Method and device for planning energy storage capacity of hybrid power distribution network

Through clustering analysis and particle swarm algorithm, the optimal capacity and location of energy storage devices in the distribution network are determined, which solves the problem of unreasonable configuration in the existing technology and improves the configuration accuracy of distributed power generation systems and energy storage systems. , optimized the operation of the distribution network.

CN119010133APending Publication Date: 2024-11-22国网河北省电力有限公司营销服务中心 +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411038977.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-31
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

When configuring distributed power generation systems and energy storage systems in the prior art, it fails to effectively represent the load conditions of real distribution network users, resulting in unreasonable configuration and affecting the operation of the power grid.

Method used

By obtaining the load data of each user in the target area for cluster analysis, the user type and typical load are determined, the site selection and capacity planning model of the energy storage device is constructed based on the daily comprehensive cost of the distribution network, and the energy storage device is determined using the particle swarm algorithm. Optimal capacity and position of the device.

Benefits of technology

The rational configuration of distributed power generation systems and energy storage systems is realized, the site selection and capacity accuracy of photovoltaic power generation and energy storage devices are improved, and the overvoltage and overload problems of the distribution network are reduced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119010133A_ABST
    Figure CN119010133A_ABST
Patent Text Reader

Abstract

The invention provides a planning method and device for the energy storage capacity of a hybrid power distribution network, and belongs to the field of power distribution network optimization scheduling. The method comprises the steps of obtaining load data of each user in a target area, performing clustering analysis on the load data, and determining a user type and a typical load corresponding to each user type; based on the daily comprehensive cost of the power distribution network, constructing a locating and sizing planning model of the energy storage device; wherein the daily comprehensive cost comprises daily investment cost and daily maintenance cost of photovoltaic and energy storage, and daily power loss cost, power purchase cost and light abandoning cost of the power distribution network; and on the basis of a particle swarm algorithm, solving the addressing and sizing planning model by using typical load data corresponding to different user types, and determining the optimal capacity and the optimal position of the energy storage device. According to the invention, reasonable configuration of the distributed power generation system and the energy storage system can be realized.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network optimization and dispatching, and in particular to a method and device for planning energy storage capacity of a hybrid distribution network. Background Art

[0002] With the rapid development of renewable energy and distributed power sources, more and more distributed renewable energy sources are connected to distribution networks, and distribution networks are gradually transitioning to active networks with interactive supply and demand.

[0003] After large-scale distributed photovoltaic power generation systems are connected to the original distribution network, problems such as overvoltage and heavy overload in the distribution network are becoming increasingly prominent. Energy storage systems play an important role in distributed photovoltaic grid connection, and energy storage systems can improve the consumption of renewable energy. Therefore, how to reasonably configure distributed power sources and energy storage systems has become one of the current research hotspots.

[0004] At present, the distributed power generation and energy storage system are mainly configured based on simulation environments such as IEEE33 nodes. However, this method cannot effectively represent the load conditions of real distribution network users, resulting in the inability to achieve reasonable configuration of distributed power generation systems and energy storage systems, which has a great impact on the power grid. Summary of the invention

[0005] The embodiment of the present invention provides a method and device for planning the energy storage capacity of a hybrid distribution network to solve the problem of unreasonable configuration of the current distributed power generation system and energy storage system.

[0006] In a first aspect, an embodiment of the present invention provides a method for planning energy storage capacity of a hybrid distribution network, comprising:

[0007] Obtain the load data of each user in the target area, and perform cluster analysis on the load data to determine the user type and the typical load corresponding to each user type;

[0008] Based on the daily comprehensive cost of the distribution network, a site selection and capacity planning model for energy storage devices is constructed; the energy storage device in the distribution network is an electrochemical energy storage system, and the daily comprehensive cost includes the daily investment cost and daily maintenance cost of photovoltaic and energy storage, the daily power loss cost of the distribution network, the electricity purchase cost, and the cost of abandoned light; the daily investment cost is converted from the total installed capacity of photovoltaic and energy storage, and the daily maintenance cost is determined based on the installed capacity, installation location, and equipment quality of photovoltaic and energy storage;

[0009] Based on the particle swarm algorithm and using typical load data corresponding to different user types, the site selection and capacity planning model is solved to determine the optimal capacity and location of the energy storage device.

[0010] In a possible implementation, before cluster analysis is performed on the load data, the following steps are further included:

[0011] Based on the load data of each user, determine the load mean vector and load variance vector of each user;

[0012] Determine the load variation coefficient of each user based on the load mean vector and the load variance vector of each user;

[0013] Based on the load variation coefficient of each user and the preset fluctuation threshold, the load data of abnormal users are eliminated to obtain the filtered load data.

