Distributed energy storage aggregation planning method based on cluster division
By adopting a two-layer optimization model based on cluster division in distributed energy storage planning, combining genetic algorithms and economic and technical indicators to optimize energy storage configuration, the problem of ineffective response to dynamic load changes in the existing technology has been solved, and the problem of improving the Overall performance and adaptability of energy storage systems.
Patent Information
- Application Number
- CN202411102525.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-12
- Publication Date
- 2024-11-26
AI Technical Summary
The existing technology lacks research on the effect of suppressing source load fluctuations in the cluster in distributed energy storage planning, and the existing methods usually aim to achieve controllable power generation of renewable energy or minimize net load fluctuations in distribution networks, and fail to Effectively respond to dynamic load changes, resulting in incomplete cluster performance evaluation.
Using distributed energy storage aggregation planning method based on cluster division, a two-layer optimization model is built, and the energy storage configuration is optimized through genetic algorithms, combining economic and technical indicators, including modularity, active balance degree and reactive balance degree, to optimize energy storage Overall performance of the system.
The economic and technical goals of the energy storage system are achieved, the overall benefits of the system are improved, the adaptability and optimization effect of the model are enhanced, and the rationality of cluster division and the performance of the energy storage system are ensured.
Smart Images

Figure CN119026947A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of energy management and energy storage planning, and mainly to a distributed energy storage aggregation planning method based on cluster division. Background Art
[0002] Traditional power systems face complex load fluctuations and distributed energy integration issues, which has led to the concept of cluster division. Existing methods for studying distributed energy storage planning mostly focus on the scale of the entire distribution system, selecting the location and sizing the energy storage. The problem has a high complexity of solution, and there is a lack of research on the role of energy storage within the cluster in smoothing source-load fluctuations. In addition, existing research usually aims to achieve controllable renewable energy generation or minimize net load fluctuations in the distribution network. There is little research on the two stages of comprehensive energy storage planning and operation to enhance the system's economic, safety, and stability benefits.
[0003] For example, CN114723052A "A multi-scenario clustering structure partitioning algorithm for energy storage planning" discloses "a multi-scenario clustering structure partitioning algorithm for energy storage planning, which first calculates the modularity index based on electrical distance, then calculates the comprehensive purity index considering multiple scenario conditions, and establishes a clustering structure optimization model with the goal of maximizing the comprehensive evaluation index of cluster partitioning, and finally solves the model based on an improved genetic algorithm for the cluster partitioning problem." However, although the method introduces modularity index and comprehensive purity index, these indexes mainly focus on the strength of electrical connection and characteristic differentiation, and do not deeply consider other technical indicators such as load fluctuations, which may lead to incomplete evaluation of cluster performance, making it difficult to cope with dynamic load changes in actual operation. Summary of the invention
[0004] In order to solve the above problems existing in the prior art, the present application provides a distributed energy storage aggregation planning method based on cluster division.
[0005] The technical solution of this application is as follows:
[0006] A distributed energy storage aggregation planning method based on cluster division, the method comprising:
[0007] Obtaining data related to distributed energy storage aggregation planning, including distribution network parameters, energy storage parameters, and distribution network cluster division results;
[0008] Construct a distributed energy storage aggregation planning indicator system, including economic indicators and technical indicators, wherein the economic indicator is calculated based on the distribution network parameters and energy storage parameters, and the economic indicator is the energy storage configuration cost; the technical indicator is calculated based on the distribution network parameters and the distribution network cluster division results, and the technical indicator is the load fluctuation indicator;
[0009] According to the distributed energy storage aggregation planning related data and distributed energy storage aggregation planning indicator system, a two-layer optimization model for energy storage aggregation planning is constructed to determine the corresponding constraints;
[0010] The energy storage aggregation planning double-layer optimization model is solved by a genetic algorithm and an optimization solver to obtain an energy storage aggregation planning scheme.
[0011] Preferably, the distribution network parameters include the topology of the distribution system, node load power, power of the new energy generator set, power of the reactive compensation device, line conductance, line admittance, maximum line transmission power, lower limit of active output of the generator set, upper limit of active output of the generator set, lower limit of reactive output of the generator set, upper limit of reactive output of the generator set, lower limit of node voltage amplitude, upper limit of node voltage amplitude and power purchase cost of the distribution network;
[0012] The energy storage parameters include the lower limit of energy storage power, the upper limit of energy storage power, the lower limit of energy storage capacity, the upper limit of energy storage capacity, the energy storage charging efficiency, the energy storage discharging efficiency, the lower limit of energy storage charge state, the upper limit of energy storage charge state and the energy storage investment cost.
