Pumped storage power station capacity configuration optimization method and device

Abstract

The invention discloses a pumped storage power station capacity configuration optimization method and device, and relates to the technical field of power system energy storage configuration.The method comprises the steps that on the basis of uncertain factors in the pumped storage power station capacity configuration optimization process, a typical daily probability distribution set based on K-Means clustering is constructed; establishing an optimization model of capacity configuration of the pumped storage power station by taking the highest satisfaction rate of the hybrid pumping storage income and the peak regulation demand of the power grid as an objective function; converting the optimization model into a double-target linear programming model by using a piecewise linear approximation method, a binary variable method and a large M method; converting the dual-target linear programming model into a single-target linear programming model by using a normalization and weighted summation method; and solving the single-target linear programming model by using a Gurobi solver to obtain the optimal configuration capacity and operation scheme of the pumped storage power station.

Description

Technical Field

[0001] This invention relates to the field of power system energy storage configuration technology, and in particular to a method and apparatus for optimizing the capacity configuration of pumped storage power stations. Background Technology

[0002] Pumped storage power stations, as the most mature and largest-scale energy storage technology currently available, play a vital role in peak shaving and valley filling, frequency regulation, and reserve capacity provision in the power system. With the increasing penetration of new energy sources, the power grid's demand for flexible resource allocation is becoming increasingly urgent, and the capacity configuration of pumped storage power stations directly affects their operational efficiency and economic benefits.

[0003] However, research on the optimization of pumped storage power station capacity configuration mainly focuses on improving the renewable energy absorption rate, increasing the overall pumping revenue of pumped storage power stations, or improving the load demand satisfaction rate. These are all single-objective optimizations, failing to consider economic efficiency, reliability, and environmental friendliness. Furthermore, they suffer from insufficient dynamic adaptability, failing to fully consider the dynamic changes in grid load and renewable energy output, leading to a mismatch between configuration schemes and actual needs. They also suffer from low computational efficiency, easily getting trapped in local optima and making it difficult to achieve global optimization. Summary of the Invention

[0004] The present invention aims to at least partially solve one of the technical problems in the related art.

[0005] This invention proposes a method for optimizing the capacity configuration of pumped storage power stations.

[0006] Another objective of this invention is to provide a capacity configuration optimization device for pumped storage power stations.

[0007] To achieve the above objectives, the present invention proposes a method for optimizing the capacity configuration of pumped storage power stations, comprising: Based on the uncertainties in the capacity configuration optimization process of pumped storage power stations, a typical daily probability distribution set based on K-Means clustering is constructed. Based on the typical daily probability distribution set, with the objective function being to maximize the mixed pumped storage revenue and the grid peak-shaving demand satisfaction rate, and according to the operational constraints including grid transmission power constraints, wind-solar-storage power output constraints, capacity constraints, power balance constraints, and wind-solar absorption rate constraints, an optimization model for the capacity configuration of pumped storage power stations is established. The optimization model is transformed into a bi-objective linear programming model using the piecewise linear approximation method, the binary variable method, and the Big M method. The bi-objective linear programming model is transformed into a single-objective linear programming model using normalization and weighted summation. The single-objective linear programming model was solved using the Gurobi solver to obtain the optimal configuration capacity and operation scheme of the pumped storage power station.

[0008] The pumped storage power station capacity configuration optimization method of this invention may also have the following additional technical features: In one embodiment of the present invention, the step of constructing a typical daily probability distribution set based on K-Means clustering, based on uncertainties in the capacity configuration optimization process of pumped storage power stations, includes: Obtain the historical time series of the theoretical maximum wind power output of the power grid system over many years: , Photovoltaic theoretical maximum output time series: , Firepower output time series: , Power grid load time series: , Time series of electricity prices for pumped storage power plants: , Time series of pumped storage power station pumped hydroelectric pricing: ; For each time period Constructing multidimensional feature vectors: The multidimensional feature vector contains the first... The theoretical maximum output of wind and solar power, the actual output of thermal power, the grid load, and the value of pumped storage power station pumping and generating electricity during the specified time period; By dividing the historical data of the power grid system into multiple sample sets using a 24-hour period as the unit, the optimal sample set is determined using the elbow method. The value is obtained by performing K-means clustering on the sample set. A typical day and its corresponding probability.