[0014] In a possible implementation, the load variation coefficient of each user is for:

[0015]

[0016] in, is the load mean vector of 24 points of the i-th user, is the load variance vector of the 24 points of the i-th user, is the load variation coefficient of the i-th user at the j-th time, is the load mean vector of the i-th user at the j-th time, is the load variance vector of the ith user at the jth time.

[0017] In a possible implementation, based on the load variation coefficient of each user and a preset fluctuation threshold, the load data of abnormal users are eliminated to obtain filtered load data, including:

[0018] Input the load variation coefficient of each user into the pre-built screening model to obtain the fluctuation value of each user;

[0019] When the fluctuation value of the target user is greater than the preset fluctuation threshold, the load data of the target user is removed from the load data of each user to obtain filtered load data.

[0020] In one possible implementation, the screening model R exec for:

[0021]

[0022]

[0023] Wherein, ε is the preset fluctuation threshold, and M=24.

[0024] In a possible implementation, the location and capacity planning model F of the energy storage device is:

[0025] F=C invest +C o +C pv-q +Closs +C buy ;

[0026] Among them, C invest is the average daily investment cost of photovoltaic and energy storage, C o is the average daily operation and maintenance cost of photovoltaic and energy storage, C pv-q is the cost of abandoned light, C loss is the daily power loss cost of the transmission line in the distribution network, C buy It is the average daily electricity purchase cost of the distribution network.

[0027] In one possible implementation, based on the particle swarm algorithm and using typical load data corresponding to different user types, the site selection and capacity planning model is solved to determine the optimal capacity and optimal location of the energy storage device, including:

[0028] Establish a distribution network node model and allocate the typical load corresponding to each user type to each node in the distribution network node model;

[0029] Based on the particle swarm algorithm, the established distribution network node model and the load of each node are used to solve the site selection and capacity planning model to determine the optimal capacity and optimal location of the energy storage device.

[0030] In a possible implementation, the distribution network node model includes a bus parameter matrix, a photovoltaic power generation system parameter matrix, and a branch parameter matrix;

[0031] The bus parameter matrix includes node number, node type, node load injection power, node conductance and node reference voltage; the photovoltaic power generation system parameter matrix includes photovoltaic power generation system number, photovoltaic power generation system power, photovoltaic power generation system voltage and photovoltaic power generation system working status; the branch parameter matrix includes branch starting node number, branch end node number, each branch resistance reactance, branch phase angle and branch working status.

[0032] In a possible implementation, the typical load corresponding to each user type is allocated to each node in the distribution network node model, including:

[0033] The total injected power of each type of load is obtained based on the injection power of each node type in the distribution network;

[0034] Calculate the ratio of the injection power allocated to the target node to the total injection power of the load type of the node, and determine it as a first ratio; wherein the target node is any node in the distribution network node model;

[0035] The 24-hour loads of all the target nodes are determined based on the product of the first ratios of all the target nodes and the 24-hour loads.

[0036] In a second aspect, an embodiment of the present invention provides a device for planning energy storage capacity of a hybrid distribution network, including:

[0037] Determine the load model to obtain the load data of each user in the target area, and perform cluster analysis on the load data to determine the user type and the typical load corresponding to each user type;

[0038] A model module is constructed to construct a site selection and capacity planning model for energy storage devices based on the daily comprehensive cost of the distribution network. The energy storage device in the distribution network is an electrochemical energy storage system. The daily comprehensive cost includes the daily investment cost and daily maintenance cost of photovoltaic and energy storage, the daily power loss cost and electricity purchase cost of the distribution network, and the cost of abandoned light. The daily investment cost is calculated by the total installed capacity of photovoltaic and energy storage, and the daily maintenance cost is determined based on the installed capacity, installation location and equipment quality of photovoltaic and energy storage.

[0039] The capacity location determination module solves the site selection and capacity planning model based on the particle swarm algorithm and uses the typical load data corresponding to different user types to determine the optimal capacity and optimal location of the energy storage device.

[0040] An embodiment of the present invention provides a method and device for planning the energy storage capacity of a hybrid distribution network. First, the load data of each user in the target area is obtained, and the load data is clustered and analyzed to determine the user type and the typical load corresponding to each user type. Then, based on the daily comprehensive cost of the distribution network, a site selection and sizing planning model for the energy storage device is constructed. Finally, based on the particle swarm algorithm and using the typical load data corresponding to different user types, the site selection and sizing planning model is solved to determine the optimal capacity and optimal location of the energy storage device. The present invention takes into account the impact of user type on the load, and therefore first clusters the load data of each user obtained to obtain the typical load corresponding to each user type. Taking into account the high installation and operation and maintenance costs of electrochemical energy storage systems, a site selection and sizing planning model for energy storage devices is set up, so that the capacity and installation location of the energy storage system can be determined more accurately, thereby achieving a reasonable configuration of the distributed power generation system and the energy storage system. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. 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 creative work.

