Pumped storage power station capacity optimization configuration method

Abstract

The invention relates to a pumped storage power station capacity optimization configuration method, and belongs to the technical field of power system planning. The method comprises the following steps: constructing a power grid system construction cost model based on installed capacities corresponding to a pumped storage power station, wind power generation, photovoltaic power generation and thermal power generation respectively and a unit construction cost predicted value of the pumped storage power station; constructing an environmental benefit model based on the output of the pumped storage unit, the wind turbine generator and the photovoltaic unit in each time period; constructing a pumped storage power station capacity optimization configuration objective function based on each model by taking the minimum total cost as an objective; and solving the target function based on system power balance constraint, pumped storage power station constraint, wind curtailment and light curtailment rate constraint, wind power photovoltaic output constraint, thermal power generating unit constraint and wind and light output simulation data to obtain an optimal capacity configuration scheme of the pumped storage power station. According to the method, the capacity configuration of the to-be-constructed pumped storage power station can be reasonably planned, the power grid operation requirement is met, the power grid clean energy utilization rate is improved, and waste of construction cost and energy waste are avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of power system planning, and particularly to a method for optimizing the capacity configuration of a pumped-storage power station. Background Technique

[0002] Pumped-storage power stations have high reliability and good economy. They are currently the most mature large-scale energy storage power sources, which can provide services such as peak shaving and frequency modulation, and can also undertake emergency backup for the power grid to improve the safe and stable operation level of the power grid. Developing pumped-storage power stations plays an important role in accelerating the construction of a new power system, promoting the large-scale and high-proportion development of renewable energy, and achieving the goals of carbon peak and carbon neutrality. However, the construction investment of pumped-storage power stations is large and the construction period is long. Blind construction will bring huge losses.

[0003] At present, domestic and foreign scholars lack research on the prediction of pumped-storage investment costs and capacity configuration planning. They mostly regard pumped-storage power stations as general energy storage, without specifically considering constraints such as reservoir capacity and the impact of building pumped-storage power stations on the utilization rate of clean energy in the power grid. The capacity configuration planning of pumped-storage power stations is not scientific and reasonable enough to avoid waste of resources and maximize the regulating role of pumped-storage power stations. Summary of the Invention

[0004] In view of the above analysis, the present invention aims to disclose a method for optimizing the capacity configuration of a pumped-storage power station. By combining historical data of the construction of pumped-storage power stations to determine multiple key factors affecting the construction cost to predict the unit construction cost, and considering the impact of building pumped-storage power stations on the utilization rate of clean energy in the power grid, the capacity configuration of the to-be-built pumped-storage power station is reasonably planned based on the construction cost and environmental benefits, meeting the operation requirements of the power grid, improving the utilization rate of clean energy in the power grid, and avoiding waste of construction costs and energy waste.

[0005] The present invention provides a method for optimizing the capacity configuration of a pumped-storage power station, which specifically includes the following steps:

[0006] Construct a power grid system construction cost model based on the installed capacities corresponding to the pumped-storage power station, wind power generation, photovoltaic power generation, and thermal power, and the predicted value of the unit construction cost of the pumped-storage power station; construct an environmental benefit model based on the power outputs of the pumped-storage units, wind turbine units, and photovoltaic units in each time period;

[0007] With the goal of minimizing the total cost, construct an objective function for optimizing the capacity configuration of the pumped-storage power station based on the construction cost model and the environmental benefit model;

[0008] Solve the objective function based on the system power balance constraint, pumped-storage power station constraint, wind and light abandonment rate constraint, wind and photovoltaic power output constraint, thermal power unit constraint, and wind and light output simulation data to obtain the optimal capacity configuration plan of the pumped-storage power station.

[0009] Further, based on the PSO-LSSVM model prediction, the predicted value of the unit construction cost of the pumped-storage power station includes:

[0010] Determining multiple key construction factors affecting the unit construction cost of the pumped-storage power station based on the historical data of multiple pumped-storage power stations;

[0011] Training the PSO-LSSVM model based on the historical data of the multiple key construction factors and the unit construction cost of multiple pumped-storage power stations to obtain a trained PSO-LSSVM model;

[0012] Predicting the predicted value of the unit construction cost per kilowatt of the pumped-storage power station by using the trained PSO-LSSVM model based on the corresponding data of the multiple key construction factors of the pumped-storage power station to be constructed.

