A distributed pumped storage capacity optimization configuration method based on shared aggregation

Abstract

This invention discloses a method for optimizing the capacity allocation of distributed pumped storage based on shared aggregation. The method includes: obtaining annual equivalent operational evaluation results; constructing a dynamic evolution model of distributed pumped storage pumping, power generation, and reservoir capacity; constructing a comprehensive economic evaluation model for distributed pumped storage; constructing a distributed pumped storage capacity optimization model with the system's annualized comprehensive net cost and the daily average fluctuation variance of the adjusted surplus load power as optimization objectives; solving the distributed pumped storage capacity optimization model to obtain a Pareto non-dominated solution set; and using fuzzy set theory to comprehensively evaluate the Pareto non-dominated solution set to obtain a recommended distributed pumped storage capacity allocation scheme. This invention addresses the problems of increased net load fluctuation after a high proportion of new energy sources are connected to the distribution network, difficulty in recovering investment in distributed pumped storage, single revenue channels, and insufficient consideration of distribution network connection and operational safety constraints in existing technologies.

Description

Technical Field

[0001] This invention relates to the fields of distribution network energy storage planning, pumped storage optimization configuration, and integrated operation of the power market, and particularly to a method for optimizing the capacity configuration of distributed pumped storage based on shared aggregation. Background Technology

[0002] With the increasing integration of new energy sources such as wind and solar power into distribution networks, the randomness, volatility, and intermittency of new energy output exacerbate fluctuations in the net load of the distribution network, easily leading to problems such as insufficient local absorption of new energy, increased peak-to-valley differences, and voltage exceeding limits at nodes. Pumped storage, as an energy storage method with the capabilities of peak shaving and valley filling, frequency and voltage regulation, backup support, and promotion of new energy absorption, has become an important technical path to improve the flexibility of the power system. However, existing pumped storage capacity configuration methods are mostly geared towards large-scale centralized power plants or regional grid dispatch scenarios. For distributed pumped storage built relying on small reservoirs, lakes, or ponds on the distribution network side, there are still problems such as insufficient consideration of physical operation constraints, single revenue channels, long investment recovery periods, and insufficient verification of distribution network access security.

[0003] To address the aforementioned issues, Chinese invention patent CN114707403B proposes a multi-energy coordinated optimization scheduling method for regional distribution networks based on pumped storage regulation. This method constructs power curves for wind power, photovoltaics, small hydropower, and loads during a typical week, using total system cost, pumped storage construction cost, and remaining load fluctuations as optimization objectives to obtain pumped storage capacity configuration and multi-energy coordinated scheduling schemes. However, this patent primarily focuses on the coordinated scheduling of wind power, photovoltaics, small hydropower, and pumped storage in regional distribution networks, emphasizing improving the operational economy of multi-energy systems and the capacity for renewable energy absorption. For scenarios where distributed pumped storage on the distribution network side relies on the transformation of small reservoir resources, the synergistic relationship between capacity configuration and shared operation, market revenue calculation, and distribution network access security verification still has room for improvement. Regarding shared energy storage configuration, Chinese invention patent CN114256836B proposes a capacity optimization configuration method for shared energy storage in renewable energy power plants. This method targets the distributed shared energy storage architecture of new energy power plant clusters, designs an energy storage sharing mechanism, establishes a cooperative game model considering both grid-connected electricity revenue and shared energy storage revenue, and uses particle swarm optimization to solve for the optimal energy storage configuration. However, this patent mainly focuses on the shared energy storage configuration among new energy power plant clusters, emphasizing cooperation and sharing between different new energy power plants and the optimization of energy storage construction scale; for energy storage forms such as distributed pumped storage, which are constrained by reservoir capacity, pumped power generation operation mode switching, site resources, and distribution network access nodes, the adaptability of this method is still insufficient.

[0004] In summary, while existing technologies have proposed corresponding methods for optimized scheduling of pumped storage and shared energy storage capacity configuration, a method for optimized configuration of distributed pumped storage capacity in distribution networks based on a shared aggregation operation mechanism is still lacking. This makes it difficult to uniformly coordinate source-load temporal fluctuations, physical operational constraints of pumped storage, multi-faceted benefit assessments, capacity planning and intraday operation scheduling, and distribution network access security verification. Therefore, how to provide a method for optimized configuration of distributed pumped storage capacity in distribution networks based on shared aggregation to improve the economy, regulateability, and operational security of distributed pumped storage capacity configuration schemes has become an urgent problem to be solved. Summary of the Invention

[0005] This invention provides a distributed pumped storage capacity optimization configuration method based on shared aggregation to overcome the technical problems of increased net load fluctuations after a high proportion of new energy sources are connected to the distribution network, difficulty in recovering investment in distributed pumped storage, single revenue channels, and insufficient consideration of the safety constraints of distribution network access and operation.

[0006] To achieve the above objectives, the technical solution of the present invention is as follows: A method for optimizing the allocation of distributed pumped storage capacity based on shared aggregation, specifically including the following steps: S1: Obtain the basic data required for the optimized configuration of distributed pumped storage capacity; Furthermore, the basic data includes power distribution network load data, wind power output data, photovoltaic power output data, pumped storage power station parameters, market price parameters, investment and construction cost parameters, and operation and maintenance cost parameters; S2: Based on the distribution network load data, wind power output data, and photovoltaic output data, typical daily scenarios are extracted from the annual source-load time series data to obtain multiple typical daily scenarios; based on the occurrence probability or time weight of each typical daily scenario in the annual sample, the annual time series production simulation equivalent processing is performed to obtain the annual equivalent operation evaluation results; S3: Based on the parameters of the pumped storage power station, construct a distributed pumped storage pumping, power generation and reservoir capacity dynamic evolution model to obtain the equivalent comprehensive power generation power and equivalent comprehensive pumping power of the pumped storage. S4: Construct a comprehensive economic evaluation model for distributed pumped storage based on market price parameters, investment and construction cost parameters, and operation and maintenance cost parameters; S5: Based on steps S2 to S4, with the annualized comprehensive net cost of the system and the daily average fluctuation variance of the adjusted residual load power as the optimization objectives, construct a distributed pumped storage capacity optimization configuration model. S6: Solve the distributed pumped storage capacity optimization configuration model to obtain the Pareto non-dominated solution set; use fuzzy set theory to comprehensively evaluate the Pareto non-dominated solutions in the Pareto non-dominated solution set to obtain the recommended scheme for distributed pumped storage capacity configuration.

[0007] Furthermore, step S2 specifically includes the following steps: S21: Extract typical daily scenarios from the annual source-load time series data to obtain daily-scale source-load feature vectors; and the daily-scale source-load feature vectors include distribution network load data, wind power output data, and photovoltaic power output data corresponding to each natural day; S22: Normalize the eigenvectors of the day-scale source load; S23: Based on S22 and combined with the K-means clustering algorithm, the daily-scale source load feature vectors are clustered to obtain several clusters; multiple typical daily scenes are defined based on the clusters. And the clustering objective function of the K-means clustering algorithm is: (1) In the formula: This represents the value of the clustering objective function; Indicates the number of cluster categories; Indicates the first The nth cluster, that is, the one that is divided into the nth cluster. The set of daily-scale source load eigenvectors of the class; Indicates the normalized result of the first... Daily-scale source load feature vectors; Indicates the first Cluster centers of sample groups; S24: Based on the number of daily-scale source load samples contained in the cluster, determine the occurrence probability or time weight of the corresponding typical daily scene in the annual sample according to the expert database, and perform equivalent processing of the annual time-series production simulation based on the occurrence probability or time weight to obtain the annual equivalent operation evaluation results. And the expression for the annual equivalent operation evaluation result is: (2) In the formula: Indicates the annual equivalent operation evaluation indicators; Indicates the number of days the system operates throughout the year; Indicates the first The probability or time weight of occurrence of typical daily scenarios; Indicates the first Daily operation evaluation results under typical daily scenarios.