[0042] Figure 1 This is an application scenario diagram of the method for planning the energy storage capacity of a hybrid distribution network provided by an embodiment of the present invention;

[0043] Figure 2 It is a flow chart of the implementation of the method for planning the energy storage capacity of the hybrid distribution network provided by the embodiment of the present invention;

[0044] Figure 3 is a flow chart of an implementation method for eliminating abnormal user load data provided by an embodiment of the present invention;

[0045] Figure 4 It is a schematic diagram of the structure of load distribution among distribution network nodes provided by an embodiment of the present invention;

[0046] Figure 5 It is a structural schematic diagram of a planning device for energy storage capacity of a hybrid distribution network provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0047] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present invention. However, it should be clear to those skilled in the art that the present invention may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present invention.

[0048] In order to make the purpose, technical solutions and advantages of the present invention more clear, specific embodiments will be described below in conjunction with the accompanying drawings.

[0049] The energy storage configuration in the distribution network is mostly based on the simulation environment such as IEEE33 nodes, and is solved through solving algorithms and objective functions. However, in the process of solving and modeling, the impact of user type on load is not considered, so that the selected typical user load cannot effectively represent the user situation of the real distribution network. Figure 1 As shown in the figure, the distribution network can be divided into industrial electricity, agricultural electricity, commercial electricity and residential electricity based on the load type. Since the demand for electricity consumption of each type of load is different, it is necessary to consider the impact of user type on the load in order to improve the accuracy of site selection and capacity determination of photovoltaic power generation and energy storage devices.

[0050] Figure 2 The implementation flow chart of the method for planning the energy storage capacity of the hybrid distribution network provided by the embodiment of the present invention is described in detail as follows:

[0051] S210: Obtain load data of each user in the target area, and perform cluster analysis on the load data to determine user types and typical loads corresponding to each user type.

[0052] The load in the distribution network can be divided into direct supply, wholesale, urban and rural markets according to the sales location and channel. According to the size of customer electricity consumption, it can be divided into large customers, medium and small customers. According to the electricity price category, it can be divided into industrial electricity users, agricultural electricity users, commercial electricity users and residential electricity users. According to the reliability requirements, it can be divided into Class I, Class II and Class III load customers.

[0053] In some embodiments, Figure 3 As shown, the number of users in the distribution network can be set to N, and the hourly load data of users in the distribution network within an analysis period can be collected.

[0054] In this embodiment, the load of a single user at 24 points can be expressed as:

[0055]

[0056] Among them, i is the user number and d is the data collection day.

[0057] After obtaining the load data of all users in the distribution network, it is necessary to process the acquired load data. The specific processing process is as follows:

[0058] First, based on the load data of each user, a load mean vector and a load variance vector of each user may be determined.

[0059] During the analysis period, the load mean vector and load variance vector of each user at 24 points are calculated in units of days. For example, when the analysis period is 7 days, the load data of each user is 7 24-hour load data. The load data at the same time every day are averaged to obtain the user's 24-hour load mean vector. Similarly, the load data at the same time every day are varianceed to obtain the user's 24-hour load variance vector.

[0060] Exemplarily, the load mean vector and load variance vector of each user can be expressed as:

[0061]

[0062]

[0063] in, represents the load mean vector of user i during the analysis period, represents the load variance vector of user i during the analysis period.

[0064] Then, based on the load mean vector and load variance vector of each user, the load variation coefficient of each user is determined.

[0065] Load variation coefficient of each user for:

[0066]

[0067] in, is the load mean vector of 24 points of the i-th user, is the load variance vector of the 24 points of the i-th user, is the load variation coefficient of the i-th user at the j-th time, is the load mean vector of the i-th user at the j-th time, is the load variance vector of the ith user at the jth time.

[0068] Finally, based on the load variation coefficient of each user and the preset fluctuation threshold, the load data of abnormal users are eliminated to obtain the filtered load data.

[0069] In some embodiments, the load variation coefficient of each user can be input into a pre-built screening model to obtain the fluctuation value of each user. Then, when the fluctuation value of the target user is greater than a preset fluctuation threshold, the load data of the target user is removed from the load data of each user to obtain the filtered load data. The target user is any one of the users.