[0013] Further, the determining multiple key construction factors affecting the unit construction cost of the pumped-storage power station based on the historical data of multiple pumped-storage power stations includes:

[0014] Constructing a historical data matrix based on the unit construction cost per kilowatt of the pumped-storage power station and all construction factors in the historical data of multiple pumped-storage power stations;

[0015] Calculating the degree of association between each of the construction factors and the unit construction cost per kilowatt of the corresponding pumped-storage power station based on the historical data matrix;

[0016] Determining multiple key construction factors based on the degree of association corresponding to each of the construction factors and a preset threshold.

[0017] Further, the construction cost model of the power grid system constructed based on the installed capacities corresponding to the pumped-storage power station, wind power generation, photovoltaic power generation, and thermal power is expressed as:

[0018] F1 = F w + F pv + F ps + F TP

[0019]

[0020] Among them, F1 is the construction cost of the power grid system; F w , F pv , F ps , F tp are the construction costs of wind power generation, photovoltaic power generation, pumped-storage power station, and thermal power respectively; D w , D pv , D tp are the unit construction costs of wind power generation, photovoltaic power generation, and thermal power respectively, Dps is the predicted value of the unit construction cost of the pumped-storage power station; c w , c pv , c ps , c tp are the installed capacities of wind power generation, photovoltaic power generation, pumped-storage power station, and thermal power respectively; O w , O pv , O ps , O tp are the unit kilowatt operation and maintenance costs of wind power generation, photovoltaic power generation, pumped-storage power station, and thermal power respectively; T w , T pv , T ps , T tp are the operation years of wind power generation, photovoltaic power generation, pumped-storage power station, and thermal power respectively; r is the discount rate; N tp is the number of thermal power units; is the output power of thermal power unit i at time t; T a is a dispatching period; τ i is the ramp cost coefficient of the i-th thermal power unit; a i , b i , c i are the coal consumption coefficients; ω is the price of unit coal consumption.

[0021] Furthermore, the environmental benefit model constructed based on the output of pumped-storage units, wind turbine units, and photovoltaic units at each time period is expressed as:

[0022]

[0023] where F2 is the environmental benefit of the power grid system, ε is the coal consumption per unit power generation of the thermal power unit, C em is the emission reduction value of unit pollutants, are the outputs of the i-th wind turbine unit, the j-th photovoltaic power plant, and the k-th pumped-storage unit within the time period t respectively, N w , N pv , N ps are the numbers of wind turbine units, photovoltaic power plants, and pumped-storage units in the system respectively.

[0024] Furthermore, the objective function of the capacity optimization configuration of the pumped-storage power station is expressed as:

[0025] minF = min(F1 - F2);

[0026] where minF represents the objective function with the minimum total cost as the goal.

[0027] Furthermore, the simulated data of wind and light output obtained based on the Copula function and historical wind and photovoltaic output data includes:

[0028] Determine the marginal probability density distribution functions of the wind power and photovoltaic power generation based on the historical data of wind power and photovoltaic power;

[0029] Construct a Copula function that describes the correlation between wind power and photovoltaic power based on the marginal probability density functions;

[0030] Determine the correlation coefficient of the Copula function based on the historical data of wind power and photovoltaic power;

[0031] Generate a set of correlated random values based on the Copula function with the determined correlation coefficient;

[0032] Obtain the corresponding simulated data of the wind and light output by inverting the marginal probability density distribution based on the random values.

[0033] Furthermore, the curtailment rate constraint of wind and light is expressed as:

[0034]

[0035] where E ne is the total power generation of wind power and photovoltaic power within the scheduling period, E ab is the curtailment power of wind and light, and ξ is the preset curtailment rate threshold of wind and light.

[0036] Furthermore, the constraints of the pumped-storage power station include the power constraint of the pumped-storage power station, the operating condition constraint of the pumped-storage power station, the reservoir capacity constraint of the pumped-storage power station, and the reservoir capacity change constraint of the pumped-storage power station, which are respectively expressed as:

[0037]

[0038] V min ≤V t ≤V max ;

[0039] V t+1 =V t +(P t pump η p -P t hydro η h )Δt;

[0040] where and respectively represent the power generation and pumping power of the jth pumped-storage unit at time t; is the maximum power of the pumped-storage unit; V t is the reservoir capacity of the upper reservoir of the pumped-storage power station at time t; V min and V max are respectively the minimum and maximum reservoir capacities of the upper reservoir; P tpump , P t hydro are respectively the total pumping and discharging powers of the pumped - storage power station at time period t; η p , η h are respectively the pumping and discharging efficiencies of the pumped - storage power station.