[0009] In one embodiment of the present invention, the objective function includes:

[0010] In the formula: Indicates the annual mixed pumped storage yield; The total number of typical days; For the first The probability of a typical day occurring; The number of time periods divided for each typical day. The unit time period length; For the first A typical day Power generation output of pumped storage power stations during certain periods; For the first A typical day Pumped storage power station pumping output during certain periods; For the first A typical day The on-grid electricity price for pumped storage power plants during certain periods; For the first A typical day Pumped storage power station pumping electricity price during specific time periods; It is a capacity-based electricity price; This is the annual investment conversion factor; The annual interest rate; The design service life of the pumped storage power station; Construction cost per unit of installed capacity; This refers to the installed capacity of pumped storage power stations; For other configuration capacity The relevant annual revenue per unit of capacity; For other configuration capacity The relevant annual unit capacity cost;

[0011] In the formula: Indicates the degree of deviation in output load; The total number of typical days; For the first The probability of a typical day occurring; The number of time periods divided for each typical day. The unit time period length; For the first A typical day System power load during specific time periods; For the first A typical day Power output of thermal power units during certain periods; For the first A typical day Solar power output during specific time periods; For the first A typical day Wind power output during certain periods; For the first A typical day Power generation output of pumped storage power stations during certain periods; For the first A typical day Pumped storage power station output during certain periods.

[0012] In one embodiment of the present invention, the operational constraints include: Power transmission constraints of the power grid: ; In the formula: For the first A typical day Solar power output during specific time periods; For the first A typical day Wind power output during certain periods; This represents the maximum capacity of the wind-solar hybrid system that the power grid can accept. Wind power output limits and constraints: ; In the formula: For the first A typical day Wind power output during certain periods; For the first A typical day Theoretical maximum wind power output during a given period; Photovoltaic power output constraints: ; In the formula: For the first A typical day Solar power output during specific time periods; For the first A typical day Theoretical maximum value of photovoltaic power output during the time period; Pumped storage power generation constraints: ; In the formula: For the first A typical day Power generation output of pumped storage power stations during certain periods; This refers to the maximum power output of the pumped storage power station, i.e., its installed capacity. Pumped storage power station pumping power constraints: ; In the formula: For the first A typical day Pumped storage power station pumping output during certain periods; This represents the maximum pumping power of the pumped storage power station. Mutual exclusion constraint between pumping and power generation in pumped storage power stations: ; In the formula: For the first A typical day Power generation output of pumped storage power station during specific time periods; where For the first A typical day Pumped storage power station pumping output during certain periods; Pumped storage power station energy storage capacity constraints: ; In the formula: For the first A typical day Power generation output of pumped storage power station during specific time periods; where For the first A typical day Pumped storage power station pumping output during certain periods; To improve the overall conversion efficiency of pumped storage power stations; The unit time period length; The maximum energy storage capacity of the pumped storage power station is determined based on water level requirements and reservoir capacity. Pumped storage power station installed capacity constraints: ; In the formula: This refers to the installed capacity of pumped storage power stations; The site selection for pumped storage power stations is based on the maximum installed capacity. Power balance constraints: ; ; ; In the formula: The electricity consumed by pumped storage power stations for pumping water; This refers to the electricity generated by the pumped storage power station; To improve the overall conversion efficiency of pumped storage power stations; The number of time periods divided for each typical day. The unit time period length; For the first A typical day Power generation output of pumped storage power stations during certain periods; For the first A typical day Pumped storage power station pumping output during certain periods; Constraints on wind and solar energy absorption rates: ; In the formula: The objective function is... The number of time periods divided for each typical day. The unit time period length; For the first A typical day Solar power output during specific time periods; For the first A typical day Wind power output during certain periods; For the first A typical day Theoretical maximum value of photovoltaic power output during the time period; For the first A typical day The theoretical maximum wind power output during a given period.

[0013] In one embodiment of the present invention, the step of transforming the optimization model into a bi-objective linear programming model using the piecewise linear approximation method, the binary variable method, and the Big M method includes: For a quadratic objective function:

[0014] It is transformed into a linear objective function using a piecewise linear approximation; The specific process is as follows: Introducing the first A typical day Supply and demand deviation over a period of time: ; The maximum deviation was obtained based on historical power grid output and load data. Add maximum deviation constraint ; Will The function is divided into 5 segments, in order to... Linearization, the linearization result is as follows:

[0015] Therefore, the quadratic objective function is linearized as follows: ; For nonlinear constraints By introducing binary logical variables The Big M method transforms it into a mixed-integer linear programming problem; the specific process is as follows: Introducing binary logical variables , Indicates that the pumped storage power station generates electricity. This indicates pumping by a pumped storage power station; the power constraints for pumping output and generating output of the pumped storage power station are modified as follows: Pumped storage power generation constraints: ; Pumped storage power station pumping power constraints: .

[0016] In one embodiment of the present invention, the step of transforming the bi-objective linear programming model into a single-objective linear programming model using normalization and weighted summation includes: Construct normalization formulas to transform the two objective functions into peak shaving satisfaction rate and profit maximization rate, respectively, and control their values ​​between 0 and 1: ; ; in, The lower limit of the mixed pumping return set based on experience or investment decisions; Introduce non-negative weighting coefficients and The comprehensive optimization objective function is obtained as follows: .