[0070] Specifically, the screening model R exec for:

[0071]

[0072]

[0073] Wherein, ε is the preset fluctuation threshold, and M=24.

[0074] Used to collect statistics on user load fluctuations. The value range of ε can be 0.7-0.9.

[0075] When the fluctuation value of the target user is greater than the preset fluctuation threshold, it indicates that the power load is abnormal and should not be used to represent the typical daily user load. The user load data is deleted from the sample set to obtain the filtered load data. For example, the preset fluctuation threshold may be 50%.

[0076] After obtaining the filtered load data, cluster analysis can be performed on it. The specific steps are as follows:

[0077] First, the filtered load data is adapted, that is, for the filtered load data, the 24-hour load mean of each user is used as a piece of data, that is, a sample point, for cluster analysis. Cluster analysis is performed based on power fluctuation characteristics, and each sample is a 24-dimensional vector.

[0078] Then, the kmeans++ algorithm is used for cluster analysis. The initial cluster center selection is optimized using the kmeans++ algorithm; the number of clusters k ranges from 2 to 5, and the squared sum of Euclidean distance is used as the loss function to measure the error.

[0079] The loss function is:

[0080] Among them, x i is the sample, μ ci is the cluster center, and M is the total number of samples.

[0081] According to the loss function J(c,μ), k with the smallest loss is selected as the number of clusters for cluster analysis. The kmeans++ algorithm is used for cluster analysis, and the cluster center is taken as the typical load for solving the energy storage scale model.

[0082] Collect the hourly load data of distribution network users, use kmeans++ clustering algorithm to cluster the user load data, and obtain the user type. According to the user type obtained by clustering, add category labels to the user load, and accumulate and sum the user load according to the type to obtain the total hourly load data of this type of user.

[0083] S220. Based on the daily comprehensive cost of the distribution network, a site selection and capacity planning model for energy storage devices is constructed.

[0084] The energy storage device in the distribution network is an electrochemical energy storage system. Electrochemical energy storage is the storage of energy by converting chemical energy and electrical energy into each other through chemical reactions. Depending on the material, it can be mainly divided into lead-acid batteries, sodium-sulfur batteries, flow batteries and lithium-ion batteries. On the one hand, battery energy storage has high energy density and energy conversion efficiency, and a fast response speed, which can effectively meet the peak and frequency regulation needs of the power system; on the other hand, its power and energy can be flexibly configured according to different application requirements, and is almost unaffected by external climate and geographical factors.

[0085] When using an electrochemical energy storage system, considering the high installation and operation costs of the electrochemical energy storage system, it is necessary to consider the cost factor in the energy storage planning. Therefore, the safety benefits of energy storage to the distribution network and its energy storage cost are comprehensively considered, and the energy storage capacity and layout planning are designed as target optimization problems, and a site selection and capacity planning model for the energy storage device is constructed.

[0086] The daily comprehensive cost includes the daily investment and maintenance costs of photovoltaic and energy storage, the daily power loss costs of the distribution network, the electricity purchase costs, and the cost of abandoned light. The daily investment cost can be converted from the total installed capacity of photovoltaic and energy storage, and the daily maintenance cost can be determined based on a variety of factors such as the installed capacity, installation location, and equipment quality of photovoltaic and energy storage.

[0087] Taking the minimization of daily comprehensive cost as the objective function, a site selection and capacity planning model for energy storage devices is established.

[0088] The location and capacity planning model F of the energy storage device is:

[0089] F=C invest +C o +C pv-q +C loss +C buy ;

[0090] Among them, C invest is the average daily investment cost of photovoltaic and energy storage, C o is the average daily operation and maintenance cost of photovoltaic and energy storage, C pv-q is the cost of abandoned light, C loss is the daily power loss cost of the transmission line in the distribution network, C buy It is the average daily electricity purchase cost of the distribution network.

[0091] As for investment costs, the investment costs mainly include photovoltaic power source investment costs C DG and energy storage investment cost C BS .

[0092] C invest =C DG +C BS , C DG =C pv S pv F pv , C BS =C bs S bs F bs ;

[0093] Among them, c pv is the unit capacity cost of photovoltaic power generation, S pv is the installed capacity of photovoltaic power generation, F pv is the average daily usage conversion rate of photovoltaic power source, which can be expressed as:

[0094]

[0095] D pv is the photovoltaic power depreciation rate (valued at 0.12), T pv is the service life of photovoltaic power source, S bs is the installed capacity of the battery, C bs The cost of the battery per unit capacity.