[0008] Furthermore, a distributed pumped storage hydroelectric power generation and reservoir capacity dynamic evolution model is constructed in S3, and its expression is: (3) (4) (5) (6) (7) (8) In the formula: express The equivalent comprehensive pumping power of the pumped storage unit at all times; and They represent The pumping capacity declared in the power market and the frequency regulation reserve market at all times; express real-time pumping flow rate; This indicates the velocity-to-power ratio under pumping conditions. Indicates the pump's operating efficiency; Indicates pipeline efficiency; This represents the density of water; This represents standard gravitational acceleration; Indicates the design head height of the pumped storage unit; express The equivalent comprehensive power generation capacity of the pumped storage unit at all times; and These represent pumped storage. The power generation capacity declared in the electricity market and the power generation capacity declared in the frequency regulation reserve market at any time; express Real-time power generation flow; This indicates the velocity-to-power ratio under power generation conditions. Indicates the generator set operating efficiency; express The water storage volume of the upstream reservoir at any given time; express The water storage volume of the upstream reservoir at any given time; Indicates the unit scheduling duration; This indicates the maximum capacity of the upstream reservoir.

[0009] Furthermore, the distributed pumped storage comprehensive economic evaluation model constructed in S4 includes a distribution and storage capacity sharing revenue model, an arbitrage revenue model for low storage and high generation in the power market, a frequency regulation and reserve revenue model for the main grid ancillary service market, a user demand control revenue model, and a life cycle cost model. The expression for the shared revenue model of the storage capacity is: (9) In the formula: This indicates the average daily capacity substitution revenue of the power plant; This indicates the approved lease rate for distributed energy storage capacity in the regional market; This indicates the total installed active power capacity of the power station. Indicates the power unit conversion constant; Indicates the capacity isolation allocation factor; Indicates the total number of days in a calendar year; The expression for the low-storage, high-inflation arbitrage profit model in the electricity market is as follows: (10) In the formula: This indicates the power plant's net revenue from low energy storage and high power generation on a single day; This indicates the total number of scheduling periods throughout the day; express Time-of-use electricity pricing for industrial and commercial users during specific time periods; This represents the minimum discount factor for the electricity purchase price for pumping water; Indicates the unit scheduling duration; The expression for the mainnet ancillary service market frequency regulation and reserve revenue model is as follows: (11) In the formula: This represents the total revenue from mainnet auxiliary services on a single day. This indicates the price for frequency modulation capacity compensation; This indicates the price for compensation for spare capacity; express The system's backup capacity must be reported in real time. and These represent the objective probability coefficients for winning and clearing bids for frequency modulation capacity and reserve capacity, respectively. express Daily forecast of mileage pricing for time-of-use frequency (TOF) radio. This represents the ratio of frequency modulation mileage to frequency modulation capacity. This represents the probability coefficient of triggering the frequency modulation command during actual operation; The expression for the user demand control revenue model is: (12) In the formula: This indicates the revenue share from managing the maximum daily demand. This indicates the profit-sharing coefficient; This indicates the maximum demand load power on the user side before the pumped storage power station was put into operation; This indicates the maximum demand load power on the user side after pumped storage participates in peak shaving and valley filling; This represents the basic electricity rate per unit of the maximum demand for industrial and commercial use, converted to a single-day basis. The expression for the life-cycle cost model is as follows: (13) (14) (15) (16) (17) In the formula: This indicates the total initial investment cost of the power plant; This indicates the unit cost of constructing the upstream reservoir capacity per cubic meter. Indicates the storage capacity; This indicates the unit price per kilowatt of installed capacity; Indicates the unit's assembly capacity; This represents the total present value of operating and maintenance costs over the entire life cycle. Indicates the overall project evaluation period for the power plant; This represents the average annual maintenance cost per unit volume of the warehouse. This represents the average annual operating and maintenance cost per unit of the unit's installed capacity. This indicates the project's discount rate; This indicates a specific sequence of years in which the project has been operating; This represents the total present value of unit replacement costs over the entire life cycle; This indicates the unit cost per kilowatt for a single replacement of a water pump turbine unit. Indicates the entire life cycle of the generator unit; Indicates the total number of replacements required within the project's lifespan; Indicates intermediate parameters; This represents the annualized equivalent comprehensive cost of a distributed pumped storage power station.

[0010] Furthermore, the distributed pumped storage capacity optimization configuration model constructed in S5 includes an objective function and constraints with the annualized comprehensive net cost of the system and the daily average fluctuation variance of the adjusted surplus load power as the optimization objectives. Furthermore, the objective function includes an economic objective function constructed with the goal of minimizing the system's annualized comprehensive net cost. The technical objective function is constructed with the goal of minimizing the daily average fluctuation variance of the adjusted surplus load power. ; The expression for the economic objective function is: (18) In the formula: This represents the annualized equivalent comprehensive cost based on distributed pumped storage power stations. Equivalent to the annual operation evaluation results The resulting annualized net cost of the system; Indicates the total number of typical scenarios; Indicates the typical day scene number; Indicates the first The annual time weighting of days for a typical scenario; They represent the first The net revenue from electricity arbitrage, power distribution and storage sharing, ancillary services, and demand management calculated by the daily model in each scenario. For the first In a single-day model under a specific scenario Daily operation evaluation results; This represents the annualized equivalent total investment cost of a distributed pumped storage power station, including replacement and operation and maintenance. The expression for the technical objective function is: (19) (20) (twenty one) In the formula: They represent the first In each scenario The original base load, wind power output, and photovoltaic power output of the power distribution network at all times; This indicates the unbalanced net load power of the distribution network system before the pumped storage system is connected; and They represent the first In each scenario The equivalent comprehensive power generation and equivalent comprehensive pumping power of the pumped storage system at all times; This represents the equivalent surplus load power after introducing distributed pumped storage for peak shaving and valley filling regulation. The constraints include pumped storage capacity constraints, upper and lower limits of operating power constraints, pumped power generation status constraints, energy evolution and state of charge constraints of energy storage systems, and power balance and interaction constraints of distribution network nodes. The pumped storage capacity constraint is: (twenty two) (twenty three) In the formula: These represent the upper limit of upstream reservoir capacity and the upper limit of installed capacity, respectively. The upper and lower limits of the operating power are constrained as follows: (twenty four) (25) (26) (27) In the formula: They represent the first In each scenario The power generation capacity of the electricity market, the reserve capacity of the frequency regulation market, and the emergency reserve capacity at any given time; These represent the pumping power in the electric energy market and the pumping reserve capacity in the frequency regulation market, respectively. Indicates installed capacity; The state constraints for pumped hydropower generation are: (28) (29) (30) (31) (32) (33) (34) The energy evolution and state of charge constraints of the energy storage system are as follows: (35) (36) (37) In the formula: Indicates the first In each scenario The pumping flow rate at any given time; Indicates the first In each scenario Power generation flow rate at any given moment; Indicates the first In each scenario The water storage volume of the upstream reservoir at any given time; Indicates the first In each scenario The water storage volume of the upstream reservoir at any given time; The power balance and interaction constraints of the distribution network nodes are as follows: (38).

[0011] Furthermore, step S6 specifically includes the following steps: S61: The installed capacity of the distributed pumped storage power station and the upstream reservoir volume are used as capacity decision variables; the NSGA-II algorithm is used to perform multi-objective capacity optimization on the constructed distributed pumped storage capacity optimization configuration model to obtain the Pareto non-dominated solution set; Furthermore, each Pareto nondominated solution in the Pareto nondominated solution set corresponds to a capacity configuration scheme for a set of capacity decision variables; S62: A comprehensive evaluation of Pareto non-dominated solutions in the Pareto non-dominated solution set is performed using fuzzy set theory, specifically including: S621: Using a linear trapezoidal fuzzy membership function, combined with the objective function of the distributed pumped storage capacity optimization configuration model, the target satisfaction of each Pareto non-dominated solution in the Pareto non-dominated solution set is obtained. And the expression for the target satisfaction level is: (39) In the formula: Indicates the first The capacity configuration scheme corresponds to the first One objective function value; They represent the first and second solutions in the Pareto non-dominated set, respectively. The minimum and maximum values ​​of the objective function; Indicates the first The capacity configuration scheme is in the first Satisfaction with each objective function; S622: Based on the target satisfaction of each Pareto non-dominated solution, obtain the comprehensive evaluation satisfaction of each capacity configuration scheme; and the expression for the comprehensive evaluation satisfaction is: (40) In the formula: Indicates the first Overall satisfaction rating of each capacity configuration scheme; This represents the total number of capacity configuration schemes in the Pareto non-dominated solution set; This represents the total number of objective functions, including economic objective functions and technical objective functions. ; S63: Select the configuration scheme with the highest overall satisfaction as the recommended configuration scheme for distributed pumped storage capacity.

