributed photovoltaic and energy storage systems in a distribution network, which can effectively deal with new problems arising in the current power system.

[0064] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0065] like Figure 1 As shown, the present invention provides a joint planning method for distributed photovoltaic and energy storage systems in a distribution network, including:

[0066] Step 100: Obtain node distributed photovoltaic and energy storage system data.

[0067] Step 101: Use information entropy to determine the number of typical distribution network scenarios based on node distributed photovoltaic and energy storage system data.

[0068] Due to the defects of the single-section data analysis method, such as large data processing amount, long data processing time, large calculation error and failure to closely combine with the electrical characteristics of the distribution network, the information entropy theory is introduced in the step to effectively process the power system uncertainty problem caused by the voltage fluctuation of the distributed photovoltaic system after being connected to the distribution network and effectively represent the power system characteristics during data processing. The method for extracting the typical planning scenario of the power system is improved by selecting the maximum entropy value change amount. The original data scenario is divided into multiple scene subclasses by using a recursive method, and then a typical data point in each scene subclass data point set is selected as the typical scene of the scene subclass. The typical scene set of all scene subclasses is the typical scene set of the planning system, as shown in Figure 2 The specific implementation steps are as follows:

[0069] Firstly, the power distribution density function information of the scene to be planned is input as the initial root node of the algorithm, and the information entropy of the initial root node is calculated:

[0070]

[0071] Wherein, H is the node information entropy, v0 is the initial root node, x is the power, x max is the maximum power, and p(x) is the corresponding probability density function.

[0072] Secondly, according to the selection principle of the node segmentation position, that is, the maximum information entropy reduction amount ΔH, the segmentation position of the initial root node is calculated and selected, and the expression is as follows:

[0073] max(ΔH)=max[H(v0)-H(v0,x0)]

[0074] Wherein, ΔH is the information entropy reduction amount.

[0075] Correspondingly, the information entropy of the segmented node is obtained by summing the information entropies of the two new nodes (i.e. the segmented node) generated after the node is segmented, and the mathematical expression is as follows:

[0076]

[0077] Wherein, x0 is the node segmentation position, H(v0,x0) is the information entropy after the node is segmented, H(v0′) and H(v0″) are the node information entropy values of the left and right child nodes respectively, and p′(x) and p″(x) are the probability density functions of the left and right child nodes respectively.

[0078] After the above calculation, after the segmentation position is selected and two new nodes are generated, the segmented node parameters are further verified. The node parameter verification method selects a comparison threshold εH To complete the judgment method, the node information entropy of the node that meets the application requirements should be less than the critical value ε H If the node information entropy value of the split node to be tested is less than the critical value ε H , the split node can be called a "leaf node." A node with low information entropy indicates a more uniform internal power distribution. This means that power acquisition at this node will yield similar results, rather than significant power discrepancies. Therefore, for leaf nodes, selecting typical scenarios for this type of node can eliminate complex calculations and directly select the expected power distribution for that node. A leaf node represents a typical scenario set.

[0079] If the node information entropy of the split node to be tested is greater than the critical value ε H , meaning it doesn't meet the application requirements, then the split node needs to be used as a new root node and the split calculation needs to be repeated using the above method until all generated nodes can be used as leaf nodes. The number of leaf nodes is the number of typical scenarios, K. Therefore, the above information entropy-based method can be used to determine the number of typical distribution network scenarios, K.

[0080] Step 102: Use the K-medoids clustering algorithm to extract typical scenarios based on the distribution network to obtain typical scenarios.

[0081] After determining the number of typical scenarios, it is necessary to further use typical scenario extraction technology to reduce the dimensionality of the existing high-dimensional data. This method extracts typical scenarios while retaining the key features of the data, thereby reducing the amount of data. In this step, the present invention mainly uses the K-medoids clustering algorithm to achieve scenario reduction. This algorithm uses the object closest to the cluster center to represent the cluster center. It can retain the data characteristics of each cluster scenario while making them dissimilar to each other. It also eliminates the negative impact of possible singular points on clustering, and has the advantages of low complexity and high efficiency.