[0013] Preferably, the distribution network cluster division result is generated based on the modularity index, the active power balance index and the reactive power balance index, and the weighted comprehensive performance index, specifically:
[0014] The modularity index ρ is used to measure the structural connection strength of the cluster and is expressed as:
[0015]
[0016] In the formula, e ij represents the connection coefficient between node i and node j; e ij Indicates spatial distance or electrical distance; if e ij =1, indicating that the two nodes are directly connected in the topology graph; if e ij =0, indicating that the two nodes are not directly connected in the topology graph; represents the sum of the connection coefficients of all nodes in the network; k i =∑ j e ij Indicates the sum of the connection coefficients of node i; when node i and node j are in the same cluster, it is recorded as δ(i,j)=1, otherwise it is recorded as δ(i,j)=0;
[0017] The active power balance index It is used to reflect the stability and balance of active power in each node or area of the power grid, and is expressed by the formula:
[0018]
[0019] Where Pk represents the active power balance degree of cluster k; c represents the number of clusters; P clu,k (t) represents the net power of cluster k in time period t; T represents the time length of a typical time-varying scenario; P load,i represents the load demand of node i; P pv,i and P wind,i represents the output power of photovoltaic and wind power in node i;
[0020] The reactive power balance index It is used to reflect the capabilities of each node or area in the power grid in terms of reactive local balance and voltage control, and is expressed in the formula:
[0021]
[0022] Q need,k =Q load,k +Q minV,k ;
[0023]
[0024] In the formula, Q k Indicates the reactive power balance degree of cluster k; Q sup,k represents the maximum reactive power supply capacity within cluster k, including the reactive power provided by the reactive compensation device on the node, the segmented voltage controller, and the reactive power of some inverters; Q cb,i and Q svc,i are the reactive powers of the group switching capacitor and static VAR compensator installed at node i; Q need,k represents the reactive power demand within cluster k; Q minV,k represents the minimum reactive power required by the overvoltage node in cluster k; ΔV i represents the voltage change of node i; S QV,ii is the reactive voltage sensitivity of node i with respect to itself; Q load,k represents the reactive load demand of all nodes in cluster k;
[0025] The comprehensive performance index γ is a combination of the modularity index, the active power balance degree and the reactive power balance degree, and is used as the optimization target, and is expressed in the formula:
[0026]
[0027] In the formula, λ1 represents the weight coefficient of the modularity index; λ2 represents the weight coefficient of the active power balance degree; λ3 represents the weight coefficient of the reactive power balance degree; ρ represents the modularity value after cluster division of the system; Indicates the active power balance degree of the system after cluster division; Indicates the reactive power balance degree of the system after cluster division.
[0028] Preferably, the economic index is calculated based on the distribution network parameters and the energy storage parameters, and the economic index is the energy storage configuration cost f1, which is expressed by the formula:
[0029]
[0030] In the formula, f inv represents the investment cost of energy storage; N represents the evaluation period; f ope represents the operating cost of the distribution network; P t r P represents the cost of purchasing electricity from the upper power grid; t up Indicates the amount of electricity purchased from the upper power grid;
[0031] The technical index is calculated based on the distribution network parameters and the distribution network cluster division results. The technical index is the load fluctuation index f tech , expressed as:
[0032]
[0033] In the formula, is the net load of cluster k in period t; C k represents the set of nodes of cluster k; and The active power of the distributed generator set and load at node i in time period t respectively; and are the charging and discharging power of the energy storage system at node i in time period t respectively.
[0034] Preferably, the two-layer optimization model for energy storage aggregation planning includes an upper layer model and a lower layer model, wherein:
[0035] The objective function f of the upper model upper , expressed as:
[0036] minf upper =f1+αf tech ;
[0037] α=Ae -B×gen ;
[0038] In the formula, α represents the variable weight coefficient; A represents the proportional coefficient; B represents the attenuation coefficient; gen represents the number of iterations;
[0039] The corresponding constraints include upper and lower limits of energy storage power and capacity, which can be expressed as follows:
[0040]
[0041] Where m represents the energy storage system index; and Indicates the rated power and rated capacity of energy storage m; and Indicates the minimum and maximum rated power of energy storage m; and Indicates the minimum and maximum rated capacity of energy storage m.
[0042] Preferably, the objective function f of the lower model is lower , expressed as:
[0043] minf lower =f tech ;
[0044] The corresponding constraints include node power balance constraints, AC power flow constraints, line transmission capacity constraints, generator output constraints, node voltage amplitude constraints, and energy storage system constraints, which can be expressed as follows:
[0045]
[0046] Where j∈Θ i represents the set of nodes connected to node i; P ij represents the transmission active power of line ij; P i G represents the active power of the generator set at node i; P i DG represents the active power of the renewable energy unit at node i; P i dis and P i ch represents the energy storage charging and discharging power at node i; P i L represents the load power at node i;
[0047]
[0048] Where U i represents the voltage amplitude of node i; G ij and B ij represents the conductance and susceptance of line ij; Θ ij represents the phase angle difference between node i and node j;
[0049]
[0050] In the formula, and represents the reactive power of the generator set, renewable energy unit and load at node i;
[0051]
[0052] In the formula, and represents the reactive power of the generator set, renewable energy unit and load at node i; represents the upper limit of the transmission power of line ij;
[0053]
[0054] Where P i G,min and P i G,max It is expressed as the minimum active power and maximum active power of the generator set at node i;
[0055]
[0056] In the formula, and It is expressed as the minimum reactive power and maximum reactive power of the generator set at node i;
[0057]
[0058] In the formula, and It is expressed as the minimum and maximum voltage of node i;
[0059]
[0060] In the formula, and A 0-1 variable indicating whether the energy storage m is in the charging or discharging state during time period t;
[0061]
[0062] In the formula, and It indicates the charging and discharging power of energy storage m in time period t;
[0063]
[0064] In the formula, and It indicates the charging and discharging efficiency of energy storage m in time period t; represents the storage capacity of energy storage m in time period t;
[0065]
[0066] In the formula, and Indicates the minimum and maximum state of charge of the energy storage m.