[0017] In one embodiment of the present invention, the step of using the Gurobi solver to solve the single-objective linear programming model to obtain the optimal configuration capacity and operation scheme of the pumped storage power station includes: The single-objective linear programming model is established in the Python environment, and the solution set is obtained by calling the Gurobi solver. The non-negative weighting coefficients are then adjusted. and Different optimization results were obtained. Based on the requirements for economy and stability, the optimal configuration capacity was selected, and the corresponding pumped storage power station operation scheme was obtained.

[0018] To achieve the above objectives, another aspect of the present invention provides a capacity configuration optimization device for pumped storage power stations, comprising: Typical day set modeling unit is used to construct a typical day probability distribution set based on K-Means clustering, based on the uncertainties in the capacity configuration optimization process of pumped storage power stations. The system modeling unit is used to establish an optimization model for the capacity configuration of pumped storage power stations based on the typical daily probability distribution set, with the objective function being the highest hybrid pumped storage revenue and the grid peak-shaving demand satisfaction rate. It is also used to establish an optimization model for the capacity configuration of pumped storage power stations based on the operational constraints including grid transmission power constraints, wind-solar-storage power output constraints, capacity constraints, power balance constraints, and wind-solar absorption rate constraints. Linearization unit, used to transform the optimization model into a bi-objective linear programming model using piecewise linear approximation, binary variable method and big M method; A normalization unit is used to transform the bi-objective linear programming model into a single-objective linear programming model using normalization and weighted summation. The solution unit is used to solve the single-objective linear programming model using the Gurobi solver to obtain the optimal configuration capacity and operation scheme of the pumped storage power station.

[0019] The embodiments of the present invention can balance economic efficiency and load fluctuations with high computational efficiency, achieve multi-objective coordination, flexibly configure grid-side wind, solar and energy storage capacity, introduce energy storage devices to absorb larger-scale wind and solar power resources, reduce wind and solar curtailment, and make up for the lack of dynamic adaptability of traditional methods.

[0020] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0021] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 A flowchart of a pumped storage power station capacity configuration optimization method provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the power generation system provided in an embodiment of the present invention; Figure 3 This is a flowchart illustrating the solution of the capacity configuration optimization model for pumped storage power stations provided in an embodiment of the present invention. Figure 4 This is a structural diagram of a pumped storage power station capacity configuration optimization device provided in an embodiment of the present invention; Detailed Implementation Embodiments of the present invention are described in detail below. Examples of these embodiments are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0022] To address the technical problems existing in the prior art, embodiments of the present invention provide a method for optimizing the capacity configuration of pumped storage power stations. Figure 1 This is a flowchart illustrating a method for optimizing the capacity configuration of a pumped storage power station according to an embodiment of the present invention. Figure 1 As shown, the method includes the following steps: Step 101: Based on the uncertainties in the capacity configuration optimization process of pumped storage power stations, construct a typical daily probability distribution set based on K-Means clustering.

[0023] In this embodiment of the invention, to accurately reflect the multiple uncertainties in the capacity configuration optimization process of pumped storage power stations, a typical daily probability distribution set is first constructed based on historical operating data to represent typical operating modes in the long term. This step provides a representative input basis with probability distribution for the entire optimization model, ensuring that the model can take into account both practical feasibility and future operational adaptability.

[0024] This invention first obtains information about the power grid system, such as... Figure 2The historical operating data shown includes data from multiple years for the power grid system, encompassing wind farms, photovoltaic power plants, thermal power plants already connected to the grid, and pumped storage power plants under planning. The data types cover key variables such as renewable energy output, traditional energy output, grid load, and electricity price changes. Details are as follows: Wind power theoretical maximum output time series Photovoltaic theoretical maximum output time series Firepower output time series Power grid load time series Time series of electricity prices for pumped storage power plants Time series of pumped storage power station pumped water electricity prices .

[0025] In this embodiment of the invention, to more accurately characterize the temporal operation state, a multi-dimensional feature vector is constructed for each time period (e.g., hour or 15 minutes):

[0026] This multidimensional feature vector contains the first... Key operating parameters such as the theoretical maximum output of wind and solar power, the actual output of thermal power, the grid load, and the value of pumped water pumping and power generation of pumped storage power stations during different time periods are highly expressive and distinguishable.

[0027] Furthermore, in this embodiment of the invention, all data are divided into samples according to a periodic pattern of "24 hours per day," forming a sample of length [length missing]. The time series matrix sample set. Each sample is a complete operating day, containing 24 sets of 6-dimensional feature vectors, forming a representative set of operating pattern data.

[0028] To select the most suitable number of clusters In this embodiment of the invention, the elbow method is used to analyze the changing trend of the clustering cost function to ensure that the selected clustering cost function is optimized. The values ​​are designed to be representative without causing overfitting in the clustering. Finally, the K-Means clustering algorithm is used to classify the sample set, resulting in K representative "typical days" and their probabilities of occurrence in all samples, thus forming a set of typical days with a probability distribution.