[0096] F bs is the average daily usage rate of the battery, which can be expressed as:

[0097]

[0098] D bs is the battery depreciation rate (0.12), T bs is the service life of the battery (20 years).

[0099] For maintenance cost C o , maintenance cost C o Including photovoltaic power supply maintenance cost M pv and the maintenance cost of energy storage M bs .

[0100] C o =M pv +M bs , M pv =μ pv P pv S pv , M bs =μ bs P bs ;

[0101] μ pv is the maintenance cost per unit output power of photovoltaic power, P pv is the total output power of a single photovoltaic panel, S pv is the photovoltaic capacity, μ bs is the maintenance cost per unit output power of the battery, P bs It is the charging and discharging power of the battery.

[0102] For the abandoned light cost C pv-q The amount of abandoned light is the main indicator to measure the new energy absorption capacity of the new energy power generation system. Considering the maximum absorption of photovoltaic power generation, the abandoned light cost is included in the objective function, and the abandoned light cost can be expressed as:

[0103]

[0104] C q is the light abandonment penalty coefficient, P pv-q (t) is the abandoned optical power in period t.

[0105] For the grid power loss cost C loss Taking into account the different electricity prices in different time periods, the electricity prices in different time periods are set according to the time.

[0106] The grid power loss cost is:

[0107] e t is the hourly electricity price corresponding to 24 hours, N is the number of branches, PF i (t) is the active power at the head end of branch i at time t, PT i(t) is the active power at the end of branch i at time t. Through a power flow calculation, the power at the beginning and end of each branch can be obtained, and then the power loss of the branch can be obtained.

[0108] For electricity purchase cost C buy , which can be expressed as

[0109] E total (t) is the power consumption of the power grid at time t obtained based on various loads, E dg (t) is the effective power generation of the photovoltaic power source at time t obtained based on the photovoltaic power source.

[0110] The constraints of the site selection and sizing planning model for energy storage devices include: power constraints, voltage constraints, installation capacity constraints and state of charge constraints.

[0111] Power constraints include:

[0112]

[0113]

[0114] P DGi and Q DGi They represent the injected active power and reactive power of the distributed generation at node i into the grid respectively; P loadi and Q loadi Respectively represent the active and reactive power of the load at node i, V i Represents the voltage increase at node i.

[0115] The voltage constraint is:

[0116] After the distribution network is connected to the distributed power source, it is necessary to ensure that the node voltage does not exceed a certain range, so there are the following constraints:

[0117]

[0118] The installation capacity constraint is:

[0119] S pvimin ≤S pvi ≤S pvimax ;

[0120] S pvimin and S pvimax Represent the lower and upper capacity limits respectively.

[0121] The state of charge constraints are:

[0122] State of charge (SIC) refers to the available state of charge remaining in the battery. This concept expresses how "full" the battery is as a percentage.

[0123] SOCmin ≤SOC≤SOC max ;

[0124] SOC min and SOC max Represent the lower and upper limits of the state of charge respectively.

[0125] S230. Based on the particle swarm algorithm and using typical load data corresponding to different user types, the site selection and capacity planning model is solved to determine the optimal capacity and optimal location of the energy storage device.

[0126] In some embodiments, a distribution network node model may be first established, and the typical load corresponding to each user type may be allocated to each node in the distribution network node model. Then, based on the particle swarm algorithm and using the typical load data corresponding to different user types, the site selection and capacity planning model may be solved to determine the optimal capacity and optimal location of the energy storage device.

[0127] In this embodiment, each particle in the PSO algorithm is described by a position and a velocity vector, where the position vector represents a possible solution to the problem, the velocity vector represents a possible solution to the problem, and the velocity vector represents the direction and magnitude of the position change. Assuming that the total number of examples is M, the position and velocity of the nth particle in dimension d are expressed as follows:

[0128]

[0129] Each particle adjusts their speed and position by tracking their previous individual best position and the group best position, which are expressed as:

[0130]

[0131] Where: P′ n is the individual optimal position of the nth particle, P′ g is the group best position obtained from all particles in the previous iteration.

[0132] The speed and position update formula of the PSO algorithm is expressed as:

[0133]

[0134] w is the inertia weight factor. Increasing the inertia weight factor is not easy to fall into the local minimum, which is convenient for global search. Reducing the inertia weight factor is conducive to local search and the convergence of the algorithm. c′1 and c′2 are learning factors, which reflect the self-learning ability and social learning ability of particles respectively. r′1 and r′2 are random numbers uniformly distributed in [0, 1]. In the standard PSO algorithm, w′=1, c′1=c′2=2.