[0082] Assume that the total number of samples in the dataset D is n, the number of clusters is the number of typical scenarios K obtained by information entropy in the previous section, and the data point m is moved to the centroid c. k The distance is recorded as d(x m ,c k ).like Figure 3 As shown in the figure, the implementation steps of the K-medoids clustering algorithm are as follows:

[0083] (1) Randomly select a data point from the data set containing photovoltaics and loads as the first initial centroid, denoted as c1.

[0084] (2) Select the kth centroid c from the data set D according to the probability P k(k≥2), the farther away the point is from the centroid, the greater the probability of being selected as the new centroid. The calculation formula of probability P is as follows:

[0085]

[0086] Where C p represents the center of mass c p The dataset of the cluster, x m and x h All belong to dataset C p , x m is the data point m, x h is the data point h.

[0087] (3) Based on the distance from each point to the centroid and the proximity principle, each point is assigned to the most appropriate centroid representative cluster.

[0088] (4) In each cluster, according to the formula Calculate the Euclidean distance d and determine the new centroid based on the principle of minimum sum of Euclidean distances. h1 Indicates the h-th dimension coordinate of the first point. h2 Represents the h-th coordinate of the second point, where w is the dimension, h = 1, 2, 3, ..., w.

[0089] (5) Repeat steps (3) and (4) until the center of mass no longer changes.

[0090] (6) Output the best clustering result.

[0091] Step 103: Based on the extracted typical scenarios and under the condition that the objective function is satisfied, a joint planning model of the distributed photovoltaic and energy storage system is generated.

[0092] Specifically, this step mainly combines the typical scenarios identified above to generate a joint planning model for distributed photovoltaic and energy storage systems with the goals of meeting grid operation reliability, environmental protection, economy, and coordination. The comprehensive objective function expression is as follows:

[0093] minC=min(C1,C2,C3,C4)

[0094] Where C1~C4 are the economic, reliability, environmental protection and coordination goals respectively.

[0095] (1) Economic objectives

[0096] In the power system's distribution network construction planning, distributed photovoltaic and energy storage systems mainly include three parts: construction, operation and maintenance, and recovery. Therefore, when constructing a joint planning model for distributed photovoltaic and energy storage systems, it is necessary to consider the investment and construction costs, operation and maintenance costs, and power purchase costs of the distribution network, as well as the benefits brought by the energy storage system. The mathematical expression of the economic objective is as follows:

[0097] C1=C invest +C operation +C buy -C pro

[0098] Where C invest is the investment and construction cost, C operation is the operation and maintenance cost, C buy is the cost of purchasing electricity from the power grid, C pro For the benefit of energy storage system.

[0099]

[0100] Where r0 is the discount rate, which is 0.06. y is the applicable planning period, which is generally 20 years for distributed photovoltaics and 10 years for energy storage systems. The unit capacity distributed photovoltaic construction cost, is the construction cost of the energy storage system per unit capacity. is the actual grid-connected capacity of distributed photovoltaics at node j in scenario k, is the actual grid-connected capacity of the energy storage system at node j in scenario k.

[0101] C operation =λ·C invest

[0102] Where λ is the conversion ratio of operation and maintenance cost, which is set to 0.1.

[0103]

[0104] Where, is the amount of electricity purchased from the grid by node j in scenario k, It is the real-time electricity price per unit capacity of electricity purchased from the main grid.

[0105]

[0106] Where, are the loads before and after the energy storage system acts on node j in scenario k, and Δt is the unit time.

[0107] (2) Reliability goals

[0108] Reliability currently uses parameters used to characterize power quality and system stability to represent C2. The mathematical expression is as follows:

[0109] C2=f(SSD U ,SSD l ).

[0110] In this step, a function characterizing power quality and system stability is derived based on the joint planning characteristics of the distributed photovoltaic and energy storage systems.

[0111] The present invention takes into account that after distributed power sources and other renewable energy generation systems are widely connected to the power system, the volatility and randomness of renewable energy generation will affect the stability of the power system, causing drastic changes in user-side load fluctuations, thereby affecting the voltage of the corresponding nodes in the power system and causing fluctuations, thus undermining the operational stability of the power system. Therefore, the present invention uses two specific parameters, node voltage stability and load stability, to quantitatively characterize the power quality and power system stability in the power system. The corresponding expressions are as follows:

[0112] 1) Node voltage stability SSD U for:

[0113]

[0114] Where: U i,t is the voltage of node i at time t. is the average voltage of node i during the cycle.