[0029] Step 102: Based on the typical daily probability distribution set, with the objective function being to maximize the mixed pumped storage revenue and the grid peak-shaving demand satisfaction rate, and according to the operational constraints including grid transmission power constraints, wind-solar-storage power output constraints, capacity constraints, power balance constraints, and wind-solar absorption rate constraints, an optimization model for the capacity configuration of pumped storage power stations is established.

[0030] In this embodiment of the invention, based on the typical daily probability distribution set constructed in step 101, in order to further achieve the optimal configuration of capacity and operation coordination of pumped storage power stations, an optimization model with the dual objectives of maximizing the benefits of hybrid pumped storage and minimizing the grid peak-shaving demand satisfaction rate is established, and combined with multiple operational constraints, an integrated decision-making framework is formed.

[0031] In this embodiment of the invention, the optimization model includes two core objectives: Objective 1: Maximize the annual return of the hybrid pumped storage system. Therefore, the following objective function is constructed:

[0032] In the formula: Indicates the annual mixed pumped storage yield; The total number of typical days; For the first The probability of a typical day occurring; The number of time periods divided for each typical day. The unit time period length; For the first A typical day Power generation output of pumped storage power stations during certain periods; For the first A typical day Pumped storage power station pumping output during certain periods; For the first A typical day The on-grid electricity price for pumped storage power plants during certain periods; For the first A typical day Pumped storage power station pumping electricity price during specific time periods; It is a capacity-based electricity price; This is the annual investment conversion factor; The annual interest rate; The design service life of the pumped storage power station; Construction cost per unit of installed capacity; This refers to the installed capacity of pumped storage power stations; For other configuration capacity The relevant annual revenue per unit of capacity; For other configuration capacity The relevant annual unit capacity cost.

[0033] The objective function comprehensively considers the daily operating revenue of pumped storage power stations (i.e., the difference between buying low-priced electricity for pumping water and selling high-priced electricity for generating electricity), capacity revenue (such as capacity electricity price, subsidies, and policy benefits), and capacity costs (investment conversion, operation and maintenance expenses, etc.), thus fully reflecting the annual economic efficiency of pumped storage investment.

[0034] Objective 2: Minimize the system output-load deviation. Therefore, the following objective function for the squared deviation is constructed:

[0035] In the formula: Indicates the degree of deviation in output load; The total number of typical days; For the first The probability of a typical day occurring; The number of time periods divided for each typical day. The unit time period length; For the first A typical day System power load during specific time periods; For the first A typical day Power output of thermal power units during certain periods; For the first A typical day Solar power output during specific time periods; For the first A typical day Wind power output during certain periods; For the first A typical day Power generation output of pumped storage power stations during certain periods; For the first A typical day Pumped storage power station output during certain periods.

[0036] The objective function characterizes the degree of matching between the combined output of multiple energy sources and the actual load of the power grid. The stronger the peak-shaving capability and the smaller the deviation, the better this indicator is.

[0037] Through the design of the dual objective function described above, the embodiments of the present invention can achieve a flexible trade-off between economy and system security, and support the generation of strategies for multiple scenarios.

[0038] In this embodiment of the invention, to ensure the feasibility and engineering feasibility of the model solution, the following set of constraints is constructed: Power transmission constraints of the power grid: ; In the formula: For the first A typical day Solar power output during specific time periods; For the first A typical day Wind power output during certain periods; This represents the maximum capacity of the wind-solar hybrid system that the power grid can accept. Wind power output limits and constraints: ; In the formula: For the first A typical day Wind power output during certain periods; For the first A typical day Theoretical maximum wind power output during a given period; Photovoltaic power output constraints: ; In the formula: For the first A typical day Solar power output during specific time periods; For the first A typical day Theoretical maximum value of photovoltaic power output during the time period; Pumped storage power generation constraints: ; In the formula: For the first A typical day Power generation output of pumped storage power stations during certain periods; This refers to the maximum power output of the pumped storage power station, i.e., its installed capacity. Pumped storage power station pumping power constraints: ; In the formula: For the first A typical day Pumped storage power station pumping output during certain periods; This represents the maximum pumping power of the pumped storage power station. Mutual exclusion constraint between pumping and power generation in pumped storage power stations: ; In the formula: For the first A typical day Power generation output of pumped storage power station during specific time periods; where For the first A typical day Pumped storage power station pumping output during certain periods; Pumped storage power station energy storage capacity constraints: ; In the formula: For the first A typical day Power generation output of pumped storage power station during specific time periods; where For the first A typical day Pumped storage power station pumping output during certain periods; To improve the overall conversion efficiency of pumped storage power stations; The unit time period length; The maximum energy storage capacity of the pumped storage power station is determined based on water level requirements and reservoir capacity. Pumped storage power station installed capacity constraints: ; In the formula: This refers to the installed capacity of pumped storage power stations; The site selection for pumped storage power stations is based on the maximum installed capacity. Power balance constraints: ; ; ; In the formula: The electricity consumed by pumped storage power stations for pumping water; This refers to the electricity generated by the pumped storage power station; To improve the overall conversion efficiency of pumped storage power stations; The number of time periods divided for each typical day. The unit time period length; For the first A typical day Power generation output of pumped storage power stations during certain periods; For the first A typical day Pumped storage power station pumping output during certain periods; Constraints on wind and solar energy absorption rates: ; In the formula: The objective function is... The number of time periods divided for each typical day. The unit time period length; For the first A typical day Solar power output during specific time periods; For the first A typical day Wind power output during certain periods; For the first A typical day Theoretical maximum value of photovoltaic power output during the time period; For the first A typical day The theoretical maximum wind power output during a given period.