[0012] This invention provides a method for optimizing the allocation of distributed pumped storage capacity based on shared aggregation, with the following beneficial effects: (1) Based on the historical time series data of load, wind power and photovoltaic, this invention extracts typical daily scenarios and performs equivalent processing of the annual time series production simulation to obtain the annual equivalent operation evaluation results. It can reduce the amount of calculation every hour throughout the year while retaining the source load fluctuation characteristics and improve the adaptability of capacity configuration results to different operation scenarios. (2) This invention constructs a dynamic evolution model of pumping operation, power generation operation and reservoir capacity of distributed pumped storage, and limits the operation boundary by means of pumping power generation state constraints, upper and lower limits of reservoir capacity constraints and first and last reservoir capacity balance constraints, so that the capacity configuration result can conform to the actual physical operation process of pumped storage power station. (3) This invention introduces a shared energy storage and aggregator collaborative operation mechanism, which incorporates the distribution and storage capacity sharing revenue model, the low storage and high generation arbitrage revenue model of the power market, the frequency regulation and reserve revenue model of the main grid auxiliary service market, and the user demand control revenue model into the comprehensive revenue evaluation, which can improve the problem of incomplete revenue evaluation of distributed pumped storage. (4) This invention incorporates the total life-cycle investment cost and multiple benefits into the economic objective, and uses the daily average fluctuation variance of the adjusted residual load power as the technical objective, which can achieve a compromise between the project's economics and the smoothness of the distribution network operation. (5) The present invention uses the NSGA-II algorithm to solve the distributed pumped storage capacity optimization configuration model to obtain the Pareto non-dominated solution set, and uses fuzzy set theory to comprehensively evaluate the Pareto non-dominated solutions in the Pareto non-dominated solution set to obtain the recommended scheme for distributed pumped storage capacity configuration. This can realize the linkage between capacity configuration and intraday operation scheduling, and improve the feasibility of the capacity configuration scheme. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 This is a flowchart of the distributed pumped storage capacity optimization configuration method based on shared aggregation of the present invention; Figure 2 This is a schematic diagram of the pumped storage capacity optimization configuration model in this embodiment; Figure 3 This is a flowchart of the double-nested optimization solution in this embodiment; Figure 4 This is a schematic diagram of the physical structure of the distributed pumped storage power station in this embodiment; Figure 5This is an equivalent flowchart of typical daily scene extraction and year-round time-series production simulation in this embodiment; Figure 6 This is a schematic diagram of the shared energy storage operation mode within the distribution network in this embodiment; Figure 7 This is a schematic diagram of the distribution network topology security verification in this embodiment; Figure 8 This is a typical daily wind-solar-load output curve for spring in this embodiment; Figure 9 This is a typical summer daytime wind-solar-load output curve in this embodiment; Figure 10 This is a typical autumn solar load output curve in this embodiment; Figure 11 This is a typical winter daytime wind-solar-load output curve in this embodiment; Figure 12 This is a Pareto solution distribution diagram of the distributed pumped storage capacity optimization configuration model for the distribution network in this embodiment; Figure 13 This is a diagram showing the proportion of diversified benefits from pumped storage in this embodiment; Figure 14 This is a comparison chart of power purchases at typical day-to-day nodes in spring, as shown in this embodiment. Figure 15 This is a comparison chart of power purchases at typical summer days at the root node in this embodiment; Figure 16 This is a comparison chart of power purchases at typical day-to-day nodes in autumn in this embodiment; Figure 17 This is a comparison chart of power purchases at typical daytime nodes in winter in this embodiment; Figure 18 This is a spatiotemporal distribution diagram of node voltage on a typical summer day in this embodiment. Detailed Implementation

[0015] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] This embodiment provides a distributed pumped storage capacity optimization configuration method based on shared aggregation. First, it collects data on distribution network load, wind power output, photovoltaic power output, pumped storage power station parameters, market price parameters, investment and construction cost parameters, and distribution network topology parameters to form the basic dataset required for capacity optimization configuration. Second, it extracts typical daily scenarios from the time-series data of load, wind power, and photovoltaic power sources, and characterizes the operational characteristics under different scenarios using an equivalent method of year-round time-series production simulation. Third, it constructs a dynamic evolution model of distributed pumped storage pumping, power generation, and reservoir capacity, combining shared energy storage and aggregation. The commercial operation model establishes revenue models for shared distribution and storage capacity, electricity market arbitrage, ancillary services, and demand control, as well as a life-cycle cost model. Then, using the system's annualized net cost and the daily average fluctuation variance of the adjusted surplus load power as optimization objectives, a distributed pumped storage capacity optimization configuration model is established, and a two-layer nested solution strategy is used to obtain the Pareto non-dominated solution set. Finally, based on fuzzy set theory, the optimal compromise capacity configuration scheme is determined, and the scheme is verified for operational safety by incorporating distribution network topology constraints, outputting a distributed pumped storage capacity configuration result that meets the requirements of economy, smoothness, and safety. Figure 1 As shown, the specific steps include: S1: Obtain the basic data required for the optimized configuration of distributed pumped storage capacity; and the basic data includes distribution network load data, wind power output data, photovoltaic power output data, pumped storage power station parameters, market price parameters, investment and construction cost parameters, operation and maintenance cost parameters, and distribution network topology parameters; Specifically, in this embodiment, the basic data required for the optimized configuration of distributed pumped storage capacity in the distribution network are collected, including distribution network load data, wind power output data, photovoltaic power output data, candidate construction data of distributed pumped storage power stations, pumped storage unit operating parameters, market price parameters, investment and construction cost parameters, operation and maintenance cost parameters, and distribution network topology parameters, to form a basic dataset for the optimized configuration of distributed pumped storage capacity in the distribution network. The parameters for distributed pumped storage power station sites and reservoirs include the maximum configurable capacity of the upper reservoir, design head height, candidate access node location, and lower reservoir construction conditions for candidate sites. Specifically, the maximum configurable capacity of the upper reservoir is used to limit the upper limit of the upper reservoir capacity decision variable during capacity optimization. The design head height is used for calculating the conversion between power and water flow under pumping and power generation conditions. The candidate access node location is used for operational safety verification in the subsequent distribution network topology model. The lower reservoir construction conditions are used to determine whether candidate sites meet the requirements for pumped storage cycle operation. Market price parameters include distribution and storage capacity leasing prices, time-of-use electricity prices, frequency regulation capacity compensation prices, standby capacity compensation prices, frequency regulation mileage prices, and maximum demand electricity charges. Investment and construction cost parameters include upstream reservoir construction costs, reversible pump-turbine investment costs, operation and maintenance costs, and equipment replacement costs. Distribution network topology parameters include node connection relationships, branch parameters, node voltage constraints, and branch power flow constraints. The site and reservoir parameters, along with the pumped storage unit operating parameters, are used to construct dynamic evolution models of pumping conditions, power generation conditions, and reservoir capacity, and to define the physical boundaries for capacity optimization. The market price parameters are used to calculate the revenue from shared distribution and storage capacity, arbitrage revenue from the electricity market, ancillary service revenue, and demand control revenue. The investment and construction cost parameters and operation and maintenance cost parameters are used to calculate the total life-cycle cost and the annualized equivalent comprehensive cost. The distribution network topology parameters are used for the safety verification of node voltage, branch power flow, and root node interactive power after distributed pumped storage is connected.