[0115] 2) Load smoothness SSD1 is:

[0116]

[0117] Where: P i,t is the load power at time t. is the average load power within the cycle.

[0118] (3) Environmental protection goals

[0119] After the distribution network is completed and put into operation, it will inevitably produce carbon emissions. However, the integration of new energy generation systems such as distributed photovoltaics can effectively achieve a certain degree of carbon emission reduction for the distribution network. Energy storage systems can also be used to improve the energy absorption capacity of the distribution network, further improving the distribution network's emission reduction capabilities and reducing carbon emissions. Therefore, this paper chooses to use carbon emission cost as the corresponding function C3 for the environmental protection objective, which is expressed as follows:

[0120]

[0121] Where, is the unit penalty cost of carbon emissions during the operation of the distribution network, Cfl is the initial carbon emission of the grid, C BESS is the carbon emission reduction of the energy storage system, C dec It is the net carbon emission reduction of photovoltaic and traditional power sources.

[0122]

[0123] Where λ c is the carbon emission conversion factor. is the device power consumption of node j in scene k, is the line loss between nodes ij in scenario k.

[0124]

[0125] Where, is the carbon emissions per unit of electricity generated by distributed photovoltaics. is the traditional power output of node j in scenario k. is the distributed photovoltaic output power of node j in scenario k. pv is the carbon emission reduction of photovoltaics, C pow is the carbon emissions per unit of power.

[0126]

[0127] In the formula, C0 is the remaining carbon content of the energy storage system, and E0 is the remaining power of the energy storage system. ES is the discharge rate of the energy storage system, which is taken as 95%. is the difference between the discharge and charging power of the energy storage system at node j in scenario k.

[0128] (4) Coordination goals

[0129] In power system planning, the development levels of sources, networks, loads, and storage must match each other. The load rates of lines and transformers in the distribution network reflect the carrying capacity of the corresponding devices. The higher the load rate, the stronger the ability of the devices to coordinate resources with each other. If there are large differences in the distribution of line or transformer load rates in the power system, there will be differences in the load-bearing capacity of the distribution network, which will bring great challenges to the load coordination and stable operation of the distribution network. If the distribution of line or transformer load rates in the system tends to be consistent, the ability of the distribution network to bear loads will be more uniform, which can better meet future load growth and adapt to random changes in power generation, power grids, and loads. This indicates that the coordination of the current system is better. The mathematical formula is:

[0130]

[0131] Where, is the load rate of line i' in the planning period under scenario m', Load rate of distribution transformer j' in planning period under scenario m'; LR l Line average load rate; LR transf Transformer average load rate; Pm' is the probability of scenario m' occurrence.

[0132] Step 104: the joint planning model of the distributed photovoltaic and energy storage system is solved by using the COBL-PSO algorithm to obtain the joint planning scheme of the distributed photovoltaic and energy storage system.

[0133] As a current widely used intelligent optimization algorithm, the PSO algorithm has good adaptability, strong robustness, simple and flexible excellent characteristics, the COBL algorithm is a good learning algorithm, and the optimization algorithm can be used by matching, and the optimization effect of the optimization algorithm can be significantly improved. Therefore, the Centroid Opposition-based Learning (COBL) and Particle Swarm Optimization (PSO) are combined, the excellent characteristics of the two algorithms are considered, and the COBL-PSO algorithm is constructed to solve the planning model, as shown in the solving process: Figure 4

[0134] (1) the PSO algorithm is initialized, and the required parameters for initialization are obtained: population size N, maximum iteration number T, learning factor c1, c2, particle flight speed range [-v m ,v m ].

[0135] (2) the calculation parameters of the joint planning model of the distributed photovoltaic and energy storage system are determined, and the period corresponding to the model is set.

[0136] (3) the initial particle swarm parameters in the PSO algorithm are configured, and the particle velocity and position are recorded after being updated.

[0137] (4) the particle velocity and position are calculated and updated by using the following formula.

[0138] v ab (t+1)=wv ab (t)+c1r1[p best,ab (t)-x ab (t)]+c2r2[g best,ab (t)-x ab (t)]

[0139] x ab (t+1)=x ab (t)+v ab (t+1) ​

[0140] If the velocity component v ab (t) > v m , let v ab (t) = v m . If v ab (t) < -v m , let v ab (t) = -v m (t) and repeat the iteration.