[0067] Preferably, the two-layer optimization model of energy storage aggregation planning is solved by genetic algorithm and optimization solver, specifically, the upper model uses genetic algorithm to optimize the cluster location, rated power and rated capacity of energy storage, and the lower model uses optimization solver to calculate the optimal power flow.
[0068] Preferably, a genetic algorithm is used in the upper model to optimize the cluster location, rated power and rated capacity of the energy storage, specifically:
[0069] Set model parameters, including the number of populations, the number of energy storages, the capacity of distributed energy storage, the maximum and minimum power, the maximum number of iterations and crossovers, and the maximum and minimum mutation probability;
[0070] Initialize the population according to the set model parameters, and each individual in the population represents an energy storage configuration scheme;
[0071] Substitute each individual into the lower model, use the optimization solver to calculate the optimal power flow to obtain the fitness function value; determine the crossover and mutation probabilities based on the fitness function value, and perform selection, crossover and mutation operations to generate a new generation of population;
[0072] Check whether the new population meets the termination condition, that is, the maximum number of iterations is reached or the fitness value changes to the fitness threshold; if the termination condition is not met, repeat the selection, crossover and mutation operations; if the termination condition is met, output the optimal solution and decode it.
[0073] Preferably, the optimal power flow is calculated using an optimization solver in the lower model as follows:
[0074] Traverse each individual in the population and calculate the charging and discharging power and capacity obtained by each individual during the optimization process;
[0075] Calculate the economic and technical indicators of each individual according to the charging and discharging strategy;
[0076] Use economic indicators, technical indicators and variable weight coefficient α to calculate the fitness function value of the individual;
[0077] Calculate the fitness function values of all individuals and output the optimal result.
[0078] A distributed energy storage aggregation planning system based on cluster division, the system includes a data acquisition module, an energy storage aggregation planning two-layer model module, an output module and a communication module, wherein:
[0079] The data acquisition module is used to acquire distributed energy storage aggregation planning related data, including distribution network parameters, energy storage parameters and distribution network cluster division results;
[0080] The energy storage aggregation planning double-layer model module is internally provided with an energy storage aggregation planning double-layer optimization model, which is constructed according to the distributed energy storage aggregation planning related data and the distributed energy storage aggregation planning indicator system, and includes an upper model and a lower model, wherein:
[0081] The distributed energy storage aggregation planning indicator system includes economic indicators and technical indicators. The economic indicator is the energy storage configuration cost, which is calculated based on the distribution network parameters and energy storage parameters; the technical indicator is the load fluctuation indicator, which is calculated based on the distribution network parameters and the distribution network cluster division results;
[0082] The energy storage aggregation planning two-layer optimization model uses a genetic algorithm in the upper model to optimize the cluster location, rated power and rated capacity of energy storage, and uses an optimization solver in the lower model to calculate the optimal power flow, and obtains the energy storage aggregation planning scheme through upper and lower optimization solutions;
[0083] The output module is used to obtain the energy storage aggregation planning scheme and transmit it to the distribution network operator through the communication module.
[0084] Compared with the prior art, the present invention has the following beneficial effects:
[0085] 1) The present invention provides a distributed energy storage aggregation planning method based on cluster division, which optimizes economic indicators and technical indicators simultaneously by constructing a two-layer optimization model, achieves a balance between economic and technical goals, and improves the overall efficiency of the energy storage system;
[0086] 2) The present invention provides a distributed energy storage aggregation planning method based on cluster division, which optimizes the upper model through genetic algorithm so that the model can adapt to different energy storage configuration requirements and operating conditions, and continuously optimizes the energy storage configuration scheme through operations such as crossover and mutation, thereby improving the adaptability and optimization effect of the model;
[0087] 3) The present invention provides a distributed energy storage aggregation planning method based on cluster division, which optimizes the energy storage aggregation planning through the distribution network cluster division results, accurately reflects the actual operation of the power grid, and through the comprehensive evaluation of modularity indicators, active balance degree and reactive balance degree, makes the cluster division more reasonable and improves the overall performance of the energy storage system. BRIEF DESCRIPTION OF THE DRAWINGS
[0088] Figure 1 It is a flow chart of a distributed energy storage aggregation planning method based on cluster division according to an embodiment of the present invention;
[0089] Figure 2 It is a distributed energy storage aggregation planning system diagram based on cluster division according to an embodiment of the present invention;
[0090] Figure 3 It is a flowchart of the upper model optimization process of an embodiment of the present invention;
[0091] Figure 4 It is a flowchart of the lower-level model optimization process according to an embodiment of the present invention. DETAILED DESCRIPTION
[0092] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.