[0017] S2: Based on distribution network load data, wind power output data, and photovoltaic output data, typical daily scenarios are extracted from the annual source-load time-series data to obtain multiple typical daily scenarios; based on the occurrence probability or time weight of each typical daily scenario in the annual sample, equivalent processing of the annual time-series production simulation is performed to obtain the annual equivalent operation evaluation results, such as... Figure 5 As shown, the specific steps include: S21: Extract typical daily scenarios from the annual source-load time series data to obtain daily-scale source-load feature vectors; and the daily-scale source-load feature vectors include distribution network load data, wind power output data, and photovoltaic power output data corresponding to each natural day; Specifically, in this embodiment, the hourly load curve, wind power output curve, and photovoltaic output curve for the whole year are divided into daily scales, and each natural day is used as a source-load sample. For any natural day, the load power, wind power output, and photovoltaic output for each time period of that day are extracted in chronological order, and the three types of time-series data are combined to form the daily-scale source-load feature vector corresponding to that natural day. Among them, when 1 hour is used as the time interval, each natural day contains 24 load power data, 24 wind power output data, and 24 photovoltaic output data, that is, each daily-scale source-load feature vector is composed of the load, wind power, and photovoltaic operation characteristics of that day for 24 consecutive hours, and is used to characterize the daily-scale combination of the source-load time-series change characteristics of that day. S22: Normalize the diurnal-scale source load feature vector to eliminate the influence of different data units on the scene segmentation results; S23: Based on S22 and combined with the K-means clustering algorithm, the daily-scale source load feature vectors are clustered to obtain several clusters; multiple typical daily scenes are defined based on the clusters. And the clustering objective function of the K-means clustering algorithm is: (1) In the formula: This represents the value of the clustering objective function; Indicates the number of cluster categories; Indicates the first The nth cluster, that is, the one that is divided into the nth cluster. The set of daily-scale source load eigenvectors of the class; Indicates the normalized result of the first... Daily-scale source load feature vectors; Indicates the first Cluster centers of sample groups; S24: Based on the number of daily-scale source load samples contained in the cluster, determine the occurrence probability or time weight of the corresponding typical daily scene in the annual sample according to the expert database, and perform equivalent processing of the annual time-series production simulation based on the occurrence probability or time weight to obtain the annual equivalent operation evaluation results. And the expression for the annual equivalent operation evaluation result is: (2) In the formula: Indicates the annual equivalent operation evaluation indicators; Indicates the number of days the system operates throughout the year; Indicates the first The probability or time weight of occurrence of typical daily scenarios; Indicates the first Daily operation evaluation results under typical daily scenarios.

[0018] S3: Based on the parameters of the pumped storage power station, a distributed pumped storage pumping, power generation, and reservoir capacity dynamic evolution model is constructed to obtain the equivalent comprehensive power generation and equivalent comprehensive pumping power of the pumped storage. The pumped storage power station parameters include, but are not limited to, pumping power, pumped storage unit design head, power generation, and pumping flow rate. Specifically, in this embodiment, based on the distributed pumped storage power station site and reservoir parameters and the pumped storage unit operating parameters, the power-flow conversion relationship under pumping and power generation conditions is constructed using the design head, unit operating efficiency, and pipeline efficiency. The real-time reservoir volume is used as the state variable, and the maximum configurable reservoir capacity is used as the capacity boundary to construct a distributed pumped storage pumping, power generation, and reservoir capacity dynamic evolution model, specifically including: S31: Model the power conversion process of pumped storage units under pumping conditions, and determine the relationship between the equivalent comprehensive pumping power, the pumping power in the electricity market, the pumping power declared in the frequency regulation reserve market, and the pumping flow rate. The expression is as follows: (3) In the formula: for The equivalent comprehensive pumping power of the pumped storage unit at all times; and They are respectively The pumping capacity declared in the power market and the frequency regulation reserve market at all times; for real-time pumping flow rate; This is the velocity-power ratio under pumping conditions. (4) In the formula: Indicates the pump's operating efficiency; Indicates pipeline efficiency; This represents the density of water; This represents standard gravitational acceleration; Indicates the design head height of the pumped storage unit; S32: Model the power conversion process of pumped storage units under power generation conditions, and determine the relationship between the equivalent comprehensive power generation, the power generation in the electricity market, the power generation declared in the frequency regulation reserve market, and the power generation flow. The expression is as follows: (5) In the formula: express The equivalent comprehensive power generation capacity of the pumped storage unit at all times; and These represent pumped storage. The power generation capacity declared in the electricity market and the power generation capacity declared in the frequency regulation reserve market at any time; express Real-time power generation flow; This indicates the velocity-to-power ratio under power generation conditions. (6) In the formula: Indicates the generator set operating efficiency; S33: Based on the pumping flow rate obtained in step S31 and the power generation flow rate obtained in step S32, and using the real-time effective water storage volume of the reservoir as the state variable, a dynamic evolution model of the reservoir capacity is constructed, and the upper and lower limits of the safe operation of the reservoir capacity are defined. Its expression is: (7) (8) In the formula: express The water storage volume of the upstream reservoir at any given time; express The water storage volume of the upstream reservoir at any given time; Indicates the unit scheduling duration; This indicates the maximum capacity of the upstream reservoir.