[0141] (4) Calculate the center of gravity of the particle swarm and further calculate the direction solution of each particle after updating

[0142] (5) The fitness function of the PSO algorithm is selected to calculate the objective function in the joint planning model of the distributed photovoltaic and energy storage system. After the fitness of the particles generated by the calculation is compared, the particle with the best fitness is updated, and the fitness of the corresponding particle and the individual optimal position of the particle are recorded. The global optimal position of the particle is further calculated, that is, the output power of the distributed photovoltaic and energy storage system in the typical scenario that can minimize the objective function value.

[0143] (6) Limit value judgment is performed on the calculated output power of the distributed photovoltaic and energy storage system to determine whether the calculated value in step (5) meets the calculation limit or the current iteration number of this calculation exceeds the maximum iteration number limit. If yes, the calculated value is output as the final output value of the distributed photovoltaic and energy storage system, and the operation is ended. If no, return to step (4) to perform iterative operation until the above requirements are met.

[0144] Further, when solving the joint planning model of the distributed photovoltaic and energy storage system, the constraint conditions are also used to constrain the calculation results of the particles. The constraint conditions include:

[0145] 1) Power flow constraint

[0146]

[0147] In the formula, P i is the active power of node i, Q i is the reactive power of node i. U i and U j are the voltages of nodes i and j, respectively. B ij is the line conductance between nodes i and j, G ij is the line susceptance between nodes i and j. θ ij is the voltage phase angle between nodes i and j.

[0148] 2) Line transmission power constraints

[0149] 0<P l ≤P l,max

[0150] Where, P l is the actual value of line transmission power, P l,max The upper limit of the line transmission power.

[0151] 3) Node voltage constraints

[0152] U i,min ≤U i ≤U i,max

[0153] Where U i,min 、U i,max are the upper and lower voltage limits of node i respectively.

[0154] 4) Single-node distributed photovoltaic grid-connected capacity constraints

[0155]

[0156] In the formula is the upper limit of the grid-connected capacity of the distributed photovoltaic node at node i, P i PV It is the grid-connected capacity of distributed photovoltaic power generation at a single node.

[0157] 5) Constraints on the total amount of distributed photovoltaic grid-connected to the distribution network

[0158]

[0159] Where, P i load is the load power of node i. PV is the ratio coefficient of distributed photovoltaic grid-connected to the total load.

[0160] 6) Energy storage system-related constraints

[0161] SOC min ≤SOC≤SOC max

[0162] 0≤P ES (t)≤P ES-max

[0163]

[0164] Where, SOC min , SOC max are the upper and lower limits of the total capacity SOC of the energy storage system respectively. ES-maxThe unit time power of the energy storage system of the node i, The upper limit value of the single node installation capacity of the energy storage system of the node i. i BESS The single node installation capacity of the energy storage system of the node i. ES (t) The single node installation capacity of the energy storage system of the node i.

[0165] Based on the above description, the present application also has the following advantages over the prior art:

[0166] 1、The present application selects a typical planning scene extraction method based on information entropy theory in multiple scenes, completes the selection of the number of typical scenes of power system planning in multiple scenes, provides a data basis for subsequent clustering calculation, and compared with a planning algorithm directly performing clustering calculation or a data analysis method based on a single section, can effectively deal with defects such as large data volume, long data processing time, large calculation error and failure to closely combine with the electrical characteristics of the distribution network.

[0167] 2、The COBL-PSO algorithm used in the present application improves the performance of the intelligent optimization algorithm compared with the typical distribution network optimization algorithm, can achieve faster convergence speed and more effective optimization scheme, and has a positive effect on the overall performance improvement of the distribution network.

[0168] 3、The present application finally proposes a distribution network planning method after the access of photovoltaic power generation and energy storage system based on the extraction of typical scenes in multiple scenes, which combines an artificial intelligence algorithm, is simple and easy to operate, fast and efficient, and can guarantee the diversity of the reduced scenes, can seek the optimal planning scheme under the probability of each scene, and makes the planning scheme have the best fitness to each scene.

[0169] In addition, the present application also provides a joint planning system for distributed photovoltaic and energy storage systems in a distribution network, which comprises:

[0170] A data acquisition module is used to acquire node distributed photovoltaic and energy storage system data.