[0041] Furthermore, solving the objective function based on system power balance constraints, pumped - storage power station constraints, curtailment rate constraints of wind and photovoltaic power, wind and photovoltaic power output constraints, thermal power unit constraints, and wind and photovoltaic output simulation data includes:

[0042] Taking wind and photovoltaic output simulation data, grid system load prediction values, and predicted unit construction cost of the pumped - storage power station as inputs, solving the optimal solutions of the objective function corresponding to each alternative capacity configuration plan and the corresponding objective function values; the optimal solutions include the operating conditions and outputs of pumped - storage units and thermal power units at each time period;

[0043] Selecting the alternative capacity configuration plan corresponding to the minimum objective function value as the optimal capacity configuration plan of the pumped - storage power station.

[0044] The present invention can at least achieve one of the following beneficial effects:

[0045] By considering the construction cost and environmental benefits of the grid system for planning and constructing a pumped - storage power station, and considering the case of meeting the clean energy utilization rate (i.e., curtailment rate constraints of wind and photovoltaic power), scientifically solving to obtain the optimal capacity configuration plan of the pumped - storage power station, meeting the grid operation requirements and improving the clean energy utilization rate of the grid while ensuring the safe and stable operation of the grid system, avoiding waste of construction costs and energy waste.

[0046] By conducting correlation analysis on various construction factors affecting the unit construction cost of the pumped - storage power station, determining multiple key construction factors affecting the construction cost, ensuring the reliability of the prediction of the unit construction cost of the pumped - storage power station, and further predicting the unit construction cost of the pumped - storage power station based on the PSO - LSSVM algorithm, improving the prediction accuracy. By obtaining wind and photovoltaic output simulation data based on the Copula function and historical output data of wind and photovoltaic power, more scientifically describing the output characteristics of wind and photovoltaic power, and ensuring the accuracy of solving the objective function.

[0047] Other features and advantages of the present invention will be described in the subsequent description, and some advantages can be made obvious from the description, or understood by implementing the present invention. The objectives and other advantages of the present invention can be realized and obtained from the content specifically pointed out in the description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The accompanying drawings are only for the purpose of showing specific embodiments and are not considered to be a limitation of the present invention. Throughout the drawings, the same reference numerals represent the same components.

[0049] Figure 1 This is the flowchart of the method of the present invention. Detailed implementation manners

[0050] The following will specifically describe the preferred embodiments of the present invention in conjunction with the accompanying drawings, wherein the accompanying drawings form a part of this application and are used together with the embodiments of the present invention to explain the principles of the present invention, rather than to limit the scope of the present invention.

[0051] An embodiment of the present invention discloses a method for optimizing the capacity configuration of a pumped-storage power station, and its research object is the power grid system in the planning area of the pumped-storage power station to be built. The method of this embodiment specifically includes steps S01 to S03.

[0052] Step S01: Construct a power grid system construction cost model based on the installed capacities corresponding to the pumped-storage power station, wind power generation, photovoltaic power generation, and thermal power, as well as the predicted value of the unit construction cost of the pumped-storage power station; construct an environmental benefit model based on the power outputs of the pumped-storage units, wind turbines, and photovoltaic units in each time period.

[0053] Specifically, the power grid system construction cost model is expressed as:

[0054] F1 = F w + F pv + F ps + F TP

[0055]

[0056] Wherein, F1 is the power grid system construction cost; F w , F pv , F ps , F tp are the construction costs of wind power generation, photovoltaic power generation, pumped-storage power station, and thermal power respectively; D w , D pv , D tp are the unit construction costs of wind power generation, photovoltaic power generation, and thermal power respectively; D ps is the predicted value of the unit construction cost of the pumped-storage power station, which needs to be obtained through prediction; c w , c pv , c ps , c tp are the installed capacities of wind power generation, photovoltaic power generation, pumped-storage power station, and thermal power respectively, where c ps is the variable to be solved; O w , O pv , O ps , Otp The unit kilowatt operation and maintenance costs of wind power generation, photovoltaic power generation, pumped-storage power stations, and thermal power respectively; T w 、T pv 、T ps 、T tp The operation years of wind power generation, photovoltaic power generation, pumped-storage power stations, and thermal power respectively; r is the discount rate; N tp is the number of thermal power units; is the output power of thermal power unit i at time t, which is a process variable for solving; T a is a scheduling period; τ i is the ramp cost coefficient of the i-th thermal power unit; a i 、b i 、c i are the coal consumption coefficients, which are the coefficients of the binary linear equation relationship between coal consumption and output power; ω is the price of unit coal consumption.