[0019] S4: Construct a comprehensive economic evaluation model for distributed pumped storage based on market price parameters, investment and construction cost parameters, and operation and maintenance cost parameters; Specifically, the distributed pumped storage comprehensive economic evaluation model constructed in this embodiment includes a shared revenue model for storage capacity, a low-storage-high-occupancy arbitrage revenue model for the power market, a frequency regulation and reserve revenue model for the main grid ancillary services market, a user demand control revenue model, and a life-cycle cost model. The expression for the shared revenue model of the storage capacity is: (9) In the formula: This indicates the average daily capacity substitution revenue of the power plant; This indicates the approved lease rate for distributed energy storage capacity in the regional market; This indicates the total installed active power capacity of the power station. This represents the power unit conversion constant, taken as (1000kW / MW). This represents the capacity isolation allocation coefficient, which indicates the proportion of power plant capacity allocated for long-term static sharing. This represents the total number of days in a natural year, taken as 365 days. In this embodiment, the distributed pumped storage capacity sharing revenue model is used to calculate the average daily capacity substitution revenue of the power station by providing capacity substitution services to new energy power stations in the region based on the operation mode of distributed pumped storage. The expression for the low-storage, high-inflation arbitrage profit model in the electricity market is as follows: (10) In the formula: This indicates the power plant's net revenue from low energy storage and high power generation on a single day; This represents the total number of scheduling periods throughout the day, taken as 24 hours. express Time-of-use electricity pricing for industrial and commercial users during specific time periods; This represents the minimum discount factor for the electricity purchase price for pumping water, taken as 0.75; This represents the unit dispatch duration. In this embodiment, the low-storage-high-generation arbitrage profit model in the electricity market is used to calculate the daily net profit of a power station based on the operation mode of distributed pumped storage hydropower pumping during low-price periods and generating electricity during high-price periods. The expression for the mainnet ancillary service market frequency regulation and reserve revenue model is as follows: (11) In the formula: This represents the total revenue from mainnet auxiliary services on a single day. This indicates the price for frequency modulation capacity compensation; This indicates the price for compensation for spare capacity; express The system's backup capacity must be reported in real time. and These represent the objective probability coefficients for winning and clearing bids for frequency modulation capacity and reserve capacity, respectively. express Daily forecast of mileage pricing for time-of-use frequency (TOF) radio. This represents the ratio of frequency modulation mileage to frequency modulation capacity. This represents the probability coefficient of triggering the frequency modulation command in actual operation; in this embodiment, the frequency modulation and reserve revenue model of the main grid ancillary service market is used to calculate the total daily revenue of the main grid ancillary services based on the operation mode of distributed pumped storage relying on aggregators to participate in the main grid ancillary service market. The expression for the user demand control revenue model is: (12) In the formula: This indicates the revenue share from managing the maximum daily demand. This represents the profit-sharing coefficient, indicating that the energy storage provider extracts a portion of the cost savings as revenue from its value-added services. This indicates the maximum demand load power on the user side before the pumped storage power station was put into operation; This indicates the maximum demand load power on the user side after pumped storage participates in peak shaving and valley filling; This represents the basic electricity price per unit for the maximum demand of industrial and commercial users converted to a single day. In this embodiment, the user demand control revenue model is used to calculate the revenue sharing of the maximum demand management for single day based on the operation mode of distributed pumped storage to reduce the maximum demand of industrial and commercial users. The expression for the life-cycle cost model is as follows: (13) (14) (15) (16) (17) In the formula: This indicates the total initial investment cost of the power plant; This indicates the unit cost of constructing the upstream reservoir capacity per cubic meter. Indicates the storage capacity; This indicates the unit price per kilowatt of installed capacity; Indicates the unit's assembly capacity; This represents the total present value of operating and maintenance costs over the entire life cycle. Indicates the overall project evaluation period for the power plant; This represents the average annual maintenance cost per unit volume of the warehouse. This represents the average annual operating and maintenance cost per unit of the unit's installed capacity. This indicates the project's discount rate; This indicates a specific sequence of years in which the project has been operating; This represents the total present value of unit replacement costs over the entire life cycle; This indicates the unit cost per kilowatt for a single replacement of a water pump turbine unit. Indicates the entire life cycle of the generator unit; Indicates the total number of replacements required within the project's lifespan; Indicates intermediate parameters; This represents the annualized equivalent comprehensive cost of a distributed pumped storage power station, which is equivalent to the total investment cost. This is converted to the annualized equivalent comprehensive cost of a distributed pumped storage power station. In this embodiment, the life-cycle cost model is used to calculate the present value of the initial construction investment cost, operation and maintenance cost, and unit equipment replacement cost based on the initial construction investment, operation and maintenance cost, and unit equipment replacement cost of distributed pumped storage power stations. S5: Based on steps S2 to S4, with the annualized comprehensive net cost of the system and the daily average fluctuation variance of the adjusted residual load power as the optimization objectives, construct a distributed pumped storage capacity optimization configuration model. Specifically, in this embodiment, based on the typical daily scenario and annual equivalent operation evaluation results obtained in step S2, the distributed pumped storage physical operation model obtained in step S3, and the comprehensive economic evaluation model obtained in step S4, a capacity optimization configuration model is constructed with the objectives of minimizing the system's annualized comprehensive net cost and minimizing the daily average fluctuation variance of the adjusted surplus load power. This specifically includes: S51: Define the installed capacity of distributed pumped storage power stations and the upstream reservoir volume as decision variables for the capacity optimization configuration model; S52: Based on the annualized equivalent comprehensive cost and diversified revenue results obtained in step S4, an economic objective function is constructed with the goal of minimizing the system's annualized comprehensive net cost. Its expression is as follows: (18) In the formula: This represents the annualized equivalent comprehensive cost based on distributed pumped storage power stations. Equivalent to the annual operation evaluation results The resulting annualized net cost of the system; Indicates the total number of typical scenarios; Indicates the typical day scene number; Indicates the first The annual time weighting of days for a typical scenario; They represent the first The net revenue from electricity arbitrage, power distribution and storage sharing, ancillary services, and demand management calculated by the daily model in each scenario. For the first In a single-day model under a specific scenario Daily operation evaluation results; This represents the annualized equivalent total investment cost of a distributed pumped storage power station, including replacement and operation and maintenance. S53: Based on the load, wind power, and typical daily photovoltaic scenario data obtained in step S2, calculate the unbalanced net load power of the distribution network system before pumped storage is connected. The expression is as follows: (19) In the formula: They represent the first In each scenario The original base load, wind power output, and photovoltaic power output of the power distribution network at all times; This indicates the unbalanced net load power of the distribution network system before the pumped storage system is connected; S54: Based on the integrated power generation and integrated pumping power of pumped storage obtained in step S3, calculate the equivalent surplus load power after peak shaving and valley filling regulation by introducing distributed pumped storage. The expression is as follows: (20) In the formula: and They represent the first In each scenario The equivalent comprehensive power generation and equivalent comprehensive pumping power of the pumped storage system at all times; S55: Based on the equivalent surplus load power obtained in step S54, and with the goal of minimizing the daily average fluctuation variance of the adjusted surplus load power, a technical objective function is constructed, the expression of which is: (twenty one) S56: Based on the physical operation model obtained in step S3 and the business operation relationship obtained in step S4, construct the capacity optimization configuration model constraints, including pumped storage capacity constraints, upper and lower limit constraints on operating power, pumped power generation state constraints, energy evolution and state of charge constraints of the energy storage system, power balance and interaction constraints of distribution network nodes, and capacity allocation constraints for multi-business scenarios. Its expression is: The pumped storage capacity constraint is: (twenty two) (twenty three) In the formula: These represent the upper limit of upstream reservoir capacity and the upper limit of installed capacity, respectively. The upper and lower limits of the operating power are constrained as follows: (twenty four) (25) (26) (27) In the formula: They represent the first In each scenario The power generation capacity of the electricity market, the reserve capacity of the frequency regulation market, and the emergency reserve capacity at any given time; These represent the pumping power in the electric energy market and the pumping reserve capacity in the frequency regulation market, respectively. Indicates installed capacity; The state constraints for pumped hydropower generation are: (28) (29) (30) (31) (32) (33) (34) The energy evolution and state of charge constraints of the energy storage system are as follows: (35) (36) (37) In the formula: Indicates the first In each scenario The pumping flow rate at any given time; Indicates the first In each scenario Power generation flow rate at any given moment; Indicates the first In each scenario The water storage volume of the upstream reservoir at any given time; Indicates the first In each scenario The water storage volume of the upstream reservoir at any given time; The power balance and interaction constraints of the distribution network nodes are as follows: (38) S6: Solve the distributed pumped storage capacity optimization configuration model to obtain the Pareto non-dominated solution set; use fuzzy set theory to comprehensively evaluate the Pareto non-dominated solutions in the Pareto non-dominated solution set to obtain the recommended scheme for distributed pumped storage capacity configuration, specifically including the following steps: S61: The installed capacity of the distributed pumped storage power station and the upstream reservoir volume are used as capacity decision variables; the NSGA-II algorithm is used to perform multi-objective capacity optimization on the constructed distributed pumped storage capacity optimization configuration model to obtain the Pareto non-dominated solution set; Furthermore, each Pareto nondominated solution in the Pareto nondominated solution set corresponds to a capacity configuration scheme for a set of capacity decision variables; Specifically, in this embodiment, the capacity optimization configuration model constructed in step S5 is solved using a double-layer nested solution strategy to obtain the Pareto non-dominated solution set for distributed pumped storage capacity configuration, which specifically includes the following steps: S611: Using the installed capacity of distributed pumped storage power stations and the upstream reservoir volume as outer capacity decision variables, the NSGA-II algorithm is used to perform multi-objective capacity optimization on the capacity optimization configuration model to obtain several sets of capacity configuration schemes. S612: For each set of capacity configuration schemes generated in step S611, pass it to the inner production simulation model, i.e., the capacity optimization configuration model. S613: Given the installed capacity and upstream storage volume, the inner production simulation model solves the intraday operation scheduling results under each typical daily scenario based on the typical daily scenario obtained in step S2, the physical operation model obtained in step S3, the revenue and cost model obtained in step S4, and the operation constraints obtained in step S5. S614: Feed back the revenue results and net load fluctuation results obtained in step S613 for each typical daily scenario to the outer capacity optimization configuration model. The inner production simulation model performs operation optimization based on the outer capacity decision variables and returns the calculated economic objective function value and technical objective function value to the outer capacity optimization configuration model. S615: The outer capacity decision variables are used to calculate economic and technical objectives based on the feedback from step S614, and a multi-objective optimization algorithm is used for iterative optimization. The specific process is as follows: Initialize the capacity configuration scheme population, each capacity configuration scheme includes the reservoir capacity, pumped storage unit capacity, and capacity allocation ratio for each application scenario; input each capacity configuration scheme into the inner production simulation model established in step S612 to obtain the corresponding economic and technical objective evaluation results; generate the next generation of capacity configuration schemes based on the non-dominated sorting, congestion distance calculation, and genetic operations of the multi-objective optimization algorithm; repeat the above process until the preset number of iterations is reached or the convergence condition is met, and finally obtain the Pareto non-dominated solution set corresponding to the capacity configuration scheme; S62: A comprehensive evaluation of Pareto non-dominated solutions in the Pareto non-dominated solution set is performed using fuzzy set theory, specifically including: S621: Using a linear trapezoidal fuzzy membership function, combined with the objective function of the distributed pumped storage capacity optimization configuration model, the target satisfaction of each Pareto non-dominated solution in the Pareto non-dominated solution set is obtained. And the expression for the target satisfaction level is: (39) In the formula: Indicates the first The capacity configuration scheme corresponds to the first One objective function value; They represent the first and second solutions in the Pareto non-dominated set, respectively. The minimum and maximum values ​​of the objective function; Indicates the first The capacity configuration scheme is in the first Satisfaction on the objective function; In this embodiment, based on the obtained objective function values, i.e., formulas (18) and (21), a linear trapezoidal fuzzy membership function is used to normalize the satisfaction of each capacity configuration scheme on different objective functions; For the first objective function... The capacity configuration scheme of the first Each objective function value Based on the minimum value of the objective function in the Pareto non-dominated solution set and Calculate their target satisfaction ; S622: Based on the target satisfaction of each Pareto non-dominated solution, obtain the comprehensive evaluation satisfaction of each capacity configuration scheme; and the expression for the comprehensive evaluation satisfaction is: (40) In the formula: Indicates the first Overall satisfaction rating of each capacity configuration scheme; This represents the total number of capacity configuration schemes in the Pareto non-dominated solution set; This represents the total number of objective functions, including economic objective functions and technical objective functions. ; S63: Select the configuration scheme with the highest overall satisfaction as the recommended configuration scheme for distributed pumped storage capacity. This embodiment also includes, based on the distribution network topology parameters obtained in step S1, performing a distribution network operation safety verification on the capacity configuration recommendation scheme obtained in step S6, and outputting the final capacity configuration scheme. Specifically, it includes the following: S71: Based on the distribution network node connection relationship, branch parameters and operation constraint parameters obtained in step S1, the distribution network power flow model is obtained, and the pumped storage access node, installed capacity, upstream reservoir volume and operating power of each time period in the capacity configuration recommendation scheme obtained in S6 are mapped to the corresponding distribution network node. S72: Based on the pre-set Dist Flow power flow model, power flow calculation is performed on the distribution network after pumped storage is connected. Based on the upper and lower limits of node voltage, branch power flow constraints and root node interactive power constraints, the operation safety is verified for node voltage, branch power flow and root node power purchase. S721: If all node voltages are within the allowable voltage range, all branch power flows do not exceed the allowable transmission capacity of the corresponding branch, and the power purchased by the root node meets the corresponding interactive power limit, then the recommended capacity configuration scheme is determined to meet the safety requirements of the distribution network operation, and the final capacity configuration scheme is output. S722: If there are node voltage exceeding limits, branch power flow exceeding limits, or root node interactive power not meeting the requirements, then the pumped storage output will be corrected by topological constraints, or the operation safety constraints will be fed back to the capacity optimization configuration model for re-solution until a capacity configuration scheme that meets the operation safety requirements of the distribution network is obtained. S723: If there are node voltage exceeding limits, branch power flow exceeding limits, or root node interaction power not meeting the requirements, then under the condition of keeping the pumped storage access node, installed capacity, and upstream reservoir volume unchanged, the pumping power and power generation power corresponding to the recommended scheme obtained in step S6 are used as a reference. The goal is to minimize the deviation between the corrected pumped storage operating power and the original operating power. Under the condition of satisfying node voltage constraints, branch power flow constraints, root node interaction power constraints, and pumped storage physical operation constraints, the operating power is re-solved to obtain the topology constraint corrected operating scheme. S73: If, after fixing the access node, installed capacity, and upstream reservoir volume in step S723, there is no feasible operating scheme that meets the safety constraints of the distribution network operation, then the node voltage constraint, branch power flow constraint, and root node interactive power constraint in step S72 are added to the capacity optimization configuration model described in step S5, and step S6 is re-executed to optimize the pumped storage installed capacity and upstream reservoir volume until a capacity configuration scheme that meets the safety constraints of the distribution network operation is obtained.