[0039] Through the aforementioned modeling and constraint system, the embodiments of this invention possess the following technical advantages and application effects: In terms of optimization results, it can achieve both high economic returns and a high proportion of renewable energy utilization, significantly enhancing the matching between the system output curve and load; in terms of model performance, through structured modeling and explicit boundary constraints, it effectively reduces the solver's search time outside the feasible region, improving computational efficiency; in terms of engineering promotion, this model has good regional adaptability and is suitable for the capacity configuration planning of pumped storage power stations under different wind and solar resource structures and load characteristics, providing a supporting tool for large-scale renewable energy grid integration.

[0040] Step 103: Use the piecewise linear approximation method, the bivariate method, and the big M method to transform the optimization model into a biobjective linear programming model.

[0041] In this embodiment of the invention, step 103 transforms the optimization model constructed in step 102 into a bi-objective linear programming model that can be directly solved by mathematical programming tools by using piecewise linear approximation, bivariate method and big M method, so as to overcome the nonlinearity and product terms in the original model and improve the solvability and practical application efficiency of the model.

[0042] Specifically, step 103 includes the following two key parts: 1031, linearization of the quadratic objective function.

[0043] In this embodiment of the invention, in order to make the objective function The sum of squares of load matching deviations is transformed from its original nonlinear form into a linear expression, and a piecewise linear approximation method is proposed.

[0044] The original quadratic objective function is as follows:

[0045] This invention employs a piecewise linear approximation to transform it into a linear objective function. The specific process is as follows: To achieve linearization, this invention introduces the first... A typical day Supply and demand deviation over a period of time .

[0046] Based on statistical analysis of historical power grid supply and demand balance data, the embodiment of this invention obtains the maximum deviation. And add maximum deviation constraint .

[0047] Next, The function is divided into 5 segments, in order to... Linearization, the linearization result is as follows:

[0048] By using the above method, the original nonlinear quadratic objective function is replaced with a piecewise weighted sum form based on a linear expression, thereby transforming the objective function into a linear form: ; Through the above processing, the embodiments of the present invention effectively preserve the changing trend of the original quadratic function, while significantly reducing the mathematical complexity, so that the entire optimization problem can be solved by linear programming algorithm.

[0049] 1032, linearization of product-type nonlinear constraints.

[0050] In this embodiment of the invention, to achieve mutually exclusive operation control of pumping and power generation, the original model introduces the following product constraint: ; This product term constitutes a typical non-convex nonlinear constraint, which is not suitable for solving standard linear programming problems. Therefore, this invention uses a binary variable and the Big M method for linearization: First, define a Boolean decision variable. , Indicates that the pumped storage power station generates electricity. This indicates that the pumped storage power station is pumping water.

[0051] Next, the power constraints for pumped storage power station pumping output and power generation are modified as follows: Pumped storage power generation constraints: ; Pumped storage power station pumping power constraints: .

[0052] By substituting the variables and introducing logical conditions, this invention successfully rewrites the mutually exclusive product terms in the original model into linear inequalities, thereby transforming the optimization model into a standard mixed integer linear programming (MILP) model, which can then be solved efficiently by directly calling mature mathematical optimization software (such as Gurobi).

[0053] The linearization modeling process in this invention not only ensures the solvability and linear structural integrity of the objective function and constraint expression, but also effectively preserves the decision-making logic and engineering characteristics of the original problem, providing a good foundation for subsequent solver applications and multi-objective unified optimization.

[0054] Step 104: Use normalization and weighted summation to transform the bi-objective linear programming model into a single-objective linear programming model.

[0055] In this embodiment of the invention, step 104 uses normalization and weighted summation to transform the bi-objective linear programming model into a single-objective linear programming model, so as to achieve a unified quantitative evaluation and trade-off optimization of revenue and peak-shaving performance, thereby simplifying the model solution process and improving solution efficiency.

[0056] Specifically, after completing the linearization process in step 103, the original biobjective model has been reconstructed into two independent but structurally tractable objective functions, namely: : Annual return of mixed pumped storage (to be maximized); : Sum of squares of system output and load deviation (to be minimized).

[0057] However, directly solving bi-objective optimization problems often faces challenges such as conflicting objectives and difficult trade-offs, especially in engineering deployments where it is difficult to define the optimal solution. Therefore, this invention introduces a normalization and weighted summation mechanism to render the two objective functions dimensionless and merge them into a single objective, constructing a unified evaluation index to achieve flexible optimization under controllable weights.