[0057] Specifically, the environmental benefit model is expressed as:

[0058]

[0059] Among them, F2 is the environmental benefit of the power grid system; ε is the coal consumption per unit power generation of the thermal power unit; C em is the emission reduction value of unit pollutants; are the outputs of the i-th wind turbine, the j-th photovoltaic power plant, and the k-th pumped-storage unit respectively within time t, all of which are process variables for solving; N w 、N pv 、N ps are the numbers of wind turbines, photovoltaic power plants, and pumped-storage units in the system respectively, where N ps is the variable to be solved.

[0060] Furthermore, in step S01, the predicted value of the unit construction cost of the pumped-storage power station obtained based on the PSO-LSSVM model includes:

[0061] Determine multiple key construction factors affecting the unit construction cost of the pumped-storage power station based on the historical data of multiple pumped-storage power stations;

[0062] Train the PSO-LSSVM model based on the historical data of the multiple key construction factors and the unit construction cost of multiple pumped-storage power stations to obtain a trained PSO-LSSVM model;

[0063] Use the trained PSO-LSSVM model to predict the predicted value of the unit kilowatt construction cost of the pumped-storage power station based on the corresponding data of the multiple key construction factors of the pumped-storage power station to be built.

[0064] Further, multiple key construction factors affecting the unit construction cost of pumped-storage power stations are determined based on the historical data of multiple pumped-storage power stations, including:

[0065] Construct a historical data matrix based on the unit construction cost per kilowatt of pumped-storage power stations and all construction factors in the historical data of multiple pumped-storage power stations;

[0066] Calculate the degree of correlation between each of the construction factors and the unit construction cost per kilowatt of the corresponding pumped-storage power station based on the historical data matrix;

[0067] Determine multiple key construction factors based on the degree of correlation corresponding to each of the construction factors and a preset threshold.

[0068] Specifically, the construction factors considered in the historical data of the present invention include: total installed capacity of the pumped-storage power station, single unit capacity, number of units, turbine specifications (tons / unit), generator specifications (tons / unit), main transformer specifications and voltage levels, inlet valve radius and weight, total floor area, and total man-hours. It should be noted that the total installed capacity of the pumped-storage power station, single unit capacity, number of units, turbine specifications (tons / unit), generator specifications (tons / unit), main transformer specifications and voltage levels are determined at the design stage of the pumped-storage power station and can be obtained from the design budget estimate of the pumped-storage power station; the total floor area and total man-hours can be obtained from engineering record materials or post-evaluation materials after the completion of the pumped-storage power station.

[0069] Further, the historical data matrix is expressed as:

[0070]

[0071] Wherein, X is the historical data matrix; m is the number of pumped-storage power stations; n is the number of the construction factors; the first column x0(k) of the matrix represents the unit construction cost per kilowatt of each pumped-storage power station and is defined as the reference sequence, where k = 1, 2,..., m; the column x i (k) are the construction factors and are defined as the comparison sequences, where i = 1, 2,..., n.

[0072] Further, calculating the degree of correlation between each of the construction factors and the unit construction cost per kilowatt of the corresponding pumped-storage power station based on the historical data matrix includes:

[0073] Normalize the matrix based on the data in the first row of the historical data matrix;

[0074] Calculate the difference between each element of each comparison sequence and the corresponding element of the reference sequence in the normalized matrix;

[0075] Calculate the correlation coefficient corresponding to each element of each comparison sequence based on each difference;

[0076] The correlation degree between each comparison sequence and the reference sequence is calculated based on the correlation coefficients corresponding to all elements of each comparison sequence.

[0077] Furthermore, the following formula is used for normalization processing: The obtained normalized matrix is expressed as:

[0078]

[0079] Furthermore, calculate the difference between each element of each comparison sequence in the normalized matrix and the corresponding element of the reference sequence:

[0080] Δx i (k) = |x i ’(k) - x0’(k)|;

[0081] Furthermore, based on each difference, calculate the correlation coefficient corresponding to each element of each comparison sequence:

[0082]

[0083] where ξ i (k) is the correlation coefficient corresponding to the k-th element in the i-th column of the comparison sequence; is the minimum value of the differences in this column of the comparison sequence; the maximum value of the differences in this column of the comparison sequence; α is the discrimination coefficient, and its value range is (0, 1);

[0084] Furthermore, the calculation formula for the correlation degree between each comparison sequence and the reference sequence is:

[0085]

[0086] where r i is the correlation degree between the i-th column of the comparison sequence and the reference sequence; r i The larger it is, the higher the correlation degree between the unit construction cost of the pumped-storage power station and this construction factor; furthermore, select the factors with a correlation degree above the set threshold as the key construction factors, and an exemplary threshold is 0.5.