[0020] This embodiment includes the following application example: The hardware device used in this embodiment includes a PC. The PC runs on a Windows 11 operating system, the simulation platform is MATLAB R2025a, the optimization modeling tool is YALMIP, and the inner-layer optimization solver is Gurobi.

[0021] The method for optimizing the allocation of distributed pumped storage capacity in the distribution network in this embodiment includes the following steps: Step 1: Collect the basic data required for the optimal configuration of distributed pumped storage capacity in the distribution network, including distribution network load data, wind power output data, photovoltaic power output data, candidate construction data of distributed pumped storage power stations, pumped storage unit operating parameters, market price parameters, investment and construction cost parameters, operation and maintenance cost parameters, and distribution network topology parameters, to form a basic dataset for the optimal configuration of distributed pumped storage capacity in the distribution network.

[0022] In this embodiment, the total installed capacity of the wind power cluster is 180MW, the total installed capacity of the photovoltaic cluster is 12.6MW, and the load capacity is 120MW; the pumped storage power station has a project life of 25 years, a turbine lifespan of 15 years, a discount rate of 0.05, a design head of 79m, a generator efficiency of 0.64, a pump efficiency of 0.65, and a pipeline efficiency of 0.95. Specific parameters are shown in Tables 1, 2, and 3.

[0023] Table 1. Simulation Example Parameter Settings

[0024] Table 2. Configuration parameters of pumped storage power stations

[0025] Table 3. Investment and construction cost parameters for pumped storage power stations

[0026] Step 2: Based on the distribution network load data, wind power output data and photovoltaic output data obtained in Step 1, extract typical daily scenarios and perform equivalent processing for year-round time-series production simulation.

[0027] In this embodiment, the K-means clustering method is used to extract typical daily scenarios from the annual load, wind power, and photovoltaic time-series data, resulting in four types of typical daily scenarios: spring, summer, autumn, and winter. The wind and solar load output curves for typical days in each season are shown below. Figures 8 to 11 As shown.

[0028] Step 3: Based on the pumped storage unit operating parameters and reservoir capacity parameters obtained in Step 1, construct a distributed pumped storage dynamic evolution model for pumping conditions, power generation conditions, and reservoir capacity, following the method described in Step 3 of the invention. In this embodiment, the distributed pumped storage power station is built relying on small reservoirs on the distribution network side or water resources with modification potential, realizing the conversion between electrical energy and water potential energy through water transfer between the upper and lower reservoirs. Its physical structure is shown in the attached figure. Figure 4 As shown.

[0029] Step 4: Based on the market price parameters, investment and construction cost parameters, and operation and maintenance cost parameters obtained in Step 1, construct the following models according to the method described in Step 4 of the invention: a shared capacity revenue model for distribution and storage, a low-storage-high-generation arbitrage revenue model for the electricity market, a frequency regulation and reserve revenue model for the main grid ancillary service market, a demand control revenue model for industrial and commercial users, and a life-cycle cost model. In this embodiment, distributed pumped storage provides capacity sharing services to new energy power plants within the distribution network through a shared energy storage model, achieves low-storage-high-generation arbitrage through the electricity market, participates in the frequency regulation and reserve ancillary service markets through an aggregator model, and obtains demand control revenue by reducing the maximum demand of industrial and commercial users. The shared energy storage operation model within the distribution network is attached. Figure 6 As shown. Step 5: Based on the typical daily scenario and annual equivalent operation evaluation results obtained in Step 2, the distributed pumped storage physical operation model obtained in Step 3, and the revenue-cost model obtained in Step 4, a capacity optimization configuration model is constructed according to the method described in Step 5 of the invention, with the objectives of minimizing the system's annualized comprehensive net cost and minimizing the daily average fluctuation variance of the adjusted surplus load power. In this embodiment, the capacity optimization configuration model uses the installed capacity of the distributed pumped storage power station and the upstream reservoir volume as decision variables, and simultaneously considers pumped storage capacity constraints, upper and lower limits of operating power constraints, pumped power generation status constraints, reservoir capacity evolution constraints, distribution network interactive power constraints, and multi-service capacity allocation constraints. The structure of the capacity optimization configuration model is attached. Figure 2 As shown.

[0030] Step 6: Based on the capacity optimization configuration model constructed in Step 5, a two-layer nested solution strategy is used to obtain the Pareto non-dominated solution set for distributed pumped storage capacity configuration. In this embodiment, the outer layer uses the NSGA-II algorithm to perform multi-objective optimization of installed capacity and upstream reservoir volume, while the inner layer uses YALMIP to establish a production simulation optimization model and calls Gurobi to solve the operation scheduling results under various typical daily scenarios. The two-layer nested optimization solution process is shown in the attached figure. Figure 3 As shown. The Pareto non-dominated solution set obtained by solving is as follows. Figure 12 As shown in the figure. Based on the Pareto non-dominated solution set obtained in step 6, fuzzy set theory is used to comprehensively evaluate each capacity configuration scheme, resulting in a recommended distributed pumped storage capacity configuration scheme. In this embodiment, after compromise decision-making using fuzzy set theory, the optimal aggregated installed capacity is 24.95MW, the optimal aggregated reservoir capacity is 936,000 m³, and the annual net profit is 12.8052 million yuan. The capacity optimization configuration results are shown in Table 4, and the diversified revenue composition is as follows: Figure 13 As shown.