[0135] In this embodiment, the steps of solving the site selection and capacity planning model by the PSO algorithm are as follows:

[0136] S2310. Input distribution network node model.

[0137] The distribution network node model includes bus parameter matrix, photovoltaic power generation system parameter matrix and branch parameter matrix. The bus parameter matrix includes node number, node type, node load injection power, node conductance and node reference voltage. The photovoltaic power generation system parameter matrix includes photovoltaic power generation system number, photovoltaic power generation system power, photovoltaic power generation system voltage and photovoltaic power generation system working status. The branch parameter matrix includes branch starting node number, branch end node number, each branch resistance reactance, branch phase angle and branch working status.

[0138] According to the ratio of the node load injection power to the sum of all node injection powers in the bus parameter matrix, the typical load data obtained in S210 is distributed to each node in the distribution network.

[0139] S2320, set the number of particle swarm iterations and the number of particles, and initialize the particle swarm.

[0140] S2330, the node model containing DG is equivalent to a PV node, and the forward and backward method is used to calculate the power flow. The fitness function is calculated, and the penalty function is used to deal with the solution that does not meet the constraint conditions.

[0141] like Figure 4 As shown, the PV node is a node with known active power and voltage, and the node model containing DG is a distribution network node model connected to a distributed photovoltaic power source. The power and voltage of the photovoltaic power source, the node line load power, and the node line voltage are all known data.

[0142] Since there are different load types in the nodes in the distribution network, the load of the node needs to be determined according to the load type.

[0143] First, the total injected power of each type of load can be obtained based on the injected power of each node type in the distribution network. Then, the ratio of the injected power allocated to the target node to the total injected power of the load type of the node is calculated and determined as a first ratio. The target node is any node in the distribution network node model. Then, the 24-hour load of all target nodes is determined based on the product of the first ratio of all target nodes and the 24-hour load.

[0144] Use MATPOWER power flow calculation tool to perform power flow calculation, and the configured distribution network data can be used for power flow calculation.

[0145] The fitness function value is the function value obtained by substituting the distribution network operation result data obtained by the power flow calculation into the function given in step 2.

[0146] S2340, perform individual optimization, the position vector corresponding to the minimum value of the particle's objective function this time is the optimal solution, and compared with the previous objective function value, the solution corresponding to the smaller one is the current individual optimal solution.

[0147] S2350, speed update, position update.

[0148] S2360, perform global optimization. Compare the optimal solution of this cycle with the current optimal solution, and the smaller one is the latest global optimal solution.

[0149] S2370: If the search termination condition is met or the maximum number of searches is reached, the optimal result is output; otherwise, S2330 is looped and the number of search iterations is increased by 1.

[0150] The planning method provided by the present invention first obtains the load data of each user in the target area, and performs cluster analysis on the load data to determine the user type and the typical load corresponding to each user type. Then, based on the daily comprehensive cost of the distribution network, a site selection and sizing planning model for the energy storage device is constructed. Finally, based on the particle swarm algorithm and using the typical load data corresponding to different user types, the site selection and sizing planning model is solved to determine the optimal capacity and optimal location of the energy storage device. The present invention takes into account the impact of user type on the load, and therefore first clusters the load data of each user obtained to obtain the typical load corresponding to each user type. Taking into account the high installation and operation and maintenance costs of electrochemical energy storage systems, a site selection and sizing planning model for energy storage devices is set up, so that the capacity and installation location of the energy storage system can be determined more accurately, thereby achieving a reasonable configuration of the distributed power generation system and the energy storage system.

[0151] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0152] The following is an embodiment of the device of the present invention. For details not described in detail, reference may be made to the corresponding method embodiments described above.

[0153] Figure 5 The schematic diagram of the structure of the planning device of the energy storage capacity of the hybrid distribution network provided by the embodiment of the present invention is shown. For the convenience of explanation, only the part related to the embodiment of the present invention is shown, which is described in detail as follows:

[0154] like Figure 5 As shown, the planning device 500 for the energy storage capacity of the hybrid distribution network includes:

[0155] Determine a load model 510, which is used to obtain load data of each user in the target area, and perform cluster analysis on the load data to determine user types and typical loads corresponding to each user type;

[0156] A model building module 520 is used to build a site selection and capacity planning model for energy storage devices based on the daily comprehensive cost of the distribution network; wherein the energy storage device in the distribution network is an electrochemical energy storage system, and the daily comprehensive cost includes the daily investment cost and daily maintenance cost of photovoltaic and energy storage, the daily power loss cost of the distribution network, the electricity purchase cost and the abandoned light cost; the daily investment cost is calculated by converting the total installed capacity of photovoltaic and energy storage, and the daily maintenance cost is determined based on the installed capacity, installation location and equipment quality of photovoltaic and energy storage;

[0157] The capacity location determination module 530 is used to solve the site selection and capacity planning model based on the particle swarm algorithm and using typical load data corresponding to different user types to determine the optimal capacity and optimal location of the energy storage device.