[0087] Furthermore, after determining multiple key construction factors affecting the unit construction cost of the pumped-storage power station, select the historical data of the pumped-storage power stations built in China in the past 10 years to train the PSO-LSSVM model. Among them, the optimization function of the LSSVM model is:

[0088]

[0089] In the formula, J is the loss function; φ(x i) is a non - linear mapping function; w is the weight vector; e i is the regression error between the actual value and the predicted value of the model output; γ is the regularization parameter that determines the trade - off between the model complexity and accuracy; b is the bias term.

[0090] Furthermore, the RBF function is used as the kernel function of LSSVM, which is expressed as:

[0091]

[0092] Furthermore, in the process of establishing the LSSVM model, the regularization parameter γ and the kernel function width σ are the key parameters affecting the accurate prediction of the model. If the preset parameter values of (γ, σ) are used and adjusted according to the prediction accuracy of the LSSVM model, the efficiency is low, and often the optimal parameter combination (γ, σ) cannot be found. The particle swarm optimization algorithm (PSO) is a swarm intelligence optimization algorithm that imitates the foraging and cooperation behavior of bird flocks. Its characteristic is to promote the comprehensive optimization of the bird flock in a dynamic and multi - objective environment based on the sharing of information among the bird flock. The core idea of the PSO - LSSVM algorithm is to optimize and obtain the LSSVM model with the optimal parameter combination (γ, σ) based on the initial LSSVM model by constructing the PSO fitness evaluation function.

[0093] Furthermore, the training process of the model includes:

[0094] Pre - process the selected historical data, eliminate the unreasonable data with large fluctuations. On this basis, divide the data into a test set and a training set, and normalize the sample data. The normalization range is [-1, 1], and the normalization formula is as follows:

[0095]

[0096] In the formula, y'0(k) is the normalized pumped - storage power station unit construction cost data of the sample; y0(k) is the original sample data; y max and y min are the maximum and minimum values of y0(k) respectively;

[0097] Set the initialization parameters of the PSO - LSSVM model, including the particle position and velocity range, the particle swarm size, the number of iterations, the inertia weight, and the learning factor, etc.;

[0098] Based on the LSSVM model data, calculate the fitness value of each particle, and update the particle position and velocity. Calculate the optimal velocity and optimal position of the particle under the optimal objective fitness within the maximum number of iterations to obtain the optimal parameter combination;

[0099] Assign the optimal parameter combination (γ, σ) to the LSSVM model, establish the LSSVM model with the optimal parameter combination (γ, σ), and select some pumped-storage power station samples as the test data set for prediction to verify the prediction accuracy of the model. When the prediction accuracy reaches the expectation, end the training; otherwise, repeat the above process until the prediction accuracy reaches the expectation.

[0100] Further, based on the corresponding data of the multiple key construction factors of the pumped-storage power station to be built, use the trained PSO-LSSVM model to predict the predicted value of the construction cost per kilowatt of the pumped-storage power station.

[0101] Step S02: With the goal of minimizing the total cost, construct the objective function for the optimal capacity configuration of the pumped-storage power station based on the construction cost model and the environmental benefit model.

[0102] Specifically, the objective function for the optimal capacity configuration of the pumped-storage power station is expressed as:

[0103] minF = min(F1 - F2);

[0104] Among them, minF represents the objective function with the goal of minimizing the total cost.

[0105] Step S03: Solve the objective function based on the system power balance constraint, the pumped-storage power station constraint, the wind and light curtailment rate constraint, the wind and photovoltaic power output constraint, the thermal power unit constraint, and the wind and light output simulation data to obtain the optimal capacity configuration plan of the pumped-storage power station.

[0106] The system power balance constraint is expressed as:

[0107]

[0108] Among them, is the pumping power of the kth pumped-storage unit in the tth time period; l t is the load of the power grid system in the tth time period.