[0058] In this embodiment of the invention, to ensure the comparability of objective functions with different dimensions and orders of magnitude, the profit and deviation objectives are normalized respectively to obtain the following dimensionless objective function, and the value is controlled between 0 and 1: ; ; in, The lower limit of the mixed pumping return is set based on experience or investment decisions.

[0059] Furthermore, based on normalization, to achieve an adjustable trade-off between the two objectives, this invention introduces non-negative weighting coefficients. and The comprehensive optimization objective function is obtained as follows: .

[0060] in, This indicates the degree of preference for maximizing profits; This indicates the level of importance placed on peak-shaving capacity matching. By adjusting the weighting coefficients, users can flexibly construct different optimization strategies according to actual project needs (such as profit-first or system security-first) to achieve customized configuration goals.

[0061] This invention, through the normalization and weight fusion design in step 104, introduces a unified optimization framework based on bi-objective optimization modeling, possessing the following significant technical advantages: It eliminates the problem of inconsistent objective function dimensions, avoiding optimization bias caused by differences in units or scales; it improves solver convergence efficiency, equivalently transforming multi-objective problems into single-objective problems, reducing variable complexity; and it supports flexible adjustment to meet the needs of various scenarios, allowing users to adjust it according to the nature of the project. and The ratio enables continuous optimization from "maximum benefit" to "maximum peak"; it enhances engineering adaptability and result interpretability, the objective function value is intuitive and clear, and the optimization results have physical understandability and investment decision-making ability.

[0062] Step 105: Use the Gurobi solver to solve the single-objective linear programming model to obtain the optimal configuration capacity and operation scheme of the pumped storage power station.

[0063] In this embodiment of the invention, step 105 uses the Gurobi solver to solve the single-objective linear programming model, thereby obtaining the optimal configuration capacity of the pumped storage power station and its corresponding operation scheme.

[0064] This step, based on the aforementioned construction of the normalized weighted objective function, combines the Python programming environment with the industrial-grade mathematical optimization engine Gurobi to complete the model solving and result extraction process. It is a key step in the implementation of the optimization method proposed in this invention.

[0065] The specific implementation process is as follows: First, in the Python programming environment, based on the single-objective linear programming model constructed in step 104, the modeling interface provided by Gurobi (such as Gurobi-Py or gurobipy) is used to accurately model the objective function, all constraints, upper and lower limits of variables, and variable types (continuous or binary variables) to construct a complete mathematical optimization problem.

[0066] Subsequently, the Gurobi solver is invoked to perform the optimization task. Gurobi supports efficient handling of linear programming (LP) and mixed-integer linear programming (MILP) problems, and exhibits extremely high solution accuracy and speed when faced with the linearized pumped optimization model proposed in this invention.

[0067] In this embodiment of the invention, to meet the differentiated needs of different application scenarios for economy and grid peak-shaving capacity, a weighting coefficient is further set. and The optimal value strategy is used to solve the model in multiple rounds. By adjusting the combination of coefficients within a controllable range, a series of optimization results are generated, forming an optimization "frontier" to describe the multi-objective trade-off between revenue and peak shaving.

[0068] Finally, in this embodiment of the invention, based on the project positioning (such as pursuing the maximum economic return or the strongest system support capability), the solutions in the solution set are comprehensively analyzed and evaluated, and the optimal configuration capacity and its operation strategy are selected as the recommended result for the planning and scheduling of pumped storage power stations.

[0069] Thus, the entire closed-loop process of the pumped storage power station capacity configuration optimization method proposed in this invention is completed. From typical day modeling, dual-objective construction, linearization processing, objective fusion to final solution, a complete set of theoretically rigorous and engineering-implementable optimization method systems is formed.

[0070] In one embodiment of the present invention, as shown in the figure Figure 3 This is a flowchart of the process for solving the capacity configuration optimization model of a pumped storage power station according to an embodiment of the present invention.

[0071] To achieve the above embodiments, the present invention also proposes a pumped storage power station capacity configuration optimization device. Figure 4 This is a schematic diagram of a pumped storage power station capacity configuration optimization device provided in an embodiment of the present invention. Figure 4 As shown, the device includes: Typical day set modeling unit 100 is used to construct a typical day probability distribution set based on K-Means clustering, based on the uncertainties in the capacity configuration optimization process of pumped storage power stations. The system model modeling unit 200 is used to establish an optimization model for the capacity configuration of pumped storage power stations based on the typical daily probability distribution set, with the objective function being the highest mixed pumped storage revenue and the grid peak-shaving demand satisfaction rate. It is based on the operational constraints including grid transmission power constraints, wind-solar-storage power output constraints, capacity constraints, power balance constraints, and wind-solar absorption rate constraints. Linearization unit 300 is used to transform the optimization model into a bi-objective linear programming model using piecewise linear approximation, binary variable method and big M method; Normalization unit 400 is used to transform a bi-objective linear programming model into a single-objective linear programming model using normalization and weighted summation methods; Solver 500 is used to solve a single-objective linear programming model using the Gurobi solver to obtain the optimal configuration capacity and operation scheme of the pumped storage power station.