[0093] The present invention provides the following technical solution: a distributed energy storage aggregation planning method based on cluster division.
[0094] Example 1
[0095] This embodiment provides a distributed energy storage aggregation planning method based on cluster division, and the specific steps include:
[0096] S1. Obtaining data related to distributed energy storage aggregation planning, including distribution network parameters, energy storage parameters, and distribution network cluster division results;
[0097] S11, the distribution network parameters include the topology of the distribution system, node load power, power of new energy generator sets, power of reactive compensation devices, line conductance, line admittance, maximum line transmission power, lower limit of active output of generator sets, upper limit of active output of generator sets, lower limit of reactive output of generator sets, upper limit of reactive output of generator sets, lower limit of node voltage amplitude, upper limit of node voltage amplitude and power purchase cost of distribution network;
[0098] S12, the energy storage parameters include energy storage power lower limit, energy storage power upper limit, energy storage capacity lower limit, energy storage capacity upper limit, energy storage charging efficiency, energy storage discharge efficiency, energy storage charge state lower limit, energy storage charge state upper limit and energy storage investment cost;
[0099] S13. The distribution network cluster division result is generated based on the modularity index, active balance index and reactive balance index, and the weighted comprehensive performance index, specifically:
[0100] S131, calculation of modularity index
[0101] The modularity index ρ is used to measure the structural connection strength of the cluster and is expressed as:
[0102]
[0103] In the formula, e ij represents the connection coefficient between node i and node j; e ij Indicates spatial distance or electrical distance; if e ij =1, indicating that the two nodes are directly connected in the topology graph; if e ij =0, indicating that the two nodes are not directly connected in the topology graph; represents the sum of the connection coefficients of all nodes in the network; k i =∑ j e ij Indicates the sum of the connection coefficients of node i; when node i and node j are in the same cluster, it is recorded as δ(i,j)=1, otherwise it is recorded as δ(i,j)=0;
[0104] S132. Calculate active power balance index
[0105] The active power balance index It is used to reflect the stability and balance of active power in each node or area of the power grid, and is expressed by the formula:
[0106]
[0107] Where P k represents the active power balance degree of cluster k; c represents the number of clusters; P clu,k (t) represents the net power of cluster k in time period t; T represents the time length of a typical time-varying scenario; P load,i represents the load demand of node i; P pv,i and P wind,i represents the output power of photovoltaic and wind power in node i;
[0108] S133, calculation of reactive power balance index
[0109] The reactive power balance index It is used to reflect the capabilities of each node or area in the power grid in terms of reactive local balance and voltage control, and is expressed in the formula:
[0110]
[0111] Q need,k =Q load,k +Q minV,k ;
[0112]
[0113] In the formula, Q k Indicates the reactive power balance degree of cluster k; Q sup,krepresents the maximum reactive power supply capacity within cluster k, including the reactive power provided by the reactive compensation device on the node, the segmented voltage controller, and the reactive power of some inverters; Q cb,i and Q svc,i are the reactive powers of the group switching capacitor and static VAR compensator installed at node i; Q need,k represents the reactive power demand within cluster k; Q minV,k represents the minimum reactive power required by the overvoltage node in cluster k; ΔV i represents the voltage change of node i; S QV,ii is the reactive voltage sensitivity of node i with respect to itself; Q load,k represents the reactive load demand of all nodes in cluster k;
[0114] S134. Calculate comprehensive performance indicators
[0115] The comprehensive performance index is a combination of the modularity index, the active power balance index and the reactive power balance index, and is used as the optimization target;
[0116] According to the specific operation characteristics and objectives of the power grid, the value of the weight coefficient is determined by the hierarchical analysis method, the importance of the three indicators is compared to construct a judgment matrix, and the value of each weight is obtained; each indicator is normalized; using the determined weight, the normalized indicators are weighted and summed to calculate the comprehensive performance indicator value of each area or node;
[0117] According to the calculated comprehensive performance index value, the distribution network cluster is divided into different clusters or areas by setting a threshold or according to the distribution characteristics of the index value;
[0118] The comprehensive performance index γ is a combination of the modularity index, the active power balance degree and the reactive power balance degree, and is used as the optimization target, and is expressed in the formula:
[0119]
[0120] In the formula, λ1 represents the weight coefficient of the modularity index; λ2 represents the weight coefficient of the active power balance degree; λ3 represents the weight coefficient of the reactive power balance degree; ρ represents the modularity value after cluster division of the system; Indicates the active power balance degree of the system after cluster division; Indicates the reactive power balance degree of the system after cluster division;
[0121] S2. Construct a distributed energy storage aggregation planning indicator system, including economic indicators and technical indicators, wherein the economic indicator is calculated based on the distribution network parameters and the energy storage parameters, and the economic indicator is the energy storage configuration cost; the technical indicator is calculated based on the distribution network parameters and the distribution network cluster division results, and the technical indicator is the load fluctuation indicator;
[0122] S21. Economic indicators
[0123] Economic indicators are used to evaluate the economic benefits of investment and operation, including payback period, internal rate of return, net present value, operating cost and savings ratio, and cost-benefit analysis. They mainly consider energy storage configuration cost, operation and maintenance, and equipment residual value to establish an energy storage life cycle model, while considering the grid’s time-of-use electricity price mechanism to establish distribution network economic indicators.