[0109] The pumped-storage power station constraints include the pumped-storage power station power constraint, the pumped-storage power station operating condition constraint, the pumped-storage power station reservoir capacity constraint, and the pumped-storage power station reservoir capacity change constraint, which are respectively expressed as:

[0110]

[0111] V min ≤V t ≤V max ;

[0112] V t+1 =V t +(P t pump ηp -P t hydro η h )Δt;

[0113] Among them, and respectively represent the power generation and pumping power of the jth pumped-storage unit in the t period; is the maximum power of the pumped-storage unit; V t is the reservoir capacity of the upper reservoir of the pumped-storage power station in the t period; V min and V max are the minimum and maximum reservoir capacities of the upper reservoir respectively; P t pump 、P t hydro are the total pumping and discharging powers of the pumped-storage power station in the t period respectively; η p 、η h are the pumping and discharging efficiencies of the pumped-storage power station respectively.

[0114] The curtailment rate of wind and light is expressed as:

[0115]

[0116] In the formula, E ne is the total power generation of wind power and photovoltaic power during the scheduling period, E ab is the curtailment power of wind and light, and ξ is the preset curtailment rate threshold of wind and light.

[0117] The output constraints of wind power and photovoltaic power are expressed as:

[0118]

[0119] Among them, are the upper limits of the output of wind power and photovoltaic power respectively.

[0120] The constraints of thermal power units are expressed as:

[0121]

[0122] Among them, are the maximum and minimum outputs of the thermal power generation units respectively; are the maximum upward and downward ramping rates of thermal power unit l in the t period respectively, and Δt is the time interval.

[0123] Specifically, in step S03, the simulated data of wind and light output is obtained based on the Copula function and the historical output data of wind power and photovoltaic power, including steps 1-5:

[0124] 1. Determine the marginal probability density distribution functions of wind power and photovoltaic power generation output respectively based on the historical data of wind power and photovoltaic power.

[0125] Specifically, the output of wind power or photovoltaic power is expressed as a random variable P, and after data standardization, it is the random variable P j , then the marginal probability density of wind power or photovoltaic power generation output is:

[0126]

[0127] In the formula, N is the sample size; h is the smoothing coefficient; K(·) is a kernel function that follows the standard normal distribution.

[0128] 2. Construct a Copula function based on the marginal probability density function to describe the correlation between wind power and photovoltaic power.

[0129] Specifically, taking the output of wind power and photovoltaic power in a year in the planning area as samples, and forming random vectors P1 and P2 with the output rates of wind power and photovoltaic power generation, then the Copula function describing the correlation between wind power and photovoltaic power is expressed as:

[0130]

[0131] Among them, θ is the Copula parameter.

[0132] 3. Determine the correlation coefficient θ of the Copula function based on the historical data of wind power and photovoltaic power.

[0133] 4. Generate a set of correlated random values based on the Copula function with the determined correlation coefficient.

[0134] 5. Invert the marginal probability density distribution based on the random values to obtain the corresponding simulated wind-solar output data.

[0135] Specifically, in step S03, taking the simulated wind-solar output data, the predicted value of the power grid system load, and the predicted value of the unit construction cost of the pumped-storage power station as inputs, solve the optimal solution of the objective function and the corresponding objective function value for each corresponding capacity configuration plan; the optimal solution includes the operating conditions and outputs of the pumped-storage units and thermal power units in each time period; among them, the predicted value of the power grid system load can be predicted from the historical load data in the planning area, and can be predicted based on methods such as regression analysis method, wavelet analysis method, time series method, neural network method, etc.; the solution algorithm can be, for example, the sparrow search algorithm;

[0136] Select the alternative capacity configuration plan corresponding to the minimum of the minimum objective function value as the optimal capacity configuration plan of the pumped-storage power station.

[0137] It should be noted that during the solution process, since the capacity configuration plan needs to meet the actual production requirements, the practical approach in line with actual production is to select the optimal one from multiple alternative capacity configuration plans made in the actual plan under the conditions of meeting the minimum planned capacity and the maximum investment capacity.