[0072] In summary, this invention addresses the uncertainties in the capacity configuration optimization process of pumped storage power stations by constructing a typical daily probability distribution set based on K-Means clustering. An optimization model for pumped storage power station capacity configuration is established with the objective functions of maximizing the combined pumped storage revenue and the grid peak-shaving demand satisfaction rate. The optimization model is transformed into a bi-objective linear programming model using piecewise linear approximation, a binary variable method, and the Big M method. The bi-objective linear programming model is further transformed into a single-objective linear programming model using normalization and weighted summation. Finally, the single-objective linear programming model is solved using the Gurobi solver to obtain the optimal configuration capacity and operation scheme for the pumped storage power station. This application can balance economic efficiency and load fluctuations with high computational efficiency, achieving multi-objective synergy and overcoming the shortcomings of traditional methods in terms of dynamic adaptability.

[0073] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0074] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A method for optimizing the capacity configuration of a pumped storage power station, characterized in that, include: Based on the uncertainties in the capacity configuration optimization process of pumped storage power stations, a typical daily probability distribution set based on K-Means clustering is constructed. Based on the typical daily probability distribution set, with the objective function being to maximize the mixed pumped storage revenue and the grid peak-shaving demand satisfaction rate, and according to the operational constraints including grid transmission power constraints, wind-solar-storage power output constraints, capacity constraints, power balance constraints, and wind-solar absorption rate constraints, an optimization model for the capacity configuration of pumped storage power stations is established. The optimization model is transformed into a bi-objective linear programming model using the piecewise linear approximation method, the binary variable method, and the Big M method. The bi-objective linear programming model is transformed into a single-objective linear programming model using normalization and weighted summation. The single-objective linear programming model was solved using the Gurobi solver to obtain the optimal configuration capacity and operation scheme of the pumped storage power station.

2. The method according to claim 1, characterized in that, Based on the uncertainties in the capacity configuration optimization process of pumped storage power stations, a typical daily probability distribution set based on K-Means clustering is constructed, including: Obtaining the theoretical maximum output time series of wind power Photovoltaic theoretical maximum output time series Firepower output time series Power grid load time series Time series of electricity prices for pumped storage power plants Time series of pumped storage power station pumped water electricity prices ; For each time period Constructing multidimensional feature vectors: The multidimensional feature vector contains the first... The theoretical maximum output of wind and solar power, the actual output of thermal power, the grid load, and the value of pumped storage power station pumping and generating electricity during the specified time period; By dividing the historical data of the power grid system into multiple sample sets using a 24-hour period as the unit, the optimal sample set is determined using the elbow method. The value is obtained by performing K-means clustering on the sample set. A typical day and its corresponding probability.

3. The method according to claim 2, characterized in that, The objective function includes: In the formula: Indicates the annual mixed pumped storage yield; The total number of typical days; For the first The probability of a typical day occurring; The number of time periods divided for each typical day. The unit time period length; For the first A typical day Power generation output of pumped storage power stations during certain periods; For the first A typical day Pumped storage power station pumping output during certain periods; For the first A typical day The on-grid electricity price for pumped storage power plants during certain periods; For the first A typical day Pumped storage power station pumping electricity price during specific time periods; It is a capacity-based electricity price; This is the annual investment conversion factor; The annual interest rate; The design service life of the pumped storage power station; Construction cost per unit of installed capacity; This refers to the installed capacity of pumped storage power stations; For other configuration capacity The relevant annual revenue per unit of capacity; For other configuration capacity The relevant annual unit capacity cost; In the formula: Indicates the degree of deviation in output load; The total number of typical days; For the first The probability of a typical day occurring; The number of time periods divided for each typical day. The unit time period length; For the first A typical day System power load during specific time periods; For the first A typical day Power output of thermal power units during certain periods; For the first A typical day Solar power output during specific time periods; For the first A typical day Wind power output during certain periods; For the first A typical day Power generation output of pumped storage power stations during certain periods; For the first A typical day Pumped storage power station output during certain periods.