[0124] The economic index is calculated based on the distribution network parameters and energy storage parameters. The economic index is the energy storage configuration cost f1, which is expressed as:
[0125]
[0126] In the formula, f inv represents the investment cost of energy storage; N represents the evaluation period; f ope represents the operating cost of the distribution network; P t r P represents the cost of purchasing electricity from the upper power grid; t up Indicates the amount of electricity purchased from the upper power grid;
[0127] S22. Technical indicators
[0128] Technical indicators are used to evaluate the performance characteristics and indicators of technical solutions or products in achieving specific functions or goals, including energy storage capacity, power output capacity, efficiency, response time, cycle life, safety and reliability, and environmental adaptability. They mainly consider the role of energy storage in improving the load curve of the distribution network and establish the load fluctuation index of the distribution network, which is crucial for evaluating and selecting the best technical solution.
[0129] The technical index is calculated based on the distribution network parameters and the distribution network cluster division results. The technical index is the load fluctuation index f tech , expressed as:
[0130]
[0131] In the formula, is the net load of cluster k in period t; C k represents the set of nodes of cluster k; and The active power of the distributed generator set and load at node i in time period t respectively; and are the charging and discharging power of the energy storage system at node i in time period t respectively.
[0132] S3. Construct a two-layer optimization model for energy storage aggregation planning based on the distributed energy storage aggregation planning related data and the distributed energy storage aggregation planning indicator system, and determine the corresponding constraints; the two-layer optimization model for energy storage aggregation planning includes an upper model and a lower model;
[0133] S31, the objective function f of the upper model upper It is a comprehensive planning performance index of energy storage that combines economic and technical aspects and is expressed in the formula:
[0134] minf upper =f1+αf tech ;
[0135] α=Ae -B×gen ;
[0136] In the formula, α represents the variable weight coefficient; A represents the proportional coefficient; B represents the attenuation coefficient; gen represents the number of iterations;
[0137] The corresponding constraints include upper and lower limits of energy storage power and capacity, which can be expressed as follows:
[0138]
[0139] Where m represents the energy storage system index; and Indicates the rated power and rated capacity of energy storage m; and Indicates the minimum and maximum rated power of energy storage m; and Indicates the minimum and maximum values of the rated capacity of the energy storage m;
[0140] S32, the objective function f of the lower model lower , expressed as:
[0141] minf lower =f tech ;
[0142] The corresponding constraints include node power balance constraints, AC power flow constraints, line transmission capacity constraints, generator output constraints, node voltage amplitude constraints, and energy storage system constraints, which can be expressed as follows:
[0143]
[0144] Where j∈Θi represents the set of nodes connected to node i; P ij represents the transmission active power of line ij; P i G represents the active power of the generator set at node i; P i DG represents the active power of the renewable energy unit at node i; P i dis and P i ch represents the energy storage charging and discharging power at node i; P i L represents the load power at node i;
[0145]
[0146] Where U i represents the voltage amplitude of node i; G ij and B ij represents the conductance and susceptance of line ij; Θ ij represents the phase angle difference between node i and node j;
[0147]
[0148] In the formula, and represents the reactive power of the generator set, renewable energy unit and load at node i;
[0149]
[0150] In the formula, and represents the reactive power of the generator set, renewable energy unit and load at node i; represents the upper limit of the transmission power of line ij;
[0151]
[0152] Where P i G,min and P i G,max It is expressed as the minimum active power and maximum active power of the generator set at node i;
[0153]
[0154] In the formula, and It is expressed as the minimum reactive power and maximum reactive power of the generator set at node i;
[0155]
[0156] In the formula, and It is expressed as the minimum and maximum voltage of node i;
[0157]
[0158] In the formula, and A 0-1 variable indicating whether the energy storage m is in the charging or discharging state during time period t;
[0159]
[0160] In the formula, and It indicates the charging and discharging power of energy storage m in time period t;
[0161]
[0162] In the formula, and It indicates the charging and discharging efficiency of energy storage m in time period t; represents the storage capacity of energy storage m in time period t;
[0163]
[0164] In the formula, and Indicates the minimum and maximum state of charge of the energy storage m;
[0165] S4, solving the energy storage aggregation planning double-layer optimization model by using a genetic algorithm and an optimization solver to obtain an energy storage aggregation planning scheme;
[0166] S41. Genetic Algorithm
[0167] Genetic algorithm is an optimization algorithm based on biological evolution theory. It searches for the optimal solution or approximate optimal solution of complex problems by simulating the "selection", "crossover" and "mutation" processes in nature. The algorithm starts with randomly generated initial individuals, evaluates and selects individuals with high fitness as parents, and generates new individuals through crossover and mutation operations, and continuously evolves and optimizes until the preset termination condition is reached.