[0031] Table 4. Solution results of the distributed pumped storage capacity optimization configuration model for distribution networks

[0032] Step 7: Based on the distribution network topology parameters obtained in Step 1, perform a distribution network operation safety verification on the capacity configuration recommendation scheme obtained in Step 7, and output the final capacity configuration scheme. In this embodiment, the IEEE 33-node distribution network standard example is used for operation safety verification, and nodes 12 and 25 are selected as distributed pumped storage access nodes. The IEEE 33-node topology is shown in the attached figure. Figure 7As shown in Table 5, node 12 is equipped with a pumped storage capacity of 1.2943 MW and a reservoir capacity of 31,545 m³; node 25 is equipped with a pumped storage capacity of 1.6967 MW and a reservoir capacity of 30,967 m³. The total installed capacity of the two pumped storage power stations is 2.9910 MW, and the total reservoir capacity is 62,512 m³.

[0033] Table 5. IEEE 33-node pumped storage capacity configuration results

[0034] This embodiment verifies the root node power purchase, node voltage, and branch power flow after pumped storage is connected. The verification results show that the node voltage and branch power flow all meet the operational safety constraints after connection. A comparison of the root node power purchase is provided below. Figures 14 to 17 As shown, the spatiotemporal distribution of typical daily node voltages in summer is as follows: Figure 18 As shown.

[0035] In summary, the method described in this embodiment completes the distributed pumped storage capacity optimization configuration process for the distribution network according to steps 1 to 7. The implementation results show that the method described in this embodiment can obtain a distributed pumped storage capacity configuration scheme that balances economy, load smoothing, and operational safety based on source-load time-series characteristics, pumped storage physical operation constraints, a multi-revenue model, and distribution network operation safety constraints.

[0036] Compared with the prior art, the beneficial effects of the method described in this embodiment are as follows: (1) The method described in this embodiment extracts typical daily scenarios based on historical time-series data of load, wind power and photovoltaic, and performs equivalent processing of annual time-series production simulation. It can reduce the amount of calculation every hour throughout the year while retaining the source-load fluctuation characteristics and improve the adaptability of capacity configuration results to different operating scenarios.

[0037] (2) The method described in this embodiment constructs a dynamic evolution model of pumping operation, power generation operation and reservoir capacity of distributed pumped storage, and limits the operation boundary by means of pumping power generation state constraints, upper and lower limits of reservoir capacity constraints and first and last reservoir capacity balance constraints, so that the capacity configuration result can conform to the actual physical operation process of pumped storage power station.

[0038] (3) The method described in this embodiment introduces a shared energy storage and aggregator collaborative operation mechanism, which incorporates the revenue from shared energy storage capacity, the arbitrage revenue from low storage and high energy generation in the power market, the revenue from frequency regulation and reserve in the main grid auxiliary service market, and the revenue from demand control of industrial and commercial users into the comprehensive revenue evaluation, which can improve the problem of incomplete revenue evaluation of distributed pumped storage.

[0039] (4) The method described in this embodiment incorporates the total life cycle investment cost and multiple benefits into the economic objective, and uses the daily average fluctuation variance of the adjusted residual load power as the technical objective, which can achieve a compromise between the project economy and the smoothness of the distribution network operation.

[0040] (5) The method described in this embodiment adopts a two-layer nested solution strategy that combines the outer layer multi-objective capacity optimization and the inner layer production simulation scheduling with the NSGA-II algorithm. This can realize the linkage between capacity configuration and intraday operation scheduling, and improve the executability of the capacity configuration scheme.

[0041] (6) The method described in this embodiment introduces a distribution network topology operation safety verification after the capacity configuration recommendation scheme is formed. It can determine whether the node voltage, branch power flow and root node power purchase meet the operation requirements after the distributed pumped storage is connected, thereby improving the operation safety of the scheme in the actual distribution network.

[0042] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for optimizing configuration of distributed pumped hydro storage capacity based on shared aggregation, characterized in that, Specifically, the following steps are included: S1: Obtain the basic data required for the optimized configuration of distributed pumped storage capacity; Furthermore, the basic data includes power distribution network load data, wind power output data, photovoltaic power output data, pumped storage power station parameters, market price parameters, investment and construction cost parameters, and operation and maintenance cost parameters; S2: Based on the distribution network load data, wind power output data, and photovoltaic output data, typical daily scenarios are extracted from the annual source-load time series data to obtain multiple typical daily scenarios; based on the occurrence probability or time weight of each typical daily scenario in the annual sample, the annual time series production simulation equivalent processing is performed to obtain the annual equivalent operation evaluation results; S3: Based on the parameters of the pumped storage power station, construct a distributed pumped storage pumping, power generation and reservoir capacity dynamic evolution model to obtain the equivalent comprehensive power generation power and equivalent comprehensive pumping power of the pumped storage. S4: Construct a comprehensive economic evaluation model for distributed pumped storage based on market price parameters, investment and construction cost parameters, and operation and maintenance cost parameters; S5: Based on steps S2 to S4, with the annualized comprehensive net cost of the system and the daily average fluctuation variance of the adjusted residual load power as the optimization objectives, construct a distributed pumped storage capacity optimization configuration model. S6: Solve the distributed pumped storage capacity optimization configuration model to obtain the Pareto non-dominated solution set; use fuzzy set theory to comprehensively evaluate the Pareto non-dominated solutions in the Pareto non-dominated solution set to obtain the recommended scheme for distributed pumped storage capacity configuration.

2. The method of claim 1, wherein, S2 specifically includes the following steps: S21: Extract typical daily scenarios from the annual source-load time series data to obtain daily-scale source-load feature vectors; and the daily-scale source-load feature vectors include distribution network load data, wind power output data, and photovoltaic power output data corresponding to each natural day; S22: Normalize the eigenvectors of the day-scale source load; S23: Based on S22 and combined with the K-means clustering algorithm, the daily-scale source load feature vectors are clustered to obtain several clusters; multiple typical daily scenes are defined based on the clusters. And the clustering objective function of the K-means clustering algorithm is: （1） In the formula: This represents the value of the clustering objective function; Indicates the number of cluster categories; Indicates the first The nth cluster, that is, the one that is divided into the nth cluster. The set of daily-scale source load eigenvectors of the class; Indicates the normalized result of the first... Daily-scale source load feature vectors; Indicates the first Cluster centers of sample groups; S24: Based on the number of daily-scale source load samples contained in the cluster, determine the occurrence probability or time weight of the corresponding typical daily scene in the annual sample according to the expert database, and perform equivalent processing of the annual time-series production simulation based on the occurrence probability or time weight to obtain the annual equivalent operation evaluation results. And the expression for the annual equivalent operation evaluation result is: （2） In the formula: Indicates the annual equivalent operation evaluation indicators; Indicates the number of days the system operates throughout the year; Indicates the first The probability or time weight of occurrence of typical daily scenarios; Indicates the first Daily operation evaluation results under typical daily scenarios.

3. The method for optimizing the allocation of distributed pumped storage capacity based on shared aggregation according to claim 2, characterized in that, In S3, a distributed pumped storage system is constructed to dynamically evolve its pumping, power generation, and reservoir capacity. The expression for this model is: （3） （4） （5） （6） （7） （8） In the formula: express The equivalent comprehensive pumping power of the pumped storage unit at all times; and They represent The pumping capacity declared in the power market and the frequency regulation reserve market at all times; express real-time pumping flow rate; This indicates the velocity-to-power ratio under pumping conditions. Indicates the pump's operating efficiency; Indicates pipeline efficiency; This represents the density of water; This represents standard gravitational acceleration; Indicates the design head height of the pumped storage unit; express The equivalent comprehensive power generation capacity of the pumped storage unit at all times; and These represent pumped storage. The power generation capacity declared in the electricity market and the power generation capacity declared in the frequency regulation reserve market at any time; express Real-time power generation flow; This indicates the velocity-to-power ratio under power generation conditions. Indicates the generator set operating efficiency; express The water storage volume of the upstream reservoir at any given time; express The water storage volume of the upstream reservoir at any given time; Indicates the unit scheduling duration; This indicates the maximum capacity of the upstream reservoir.