4. The method according to claim 3, characterized in that, The operational constraints include: Power transmission constraints of the power grid: ； In the formula: For the first A typical day Solar power output during specific time periods; For the first A typical day Wind power output during certain periods; This represents the maximum capacity of the wind-solar hybrid system that the power grid can accept. Wind power output limits and constraints: ； In the formula: For the first A typical day Wind power output during certain periods; For the first A typical day Theoretical maximum wind power output during a given period; Photovoltaic power output constraints: ； In the formula: For the first A typical day Solar power output during specific time periods; For the first A typical day Theoretical maximum value of photovoltaic power output during the time period; Pumped storage power generation constraints: ； In the formula: For the first A typical day Power generation output of pumped storage power stations during certain periods; This refers to the maximum power output of the pumped storage power station, i.e., its installed capacity. Pumped storage power station pumping power constraints: ； In the formula: For the first A typical day Pumped storage power station pumping output during certain periods; This represents the maximum pumping power of the pumped storage power station. Mutual exclusion constraint between pumping and power generation in pumped storage power stations: ； In the formula: For the first A typical day Power generation output of pumped storage power station during specific time periods; where For the first A typical day Pumped storage power station pumping output during certain periods; Pumped storage power station energy storage capacity constraints: ； In the formula: For the first A typical day Power generation output of pumped storage power station during specific time periods; where For the first A typical day Pumped storage power station pumping output during certain periods; To improve the overall conversion efficiency of pumped storage power stations; The unit time period length; The maximum energy storage capacity of the pumped storage power station is determined based on water level requirements and reservoir capacity. Pumped storage power station installed capacity constraints: ； In the formula: This refers to the installed capacity of pumped storage power stations; The site selection for pumped storage power stations is based on the maximum installed capacity. Power balance constraints: ； ； ； In the formula: The electricity consumed by pumped storage power stations for pumping water; This refers to the electricity generated by the pumped storage power station; To improve the overall conversion efficiency of pumped storage power stations; The number of time periods divided for each typical day. The unit time period length; For the first A typical day Power generation output of pumped storage power stations during certain periods; For the first A typical day Pumped storage power station pumping output during certain periods; Constraints on wind and solar energy absorption rates: ； In the formula: The objective function is... The number of time periods divided for each typical day. The unit time period length; For the first A typical day Solar power output during specific time periods; For the first A typical day Wind power output during certain periods; For the first A typical day Theoretical maximum value of photovoltaic power output during the time period; For the first A typical day The theoretical maximum wind power output during a given period.

5. The method according to claim 4, characterized in that, The process of transforming the optimization model into a bi-objective linear programming model using piecewise linear approximation, binary variable method, and Big M method includes: For a quadratic objective function: It is transformed into a linear objective function using a piecewise linear approximation; The specific process is as follows: Introducing the first A typical day Supply and demand deviation over a period of time: ； The maximum deviation was obtained based on historical power grid output and load data. Add maximum deviation constraint ; Will The function is divided into 5 segments, in order to... Linearization, the linearization result is as follows: Therefore, the quadratic objective function is linearized as follows: ； For nonlinear constraints By introducing binary logical variables The Big M method transforms it into a mixed-integer linear programming problem; the specific process is as follows: Introducing binary logical variables , Indicates that the pumped storage power station generates electricity. This indicates pumping by a pumped storage power station; the power constraints for pumping output and generating output of the pumped storage power station are modified as follows: Pumped storage power generation constraints: ； Pumped storage power station pumping power constraints: 。 6. The method according to claim 5, characterized in that, The process of transforming the bi-objective linear programming model into a single-objective linear programming model using normalization and weighted summation includes: Construct normalization formulas to transform the two objective functions into peak shaving satisfaction rate and profit maximization rate, respectively, and control their values ​​between 0 and 1: ； ； in, The lower limit of the mixed pumping return set based on experience or investment decisions; Introduce non-negative weighting coefficients and The comprehensive optimization objective function is obtained as follows: 。 7. The method according to claim 6, characterized in that, The process of solving the single-objective linear programming model using the Gurobi solver to obtain the optimal configuration capacity and operation scheme of the pumped storage power station includes: The single-objective linear programming model is established in the Python environment, and the solution set is obtained by calling the Gurobi solver. The non-negative weighting coefficients are then adjusted. and Different optimization results were obtained. Based on the requirements for economy and stability, the optimal configuration capacity was selected, and the corresponding pumped storage power station operation scheme was obtained.

8. A capacity configuration optimization device for a pumped storage power station, characterized in that, include: Typical day set modeling unit is used to construct a typical day probability distribution set based on K-Means clustering, based on the uncertainties in the capacity configuration optimization process of pumped storage power stations. The system modeling unit is used to establish an optimization model for the capacity configuration of pumped storage power stations based on the typical daily probability distribution set, with the objective function being the highest hybrid pumped storage revenue and the grid peak-shaving demand satisfaction rate. It is also used to establish an optimization model for the capacity configuration of pumped storage power stations based on the operational constraints including grid transmission power constraints, wind-solar-storage power output constraints, capacity constraints, power balance constraints, and wind-solar absorption rate constraints. Linearization unit, used to transform the optimization model into a bi-objective linear programming model using piecewise linear approximation, binary variable method and big M method; A normalization unit is used to transform the bi-objective linear programming model into a single-objective linear programming model using normalization and weighted summation. The solution unit is used to solve the single-objective linear programming model using the Gurobi solver to obtain the optimal configuration capacity and operation scheme of the pumped storage power station.

Images