[0168] S42, Optimization Solver
[0169] Optimization solvers are a type of computing tool used to solve mathematical optimization problems. They use mathematical models and algorithms to find the optimal solution that minimizes or maximizes the objective function by adjusting the values of decision variables under constraints.
[0170] S43, please refer to Figure 3In the upper model, a genetic algorithm is used to optimize the cluster location, rated power and rated capacity of energy storage, specifically:
[0171] S431, setting model parameters, including the number of populations, the number of energy storages, the capacity of distributed energy storage, the maximum and minimum values of power, the maximum number of iterations and crossover, and the maximum and minimum values of mutation probability;
[0172] S432, initializing the population according to the set model parameters, each individual in the population represents an energy storage configuration scheme, and traversing each individual in the population to calculate the corresponding fitness function;
[0173] S433, substituting each individual into the lower model, using the comprehensive economic and technical energy storage planning performance indicators, using the optimization solver to calculate the economic indicators and technical indicators of each individual for optimization, judging the crossover and mutation probabilities according to the fitness function value, and performing selection, crossover and mutation operations to generate a new generation of population;
[0174] S434, check whether the new population meets the termination condition, that is, reaches the maximum number of iterations or the fitness value changes to reach the fitness threshold; if the termination condition is not met, repeat the selection, crossover and mutation operations; if the termination condition is met, output the optimal solution and perform decoding;
[0175] S44, please refer to Figure 4 , the optimal power flow is calculated using the optimization solver in the lower model, specifically:
[0176] S441. According to the input population, each individual is traversed in order, and the operation strategies of different individuals are optimized to calculate the economic indicators and technical indicators of different individuals, as well as the fitness function of the corresponding individuals;
[0177] S442, performing optimization calculation of the optimization solver according to the cluster position, power and capacity of each individual energy storage, optimizing with cluster load volatility as the objective function, obtaining the charging and discharging power and capacity of each energy storage every hour, and generating a charging and discharging strategy;
[0178] S443, determining the economic index of configuring energy storage according to the charging and discharging strategy, and determining the value of the individual fitness function according to the economic index, the technical index, and the value of the α variable weight coefficient;
[0179] S444. Check and determine whether all individuals in the population have been traversed. If not, calculate the next individual; if satisfied, output the fitness function corresponding to all individuals, which is the desired energy storage aggregation planning scheme.
[0180] Example 2
[0181] A distributed energy storage aggregation planning system based on cluster division, the system includes a data acquisition module, an energy storage aggregation planning two-layer model module, an output module and a communication module, wherein:
[0182] The data acquisition module is used to acquire distributed energy storage aggregation planning related data, including distribution network parameters, energy storage parameters and distribution network cluster division results;
[0183] The energy storage aggregation planning double-layer model module is internally provided with an energy storage aggregation planning double-layer optimization model, which is constructed according to the distributed energy storage aggregation planning related data and the distributed energy storage aggregation planning indicator system, and includes an upper model and a lower model, wherein:
[0184] The distributed energy storage aggregation planning indicator system includes economic indicators and technical indicators. The economic indicator is the energy storage configuration cost, which is calculated based on the distribution network parameters and energy storage parameters; the technical indicator is the load fluctuation indicator, which is calculated based on the distribution network parameters and the distribution network cluster division results;
[0185] The energy storage aggregation planning two-layer optimization model uses a genetic algorithm in the upper model to optimize the cluster location, rated power and rated capacity of energy storage, and uses an optimization solver in the lower model to calculate the optimal power flow, and obtains the energy storage aggregation planning scheme through upper and lower optimization solutions;
[0186] The output module is used to obtain the energy storage aggregation planning scheme and transmit it to the distribution network operator through the communication module.
[0187] It is worth noting that the system, electronic device and computer-readable storage medium described in the present invention are based on the same principles as the method described in Example 1, and will not be described in detail here.
[0188] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.
Claims
1. A distributed energy storage aggregation planning method based on cluster division, characterized in that: The method comprises: Obtaining data related to distributed energy storage aggregation planning, including distribution network parameters, energy storage parameters, and distribution network cluster division results; Construct a distributed energy storage aggregation planning indicator system, including economic indicators and technical indicators, wherein the economic indicator is calculated based on the distribution network parameters and energy storage parameters, and the economic indicator is the energy storage configuration cost; the technical indicator is calculated based on the distribution network parameters and the distribution network cluster division results, and the technical indicator is the load fluctuation indicator; According to the distributed energy storage aggregation planning related data and distributed energy storage aggregation planning indicator system, a two-layer optimization model for energy storage aggregation planning is constructed to determine the corresponding constraints; The energy storage aggregation planning double-layer optimization model is solved by a genetic algorithm and an optimization solver to obtain an energy storage aggregation planning scheme.