[0171] A number determination module is used to determine the number of typical scenes of the distribution network according to the node distributed photovoltaic and energy storage system data by using information entropy.

[0172] A scene extraction module is used to extract typical scenes by using a K-medoids clustering algorithm according to the number of typical scenes of the distribution network.

[0173] A model generation module is used to generate a joint planning model for distributed photovoltaic and energy storage systems under the condition of meeting a target function based on the extracted typical scenes.

[0174] The scheme generation module is used to solve the joint planning model of distributed photovoltaic and energy storage systems using the COBL-PSO algorithm to obtain a joint planning scheme for the distributed photovoltaic and energy storage systems.

[0175] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Reference can be made to the common and similar parts between the various embodiments. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method description.

[0176] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only intended to help understand the method and core concept of the present invention. At the same time, those skilled in the art will find that the specific implementation methods and application scopes may vary based on the concept of the present invention. In summary, the contents of this specification should not be construed as limiting the present invention.

Claims

1. A joint planning method for distributed photovoltaic and energy storage systems in a distribution network, characterized in that: include: Obtain node distributed photovoltaic and energy storage system data; Using information entropy to determine the number of typical distribution network scenarios based on the distributed photovoltaic and energy storage system data at the node; The K-medoids clustering algorithm is used to extract typical scenarios based on the typical scenarios of the distribution network to obtain typical scenarios; Based on the extracted typical scenarios, a joint planning model of distributed photovoltaic and energy storage systems is generated under the condition that the objective function is satisfied. In particular, the joint planning model of distributed photovoltaic and energy storage systems is generated based on the typical scenarios with the goals of satisfying grid operation reliability, environmental protection, economy, and coordination. The comprehensive objective function expression is: minC=min(C1,C2,C3,C4); Among them, C is the objective function value, C1 is the economic target, C2 is the reliability target, C3 is the environmental target, and C4 is the coordination target. is the load rate of line i' during the planning period under scenario m'; is the load rate of distribution transformer j' during the planning period under scenario m'; LR l is the average load rate of the line; LR transf is the average load rate of the transformer; Pm' is the probability of occurrence of scenario m'; The COBL-PSO algorithm is used to solve the joint planning model of the distributed photovoltaic and energy storage system to obtain a joint planning scheme of the distributed photovoltaic and energy storage system.

2. The method for joint planning of distributed photovoltaic and energy storage systems in a distribution network according to claim 1, characterized in that: The method of using information entropy to determine the number of typical distribution network scenarios based on the node distributed photovoltaic and energy storage system data specifically includes: Obtaining a power distribution density function of a scene to be planned, and using the scene power distribution density function as an initial root node; Determining the information entropy of the initial root node; Determining the split position of the initial root node with the goal of maximizing the reduction of the information entropy; Splitting the root node based on the split position to obtain a split node; Determining the information entropy of the segmentation node; Determine whether the information entropy of the split node meets the stopping condition; When the information entropy of the split node meets the stopping condition, the split node is used as a leaf node; When the information entropy of the split node does not meet the stopping condition, the current split node is used as the new initial root node and the process returns to "determining the information entropy of the initial root node" until the information entropy of all split nodes meets the stopping condition; The number of leaf nodes is determined, and the number of leaf nodes is used as the number of typical scenarios of the distribution network.

3. The method for joint planning of distributed photovoltaic and energy storage systems in a distribution network according to claim 2, characterized in that: The stopping condition is: the information entropy of the split node is less than a critical value.

4. The method for joint planning of distributed photovoltaic and energy storage systems in a distribution network according to claim 2, characterized in that: The K-medoids clustering algorithm is used to extract typical scenarios based on the typical scenarios of the distribution network to obtain typical scenarios, specifically including: Take a leaf node as a typical scene set to generate dataset D; The K-medoids clustering algorithm is used to cluster the dataset D to obtain typical scenarios.

5. The method for joint planning of distributed photovoltaic and energy storage systems in a distribution network according to claim 1, characterized in that: The process of determining reliability target C2 is as follows: Determine node voltage stability and load stability; Generate a reliability target C2 based on the node voltage smoothness and the load smoothness: C2=f(SSD U ,SSD l ); Among them, f(*) objective function, SSD U is the node voltage stability, SSD l For load stability.