[0158] In a possible implementation, a load model 510 is determined to determine a load mean vector and a load variance vector for each user based on the load data of each user;

[0159] Determine the load variation coefficient of each user based on the load mean vector and the load variance vector of each user;

[0160] Based on the load variation coefficient of each user and the preset fluctuation threshold, the load data of abnormal users are eliminated to obtain the filtered load data.

[0161] In a possible implementation, the load variation coefficient of each user is for:

[0162]

[0163] in, is the load mean vector of 24 points of the i-th user, is the load variance vector of the 24 points of the i-th user, is the load variation coefficient of the i-th user at the j-th time, is the load mean vector of the i-th user at the j-th time, is the load variance vector of the ith user at the jth time.

[0164] In one possible implementation, a load model 510 is determined to input the load variation coefficient of each user into a pre-built screening model to obtain a fluctuation value of each user;

[0165] When the fluctuation value of the target user is greater than the preset fluctuation threshold, the load data of the target user is removed from the load data of each user to obtain filtered load data.

[0166] In one possible implementation, the screening model R exec for:

[0167]

[0168]

[0169] Wherein, ε is the preset fluctuation threshold, and M=24.

[0170] In a possible implementation, the location and capacity planning model F of the energy storage device is:

[0171] F=C invest +C o +C pv-q +C loss +C buy ;

[0172] Among them, C invest is the average daily investment cost of photovoltaic and energy storage, C o is the average daily operation and maintenance cost of photovoltaic and energy storage, C pv-q is the cost of abandoned light, C loss is the daily power loss cost of the transmission line in the distribution network, C buy It is the average daily electricity purchase cost of the distribution network.

[0173] In a possible implementation, the capacity location determination module 530 is used to establish a distribution network node model and allocate the typical load corresponding to each user type to each node in the distribution network node model;

[0174] Based on the particle swarm algorithm and using typical load data corresponding to different user types, the site selection and capacity planning model is solved to determine the optimal capacity and location of the energy storage device.

[0175] In a possible implementation, the distribution network node model includes a bus parameter matrix, a photovoltaic power generation system parameter matrix, and a branch parameter matrix;

[0176] The bus parameter matrix includes node number, node type, node load injection power, node conductance and node reference voltage; the photovoltaic power generation system parameter matrix includes photovoltaic power generation system number, photovoltaic power generation system power, photovoltaic power generation system voltage and photovoltaic power generation system working status; the branch parameter matrix includes branch starting node number, branch end node number, each branch resistance reactance, branch phase angle and branch working status.

[0177] In a possible implementation, the capacity location determination module 530 is configured to obtain a total injected power of each type of load based on the injected power of each node type in the distribution network;

[0178] Calculate the ratio of the injection power allocated to the target node to the total injection power of the load type of the node, and determine it as a first ratio; wherein the target node is any node in the distribution network node model;

[0179] The 24-hour loads of all the target nodes are determined based on the product of the first ratios of all the target nodes and the 24-hour loads.

[0180] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0181] Those of ordinary skill in the art will appreciate that the templates, units, and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0182] If the module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, it can implement the steps of the above-mentioned various hybrid distribution network energy storage capacity planning method embodiments. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device that can carry the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory, random access memory, electrical carrier signal, telecommunication signal and software distribution medium.

[0183] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for planning energy storage capacity of a hybrid distribution network, characterized in that: include: Obtaining load data of each user in the target area, and performing cluster analysis on the load data to determine user types and typical loads corresponding to each user type; Based on the daily comprehensive cost of the distribution network, a site selection and capacity planning model for energy storage devices is constructed; wherein the energy storage device in the distribution network is an electrochemical energy storage system, and the daily comprehensive cost includes the daily investment cost and daily maintenance cost of photovoltaic and energy storage, the daily power loss cost of the distribution network, the electricity purchase cost and the cost of abandoned light; the daily investment cost is converted from the total installed capacity of photovoltaic and energy storage, and the daily maintenance cost is determined based on the installed capacity, installation location and equipment quality of photovoltaic and energy storage; Based on the particle swarm algorithm and using typical load data corresponding to different user types, the site selection and capacity planning model is solved to determine the optimal capacity and location of the energy storage device.