2. A distributed energy storage aggregation planning method based on cluster division according to claim 1, characterized in that: The distribution network parameters include the topology of the distribution system, node load power, power of new energy generator sets, power of reactive compensation devices, line conductance, line admittance, maximum line transmission power, lower limit of active output of generator sets, upper limit of active output of generator sets, lower limit of reactive output of generator sets, upper limit of reactive output of generator sets, lower limit of node voltage amplitude, upper limit of node voltage amplitude and power purchase cost of distribution network; The energy storage parameters include the lower limit of energy storage power, the upper limit of energy storage power, the lower limit of energy storage capacity, the upper limit of energy storage capacity, the energy storage charging efficiency, the energy storage discharging efficiency, the lower limit of energy storage charge state, the upper limit of energy storage charge state and the energy storage investment cost.
3. A distributed energy storage aggregation planning method based on cluster division according to claim 2, characterized in that: The distribution network cluster division results are generated based on the modularity index, active power balance index and reactive power balance index, and the weighted comprehensive performance index, which are as follows: The modularity index ρ is used to measure the structural connection strength of the cluster and is expressed as: In the formula, e ij represents the connection coefficient between node i and node j; e ij Indicates spatial distance or electrical distance; if e ij =1, indicating that the two nodes are directly connected in the topology graph; if e ij =0, indicating that the two nodes are not directly connected in the topology graph; represents the sum of the connection coefficients of all nodes in the network; k i =∑ j e ij Indicates the sum of the connection coefficients of node i; when node i and node j are in the same cluster, it is recorded as δ(i,j)=1, otherwise it is recorded as δ(i,j)=0; The active power balance index It is used to reflect the stability and balance of active power in each node or area of the power grid, and is expressed by the formula: Where P k represents the active power balance degree of cluster k; c represents the number of clusters; P clu,k (t) represents the net power of cluster k in time period t; T represents the time length of a typical time-varying scenario; P load,i represents the load demand of node i; P pv,i and P wind,i represents the output power of photovoltaic and wind power in node i; The reactive power balance index It is used to reflect the capabilities of each node or area in the power grid in terms of reactive local balance and voltage control, and is expressed in the formula: Q need,k =Q load,k +Q minV,k ; In the formula, Q k Indicates the reactive power balance degree of cluster k; Q sup,k represents the maximum reactive power supply capacity within cluster k, including the reactive power provided by the reactive compensation device on the node, the segmented voltage controller, and the reactive power of some inverters; Q cb,i and Q svc,i are the reactive powers of the group switching capacitor and static VAR compensator installed at node i; Q need,k represents the reactive power demand within cluster k; Q minV,k represents the minimum reactive power required by the overvoltage node in cluster k; ΔV i represents the voltage change of node i; S QV,ii is the reactive voltage sensitivity of node i with respect to itself; Q load,k represents the reactive load demand of all nodes in cluster k; The comprehensive performance index γ is a combination of the modularity index, the active power balance degree and the reactive power balance degree, and is used as the optimization target, and is expressed in the formula: In the formula, λ1 represents the weight coefficient of the modularity index; λ2 represents the weight coefficient of the active power balance degree; λ3 represents the weight coefficient of the reactive power balance degree; ρ represents the modularity value after cluster division of the system; Indicates the active power balance degree of the system after cluster division; Indicates the reactive power balance degree of the system after cluster division.
4. A distributed energy storage aggregation planning method based on cluster division according to claim 3, characterized in that: The economic index is calculated based on the distribution network parameters and energy storage parameters. The economic index is the energy storage configuration cost f1, which is expressed by the formula: In the formula, f inv represents the investment cost of energy storage; N represents the evaluation period; f ope represents the operating cost of the distribution network; P t r P represents the cost of purchasing electricity from the upper power grid; t up Indicates the amount of electricity purchased from the upper power grid; The technical index is calculated based on the distribution network parameters and the distribution network cluster division results. The technical index is the load fluctuation index f tech , expressed as: In the formula, is the net load of cluster k in period t; C k represents the set of nodes of cluster k; and The active power of the distributed generator set and load at node i in time period t respectively; and are the charging and discharging power of the energy storage system at node i in time period t respectively.
5. A distributed energy storage aggregation planning method based on cluster division according to claim 4, characterized in that: The two-layer optimization model for energy storage aggregation planning includes an upper-layer model and a lower-layer model, where: The objective function f of the upper model upper It is a comprehensive planning performance index of energy storage that combines economic and technical aspects and is expressed in the formula: minf upper =f1+αf tech ; α=Ae -B×gen ; In the formula, α represents the variable weight coefficient; A represents the proportional coefficient; B represents the attenuation coefficient; gen represents the number of iterations; The corresponding constraints include upper and lower limits of energy storage power and capacity, which can be expressed as follows: Where m represents the energy storage system index; and Indicates the rated power and rated capacity of energy storage m; and Indicates the minimum and maximum rated power of energy storage m; and Indicates the minimum and maximum rated capacity of energy storage m.