4. The method for optimizing the allocation of distributed pumped storage capacity based on shared aggregation according to claim 3, characterized in that, The distributed pumped storage comprehensive economic evaluation model constructed in S4 includes a shared revenue model for storage capacity, a low-storage-high-occupancy arbitrage revenue model for the power market, a frequency regulation and reserve revenue model for the main grid ancillary service market, a user demand control revenue model, and a life cycle cost model. The expression for the shared revenue model of the storage capacity is: （9） In the formula: This indicates the average daily capacity substitution revenue of the power plant; This indicates the approved lease rate for distributed energy storage capacity in the regional market; This indicates the total installed active power capacity of the power station. Indicates the power unit conversion constant; Indicates the capacity isolation allocation factor; Indicates the total number of days in a calendar year; The expression for the low-storage, high-inflation arbitrage profit model in the electricity market is as follows: （10） In the formula: This indicates the power plant's net revenue from low energy storage and high power generation on a single day; This indicates the total number of scheduling periods throughout the day; express Time-of-use electricity pricing for industrial and commercial users during specific time periods; This represents the minimum discount factor for the electricity purchase price for pumping water; Indicates the unit scheduling duration; The expression for the mainnet ancillary service market frequency regulation and reserve revenue model is as follows: （11） In the formula: This represents the total revenue from mainnet auxiliary services on a single day. This indicates the price for frequency modulation capacity compensation; This indicates the price for compensation for spare capacity; express The system's backup capacity must be reported in real time. and These represent the objective probability coefficients for winning and clearing bids for frequency modulation capacity and reserve capacity, respectively. express Daily forecast of mileage pricing for time-of-use frequency (TOF) radio. This represents the ratio of frequency modulation mileage to frequency modulation capacity. This represents the probability coefficient of triggering the frequency modulation command during actual operation; The expression for the user demand control revenue model is: （12） In the formula: This indicates the revenue share from managing the maximum daily demand. This indicates the profit-sharing coefficient; This indicates the maximum demand load power on the user side before the pumped storage power station was put into operation; This indicates the maximum demand load power on the user side after pumped storage participates in peak shaving and valley filling; This represents the basic electricity rate per unit of the maximum demand for industrial and commercial use, converted to a single-day basis. The expression for the life-cycle cost model is as follows: （13） （14） （15） （16） （17） In the formula: This indicates the total initial investment cost of the power plant; This indicates the unit cost of constructing the upstream reservoir capacity per cubic meter. Indicates the storage capacity; This indicates the unit price per kilowatt of installed capacity; Indicates the unit's assembly capacity; This represents the total present value of operating and maintenance costs over the entire life cycle. Indicates the overall project evaluation period for the power plant; This represents the average annual maintenance cost per unit volume of the warehouse. This represents the average annual operating and maintenance cost per unit of the unit's installed capacity. This indicates the project's discount rate; This indicates a specific sequence of years in which the project has been operating; This represents the total present value of unit replacement costs over the entire life cycle; This indicates the unit cost per kilowatt for a single replacement of a water pump turbine unit. Indicates the entire life cycle of the generator unit; Indicates the total number of replacements required within the project's lifespan; Indicates intermediate parameters; This represents the annualized equivalent comprehensive cost of a distributed pumped storage power station.

5. The method for optimizing the allocation of distributed pumped storage capacity based on shared aggregation according to claim 4, characterized in that, The distributed pumped storage capacity optimization configuration model constructed in S5 includes an objective function and constraints with the annualized comprehensive net cost of the system and the daily average fluctuation variance of the adjusted surplus load power as the optimization objectives. Furthermore, the objective function includes an economic objective function constructed with the goal of minimizing the system's annualized comprehensive net cost. The technical objective function is constructed with the goal of minimizing the daily average fluctuation variance of the adjusted surplus load power. ; The expression for the economic objective function is: （18） In the formula: This represents the annualized equivalent comprehensive cost based on distributed pumped storage power stations. Equivalent to the annual operation evaluation results The resulting annualized net cost of the system; Indicates the total number of typical scenarios; Indicates the typical day scene number; Indicates the first The annual time weighting of days for a typical scenario; They represent the first The net revenue from electricity arbitrage, power distribution and storage sharing, ancillary services, and demand management calculated by the daily model in each scenario. For the first In a single-day model under a specific scenario Daily operation evaluation results; This represents the annualized equivalent total investment cost of a distributed pumped storage power station, including replacement and operation and maintenance. The expression for the technical objective function is: （19） （20） （21） In the formula: They represent the first In each scenario The original base load, wind power output, and photovoltaic power output of the power distribution network at all times; This indicates the unbalanced net load power of the distribution network system before the pumped storage system is connected; and They represent the first In each scenario The equivalent comprehensive power generation and equivalent comprehensive pumping power of the pumped storage system at all times; This represents the equivalent surplus load power after introducing distributed pumped storage for peak shaving and valley filling regulation. The constraints include pumped storage capacity constraints, upper and lower limits of operating power constraints, pumped power generation status constraints, energy evolution and state of charge constraints of energy storage systems, and power balance and interaction constraints of distribution network nodes. The pumped storage capacity constraint is: （22） （23） In the formula: These represent the upper limit of upstream reservoir capacity and the upper limit of installed capacity, respectively. The upper and lower limits of the operating power are constrained as follows: （24） （25） （26） （27） In the formula: They represent the first In each scenario The power generation capacity of the electricity market, the reserve capacity of the frequency regulation market, and the emergency reserve capacity at any given time; These represent the pumping power in the electric energy market and the pumping reserve capacity in the frequency regulation market, respectively. Indicates installed capacity; The state constraints for pumped hydropower generation are: （28） （29） （30） （31） （32） （33） （34） The energy evolution and state of charge constraints of the energy storage system are as follows: （35） （36） （37） In the formula: Indicates the first In each scenario The pumping flow rate at any given time; Indicates the first In each scenario Power generation flow rate at any given moment; Indicates the first In each scenario The water storage volume of the upstream reservoir at any given time; Indicates the first In each scenario The water storage volume of the upstream reservoir at any given time; The power balance and interaction constraints of the distribution network nodes are as follows: （38）。 6. The method for optimizing the allocation of distributed pumped storage capacity based on shared aggregation according to claim 5, characterized in that, S6 specifically includes the following steps: S61: The installed capacity of the distributed pumped storage power station and the upstream reservoir volume are used as capacity decision variables; the NSGA-II algorithm is used to perform multi-objective capacity optimization on the constructed distributed pumped storage capacity optimization configuration model to obtain the Pareto non-dominated solution set; Furthermore, each Pareto nondominated solution in the Pareto nondominated solution set corresponds to a capacity configuration scheme for a set of capacity decision variables; S62: A comprehensive evaluation of Pareto non-dominated solutions in the Pareto non-dominated solution set is performed using fuzzy set theory, specifically including: S621: Using a linear trapezoidal fuzzy membership function, combined with the objective function of the distributed pumped storage capacity optimization configuration model, the target satisfaction of each Pareto non-dominated solution in the Pareto non-dominated solution set is obtained. And the expression for the target satisfaction level is: （39） In the formula: Indicates the first The capacity configuration scheme corresponds to the first One objective function value; They represent the first and second solutions in the Pareto non-dominated set, respectively. The minimum and maximum values ​​of the objective function; Indicates the first The capacity configuration scheme is in the first Satisfaction with each objective function; S622: Based on the target satisfaction of each Pareto non-dominated solution, obtain the comprehensive evaluation satisfaction of each capacity configuration scheme; and the expression for the comprehensive evaluation satisfaction is: （40） In the formula: Indicates the first Overall satisfaction rating of each capacity configuration scheme; This represents the total number of capacity configuration schemes in the Pareto non-dominated solution set; This represents the total number of objective functions, including economic objective functions and technical objective functions. ; S63: Select the configuration scheme with the highest overall satisfaction as the recommended configuration scheme for distributed pumped storage capacity.

Images