[0138] Exemplarily, for the current power grid system in a certain province, the thermal power is 29.194 million kilowatts, the hydropower is 32.276 million kilowatts, the wind power is 4.05 million kilowatts, and the photovoltaic power is 4.25 million kilowatts. The standard coal consumption is 300 g / kwh, and the coal price is 1000 yuan / ton. To scientifically select a suitable pumped-storage capacity planning plan, under the conditions of meeting various constraints such as reservoir capacity, investment capacity, and operation constraints, assume that the minimum planned capacity of the pumped-storage power station is 30 MW, and the maximum planned capacity is the upper limit of the investment capacity of the power grid company, which is 300 MW. Each planning plan increases by 30 MW with the equivalent installed capacity of a single unit. A total of 10 decision-making plans are obtained. Substitute the method of this embodiment into the objective function for solution, obtain the optimal solutions corresponding to each plan, and select the alternative capacity configuration plan corresponding to the minimum value of the objective function among the optimal solutions of each plan as the optimal capacity configuration plan of the pumped-storage power station.

[0139] A method for optimizing the capacity configuration of a pumped-storage power station disclosed in this embodiment, by considering the construction cost and environmental benefits of the power grid system for planning and constructing a pumped-storage power station, and considering the situation of meeting the clean energy utilization rate (i.e., the curtailment rate constraint), scientifically solves to obtain the optimal capacity configuration plan of the pumped-storage power station, meets the power grid operation requirements, and improves the clean energy utilization rate of the power grid while ensuring the safe and stable operation of the power grid system, and avoids the waste of construction costs and energy waste.

[0140] By conducting a correlation analysis on various construction factors affecting the unit construction cost of the pumped-storage power station, multiple key construction factors affecting the construction cost are determined, which ensures the reliability of the prediction of the unit construction cost of the pumped-storage power station. Further, by predicting the unit construction cost of the pumped-storage power station based on the PSO-LSSVM algorithm, the prediction accuracy is improved. By obtaining the simulated wind and light output data based on the Copula function and the historical output data of wind power and photovoltaic power, the output characteristics of wind power and photovoltaic power are described more scientifically, which ensures the accuracy of the solution of the objective function.

[0141] The above is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present invention should be covered by the protection scope of the present invention.

Claims

1. A method for optimizing the capacity of a pumped storage power station, characterized in that: The steps include: A grid system construction cost model is constructed based on the installed capacity of pumped storage power stations, wind power generation, photovoltaic power generation and thermal power generation, as well as the predicted unit construction cost of pumped storage power stations; an environmental benefit model is constructed based on the output of pumped storage units, wind turbine units and photovoltaic units in different time periods; Taking the minimum total cost as the goal, constructing the objective function of optimal configuration of pumped storage power station capacity based on the construction cost model and the environmental benefit model; Based on the system power balance constraints, pumped-storage power station constraints, wind and solar power abandonment rate constraints, wind power and photovoltaic output constraints, thermal power unit constraints and wind and solar power output simulation data, the objective function is solved to obtain the optimal capacity configuration plan of the pumped-storage power station.

2. The method for optimizing the capacity configuration of a pumped storage power station according to claim 1, characterized in that: The unit construction cost forecast of the pumped storage power station is obtained based on the PSO-LSSVM model, including: Determine multiple key construction factors that affect the unit construction cost of pumped storage power stations based on historical data of multiple pumped storage power stations; Training a PSO-LSSVM model based on the historical data of the multiple key construction factors and unit construction costs of multiple pumped storage power stations to obtain a trained PSO-LSSVM model; Based on the corresponding data of the multiple key construction factors of the pumped-storage power station to be constructed, the trained PSO-LSSVM model is used to predict the predicted value of the unit kilowatt construction cost of the pumped-storage power station.

3. The method for optimizing the capacity configuration of a pumped storage power station according to claim 2, characterized in that: The multiple key construction factors affecting the unit construction cost of the pumped storage power station determined based on the historical data of the multiple pumped storage power stations include: Construct a historical data matrix based on the unit kilowatt construction cost of pumped storage power stations and all construction factors in the historical data of multiple pumped storage power stations; Calculate the correlation between each of the construction factors and the unit kilowatt construction cost of the corresponding pumped storage power station based on the historical data matrix; A plurality of key construction factors are determined based on the correlation degree corresponding to each of the construction factors and a preset threshold.