6. A distributed energy storage aggregation planning method based on cluster division according to claim 5, characterized in that: The objective function f of the lower model lower , expressed as: min f lower =f tech ; The corresponding constraints include node power balance constraints, AC power flow constraints, line transmission capacity constraints, generator output constraints, node voltage amplitude constraints, and energy storage system constraints, which can be expressed as follows: Where j∈Θ i represents the set of nodes connected to node i; P ij represents the transmission active power of line ij; P i G represents the active power of the generator set at node i; P i DG represents the active power of the renewable energy unit at node i; P i dis and P i ch represents the energy storage charging and discharging power at node i; P i L represents the load power at node i; Where U i represents the voltage amplitude of node i; G ij and B ij represents the conductance and susceptance of line ij; θ ij represents the phase angle difference between node i and node j; In the formula, and represents the reactive power of the generator set, renewable energy unit and load at node i; In the formula, and represents the reactive power of the generator set, renewable energy unit and load at node i; represents the upper limit of the transmission power of line ij; Where P i G,min and P i G,max It is expressed as the minimum active power and maximum active power of the generator set at node i; In the formula, and It is expressed as the minimum reactive power and maximum reactive power of the generator set at node i; In the formula, and It is expressed as the minimum and maximum voltage of node i; In the formula, and A 0-1 variable indicating whether the energy storage m is in the charging or discharging state during time period t; In the formula, and It indicates the charging and discharging power of energy storage m in time period t; In the formula, and It indicates the charging and discharging efficiency of energy storage m in time period t; represents the storage capacity of energy storage m in time period t; In the formula, and Indicates the minimum and maximum state of charge of the energy storage m.
7. A distributed energy storage aggregation planning method based on cluster division according to claim 6, characterized in that: The two-layer optimization model of energy storage aggregation planning is solved by genetic algorithm and optimization solver. Specifically, the genetic algorithm is used in the upper model to optimize the cluster location, rated power and rated capacity of energy storage, and the optimization solver is used in the lower model to calculate the optimal power flow.
8. A distributed energy storage aggregation planning method based on cluster division according to claim 7, characterized in that: In the upper model, a genetic algorithm is used to optimize the cluster location, rated power and rated capacity of energy storage, specifically: Set model parameters, including the number of populations, the number of energy storages, the capacity of distributed energy storage, the maximum and minimum power, the maximum number of iterations and crossovers, and the maximum and minimum mutation probability; Initialize the population according to the set model parameters, and each individual in the population represents an energy storage configuration scheme; Substitute each individual into the lower model, use the optimization solver to calculate the optimal power flow to obtain the fitness function value; determine the crossover and mutation probabilities based on the fitness function value, and perform selection, crossover and mutation operations to generate a new generation of population; Check whether the new population meets the termination condition, that is, the maximum number of iterations is reached or the fitness value changes to the fitness threshold; if the termination condition is not met, repeat the selection, crossover and mutation operations; if the termination condition is met, output the optimal solution and decode it.
9. A distributed energy storage aggregation planning method based on cluster division according to claim 8, characterized in that: The optimal power flow is calculated using the optimization solver in the lower model as follows: Traverse each individual in the population and use the optimization solver to optimize the operation strategy of the traversed individuals; Calculate the charging and discharging power and capacity obtained by each individual during the optimization process; Calculate the economic and technical indicators of each individual according to the charging and discharging strategy; Use economic indicators, technical indicators and variable weight coefficient α to calculate the fitness function value of the individual; Calculate the fitness function values of all individuals and output the optimal result.
10. A distributed energy storage aggregation planning system based on cluster division, characterized in that: The system includes a data acquisition module, an energy storage aggregation planning double-layer model module, an output module and a communication module, wherein: The data acquisition module is used to acquire distributed energy storage aggregation planning related data, including distribution network parameters, energy storage parameters and distribution network cluster division results; The energy storage aggregation planning double-layer model module is internally provided with an energy storage aggregation planning double-layer optimization model, which is constructed according to the distributed energy storage aggregation planning related data and the distributed energy storage aggregation planning indicator system, and includes an upper model and a lower model, wherein: The distributed energy storage aggregation planning indicator system includes economic indicators and technical indicators. The economic indicator is the energy storage configuration cost, which is calculated based on the distribution network parameters and energy storage parameters; the technical indicator is the load fluctuation indicator, which is calculated based on the distribution network parameters and the distribution network cluster division results; The energy storage aggregation planning two-layer optimization model uses a genetic algorithm in the upper model to optimize the cluster location, rated power and rated capacity of energy storage, and uses an optimization solver in the lower model to calculate the optimal power flow, and obtains the energy storage aggregation planning scheme through upper and lower optimization solutions; The output module is used to obtain the energy storage aggregation planning scheme and transmit it to the distribution network operator through the communication module.
Citation Information
Cited By
Active power distribution network balance area division method, equipment and medium
CN119853168A
Source network load storage clustering joint optimization method and device based on distributed collaboration
CN120710135A
Energy storage planning method and system in novel power system
CN120896195A