2. The method for planning energy storage capacity of a hybrid distribution network according to claim 1, characterized in that: Before performing cluster analysis on the load data, the method further includes: Determine a load mean vector and a load variance vector of each user based on the load data of each user; Determining a load variation coefficient of each user based on the load mean vector and the load variance vector of each user; Based on the load variation coefficient of each user and a preset fluctuation threshold, the load data of abnormal users are eliminated to obtain filtered load data.

3. The method for planning the energy storage capacity of a hybrid distribution network according to claim 2, characterized in that: The load variation coefficient of each user for: in, is the load mean vector of 24 points of the i-th user, is the load variance vector of the 24 points of the i-th user, is the load variation coefficient of the i-th user at the j-th time, is the load mean vector of the i-th user at the j-th time, is the load variance vector of the ith user at the jth time.

4. The method for planning energy storage capacity of a hybrid distribution network according to claim 2, characterized in that: The load data of abnormal users are eliminated based on the load variation coefficient of each user and the preset fluctuation threshold to obtain filtered load data, including: Inputting the load variation coefficient of each user into a pre-built screening model to obtain a fluctuation value of each user; When the fluctuation value of the target user is greater than a preset fluctuation threshold, the load data of the target user is removed from the load data of each user to obtain filtered load data.

5. The method for planning the energy storage capacity of a hybrid distribution network according to claim 4, characterized in that: The screening model R exec for: Wherein, ε is the preset fluctuation threshold, and M=24.

6. The method for planning energy storage capacity of a hybrid distribution network according to claim 1, characterized in that: The location and capacity planning model F of the energy storage device is: F=C invest +C o +C pv-q +C loss +C buy ; Among them, C invest is the average daily investment cost of photovoltaic and energy storage, C o is the average daily operation and maintenance cost of photovoltaic and energy storage, C pv-q is the cost of abandoned light, C loss is the daily power loss cost of the transmission line in the distribution network, C buy It is the average daily electricity purchase cost of the distribution network.

7. The method for planning energy storage capacity of a hybrid distribution network according to claim 1, characterized in that: The particle swarm algorithm is used to solve the site selection and capacity planning model by using typical load data corresponding to different user types to determine the optimal capacity and optimal location of the energy storage device, including: Establishing a distribution network node model, and allocating typical loads corresponding to each user type to each node in the distribution network node model; Based on the particle swarm algorithm, the established distribution network node model and the load of each node, the site selection and capacity planning model is solved to determine the optimal capacity and optimal location of the energy storage device.

8. The method for planning energy storage capacity of a hybrid distribution network according to claim 7, characterized in that: The distribution network node model includes a bus parameter matrix, a photovoltaic power generation system parameter matrix and a branch parameter matrix; The bus parameter matrix includes node number, node type, node load injection power, node conductance and node reference voltage; the photovoltaic power generation system parameter matrix includes photovoltaic power generation system number, photovoltaic power generation system power, photovoltaic power generation system voltage, photovoltaic power generation system working status; the branch parameter matrix includes branch starting node number, branch end node number, each branch resistance reactance, branch phase angle and branch working status.

9. The method for planning energy storage capacity of a hybrid distribution network according to claim 7, characterized in that: The method of allocating the typical load corresponding to each user type to each node in the distribution network node model includes: The total injected power of each type of load is obtained based on the injection power of each node type in the distribution network; Calculate the ratio of the injection power allocated to the target node to the total injection power of the load type of the node, and determine it as a first ratio; wherein the target node is any node in the distribution network node model; The 24-hour loads of all the target nodes are determined based on the product of the first ratios of all the target nodes and the 24-hour loads.

10. A device for planning energy storage capacity of a hybrid distribution network, characterized in that: include: Determine a load model for obtaining load data of each user in the target area, and perform cluster analysis on the load data to determine user types and typical loads corresponding to each user type; A model module is constructed to construct a site selection and capacity planning model for energy storage devices based on the daily comprehensive cost of the distribution network; wherein the energy storage device in the distribution network is an electrochemical energy storage system, and the daily comprehensive cost includes the daily investment cost and daily maintenance cost of photovoltaic and energy storage, the daily power loss cost of the distribution network, the electricity purchase cost and the abandoned light cost; the daily investment cost is converted from the total installed capacity of photovoltaic and energy storage, and the daily maintenance cost is determined based on the installed capacity, installation location and equipment quality of photovoltaic and energy storage; The capacity location determination module is used to solve the site selection and capacity planning model based on the particle swarm algorithm and use the typical load data corresponding to different user types to determine the optimal capacity and optimal location of the energy storage device.