6. The method for joint planning of distributed photovoltaic and energy storage systems in a distribution network according to claim 1, characterized in that: The COBL-PSO algorithm is used to solve the joint planning model of the distributed photovoltaic and energy storage system based on the constraints to obtain a joint planning scheme of the distributed photovoltaic and energy storage system; the constraints include: power flow constraints, line transmission power constraints, node voltage constraints, single-node distributed photovoltaic grid-connected capacity constraints, distribution network distributed photovoltaic grid-connected total amount constraints and energy storage system related constraints.

7. The method for joint planning of distributed photovoltaic and energy storage systems in a distribution network according to claim 6, characterized in that: The COBL-PSO algorithm is used to solve the joint planning model of the distributed photovoltaic and energy storage system based on the constraints to obtain a joint planning scheme for the distributed photovoltaic and energy storage system, which specifically includes: Use initialization parameters to initialize the particle swarm optimization algorithm; Determining calculation parameters of the joint planning model of the distributed photovoltaic and energy storage system, and setting a period corresponding to the joint planning model of the distributed photovoltaic and energy storage system; Configure the initial particle swarm parameters in the particle swarm optimization algorithm, update the particle speed and particle position, and record the updated particle speed and position; Using the objective function in the joint planning model of the distributed photovoltaic and energy storage system as the fitness function of the particle swarm optimization algorithm to determine the fitness of particle generation; Select the particle with the best fitness to update and obtain the updated particle, and record the fitness of the updated particle and the individual optimal position of the updated particle; Determining a global optimal position based on the fitness of the updated particle and the individual optimal position of the updated particle, wherein the global optimal position is the output power of the distributed photovoltaic and energy storage system in a typical scenario that minimizes the objective function value; Determine whether the output power of the distributed photovoltaic and energy storage system meets the calculation limit or whether the current number of iterations exceeds the maximum number of iterations; If the output power of the distributed photovoltaic and energy storage system meets the calculation limit or the current number of iterations does not exceed the maximum number of iterations, the output power of the distributed photovoltaic and energy storage system is output as the corresponding output value of the distributed photovoltaic and energy storage system; If the output power of the distributed photovoltaic and energy storage system does not meet the calculation limit or the current number of iterations exceeds the maximum number of iterations, the process returns to executing "update particle speed and particle position".

8. The method for joint planning of distributed photovoltaic and energy storage systems in a distribution network according to claim 7, characterized in that: The process of updating the particle velocity and particle position is: If the b-th velocity component of particle a at the time of iteration t is greater than the preset velocity value, then the b-th velocity component of particle a at the time of iteration t is set to be equal to the preset velocity value; If the b-th velocity component of particle a at the time of iteration t is less than the opposite value of the preset velocity value, then the b-th velocity component of particle a at the time of iteration t is equal to the opposite value of the preset velocity value, and the iteration is repeated; The particle swarm center of gravity is determined, and the updated direction solution of each particle is determined based on the particle swarm center of gravity.

9. A joint planning system for distributed photovoltaic and energy storage systems in a distribution network, characterized in that: include: Data acquisition module, used to obtain node distributed photovoltaic and energy storage system data; A number determination module is used to determine the number of typical distribution network scenarios based on the node distributed photovoltaic and energy storage system data using information entropy; A scenario extraction module is used to extract typical scenarios based on the typical scenarios of the distribution network using a K-medoids clustering algorithm; The model generation module is used to generate a joint planning model for distributed photovoltaic and energy storage systems based on the extracted typical scenarios and under the condition that the objective function is satisfied. The joint planning model for distributed photovoltaic and energy storage systems is generated based on the typical scenarios and with the goals of satisfying grid operation reliability, environmental protection, economy, and coordination. The comprehensive objective function expression is: minC=min(C1,C2,C3,C4); Among them, C is the objective function value, C1 is the economic target, C2 is the reliability target, C3 is the environmental target, and C4 is the coordination target. is the load rate of line i' during the planning period under scenario m'; is the load rate of distribution transformer j' during the planning period under scenario m'; LR l is the average load rate of the line; LR transf is the average load rate of the transformer; Pm' is the probability of occurrence of scenario m'; The scheme generation module is used to solve the joint planning model of the distributed photovoltaic and energy storage system by using the COBL-PSO algorithm to obtain a joint planning scheme of the distributed photovoltaic and energy storage system.

Images