4. The method for optimizing the capacity configuration of a pumped storage power station according to claim 3, characterized in that: The cost model for constructing a power grid system based on the installed capacity of pumped storage power stations, wind power generation, photovoltaic power generation and thermal power generation is expressed as follows: F1=F w +F pv +F ps +F TP Among them, F1 is the construction cost of the power grid system; F w 、F pv 、F ps 、F tp are the construction costs of wind power generation, photovoltaic power generation, pumped storage power station and thermal power generation respectively; D w , D pv , D tp are the unit construction costs of wind power generation, photovoltaic power generation and thermal power generation, respectively, ps is the unit construction cost forecast value of the pumped storage power station; c w 、c pv 、c ps 、c tp are the installed capacity of wind power generation, photovoltaic power generation, pumped storage power station and thermal power generation; w , O pv , O ps , O tp are the unit kilowatt operation and maintenance costs of wind power generation, photovoltaic power generation, pumped storage power station and thermal power generation; T w , T pv , T ps , T tp are the operating years of wind power generation, photovoltaic power generation, pumped storage power station and thermal power generation respectively; r is the discount rate; N tp is the number of thermal power units; is the output power of thermal power unit i in period t; T a is a scheduling cycle; τ i is the ramp cost coefficient of the i-th thermal power unit; a i , b i 、c i is the coal consumption coefficient; ω is the price of unit coal consumption.

5. The method for optimizing the capacity configuration of a pumped storage power station according to claim 4, characterized in that: The environmental benefit model constructed based on the output of pumped storage units, wind turbine units, and photovoltaic units in each period is expressed as: Among them, F2 is the environmental benefit of the power grid system, ε is the coal consumption per unit power generation of the thermal power unit, C em is the emission reduction value of a unit pollutant, are the outputs of the i-th wind turbine, the j-th photovoltaic power plant, and the k-th pumped storage unit in time period t, respectively. w 、N pv 、N ps They are the number of wind turbines, photovoltaic power plants and pumped storage units in the system respectively.

6. The method for optimizing the capacity configuration of a pumped storage power station according to claim 5, characterized in that: The objective function of the optimal configuration of the pumped storage power station capacity is expressed as: minF = min(F1 - F2); Among them, minF represents the objective function with the goal of minimizing the total cost.

7. The method for optimizing the capacity configuration of a pumped storage power station according to any one of claims 1 to 6, characterized in that: Based on the Copula function and the historical output data of wind power and photovoltaic power, the wind and solar power output simulation data is obtained, including: Determine the marginal probability density distribution function of wind power and photovoltaic power generation output based on historical data of wind power and photovoltaic power generation; Constructing a Copula function describing the correlation between wind power and photovoltaic power based on the marginal probability density function; Determine the correlation coefficient of the Copula function based on historical data of wind power and photovoltaic power; Generate a set of correlated random values ​​based on a Copula function with a determined correlation coefficient; Based on the random value, the inverse marginal probability density distribution is obtained to obtain the corresponding wind and solar power output simulation data.

8. The method for optimizing the capacity configuration of a pumped storage power station according to claim 6, characterized in that: The wind and solar power abandonment rate constraint is expressed as: In the formula, E ne is the total power generation of wind power and photovoltaic power in the dispatching period, E ab is the amount of wind and solar power abandoned, and ξ is the preset wind and solar power abandonment rate threshold.

9. The method for optimizing the capacity configuration of a pumped storage power station according to claim 6, characterized in that: The pumped storage power station constraints include pumped storage power station power constraints, pumped storage power station operating conditions constraints, pumped storage power station storage capacity constraints and pumped storage power station storage capacity change constraints, which are respectively expressed as: In min ≤V t ≤V max ; 5 t+1 =V t +(P t pump the p -P t hydro the h )Δt; in, and They represent the power generation and pumping power of the jth pumped storage unit in period t respectively; is the maximum power of the pumped storage unit; V t V is the storage capacity of the upper reservoir of the pumped storage power station in period t; min and V max are the minimum and maximum storage capacities of the upper reservoir respectively; P t pump , P t hydro are the total pumping and discharging power of the pumped storage power station during period t; η p and ηh are the pumping and discharging efficiencies of the pumped-storage power station, respectively.

10. The method for optimizing the capacity configuration of a pumped storage power station according to claim 6, characterized in that: The objective function is solved based on system power balance constraints, pumped storage power station constraints, wind and solar power abandonment rate constraints, wind power and photovoltaic output constraints, thermal power unit constraints and wind and solar power output simulation data, including: Taking wind and solar power output simulation data, power grid system load forecast value, and pumped storage power station unit construction cost forecast value as input, the optimal solution of the objective function and the corresponding objective function value corresponding to each alternative capacity configuration scheme are solved; the optimal solution includes the operating conditions and output of the pumped storage unit and the thermal power unit in each time period; The alternative capacity configuration scheme corresponding to the minimum objective function value is selected as the optimal capacity configuration scheme of the pumped storage power station.

Images