Time domain and frequency domain combined multi-type energy storage collaborative configuration method

Abstract

The invention discloses a time-frequency domain combined multi-type energy storage collaborative configuration method, which is applied to the technical field of power system planning and operation. The method comprises the following steps: firstly, reading wind and light load time sequence data, constructing an approximate load duration curve taking week as time resolution, and determining discrete probability distribution of wind power and photovoltaic output; secondly, based on the data, annual or multi-year time domain production simulation is carried out, and the maintenance arrangement, the seasonal hydrogen storage configuration scheme, the energy distribution plan, the wind abandoning rate and the light abandoning rate of the thermal power generating unit are determined. Then, taking a year or multi-year time domain production simulation result as a boundary condition, and constructing a preprocessing module smooth net load curve; and finally, combining the maintenance arrangement of the thermal power generating unit and the smoothed net load curve, adopting a frequency spectrum splitting method, constructing a frequency domain multi-type energy storage collaborative configuration model, and outputting a multi-type energy storage collaborative configuration scheme. Through coupling time domain simulation and frequency domain configuration, thermal power overhaul and hydrogen storage adjustment are considered, and the reliability and economical efficiency of energy storage configuration are improved.

Description

Technical Field

[0001] This invention relates to the technical field of power system planning and operation, and in particular to a time-frequency domain combined multi-type energy storage collaborative configuration method, system, storage medium and computing device. Background Technology

[0002] With the accelerated global energy transition, the penetration rate of renewable energy sources such as wind and solar power in the power system is continuously increasing, bringing significant volatility and uncertainty to power system operation. The integration of a high proportion of renewable energy presents the power system with multiple challenges in supply and demand balance, frequency stability, and energy management. Energy storage systems, as a key flexibility resource, can effectively mitigate the fluctuations in renewable energy and improve system reliability and economics. However, energy storage configuration needs to comprehensively consider the diversity of time scales, investment costs, operational constraints, and the policy environment. Traditional single-time-scale configuration methods often ignore medium- and long-term factors such as unit maintenance and seasonal hydrogen storage allocation, leading to configuration results that deviate from actual needs, resulting in investment waste or operational risks.

[0003] Current research on energy storage configuration mainly employs time-domain or frequency-domain models. Time-domain models configure energy storage by simulating actual operation, but their computational complexity increases significantly with time, making it difficult to account for the impact of seasonal variations in source and load on thermal power units and various types of energy storage on an annual scale. While frequency-domain methods are less constrained by time scales, they typically employ fixed boundary condition assumptions, failing to effectively integrate with time-domain production simulations that incorporate uncertainty handling, and often neglecting the impact of thermal power unit maintenance plans on system available capacity. Furthermore, existing preprocessing methods mostly use simple smoothing or fixed wind and solar curtailment strategies, making it difficult to achieve an optimal balance between ensuring renewable energy consumption and system economics. Summary of the Invention

[0004] The primary objective of this invention is to overcome the shortcomings and deficiencies of the prior art and provide a time-frequency domain combined multi-type energy storage collaborative configuration method to solve the economic and reliability problems of energy storage configuration under a high proportion of renewable energy.

[0005] The second objective of this invention is to provide a multi-type energy storage collaborative configuration system that combines time and frequency domains.

[0006] A third objective of this invention is to provide a storage medium.

[0007] A fourth objective of this invention is to provide a computing device.

[0008] The first objective of this invention is achieved through the following technical solution: a method for coordinated configuration of multiple types of energy storage combining time and frequency domains, comprising the following steps: Step 1): Read the load time series curve and construct an approximate load duration curve with a weekly time resolution; read the wind power and photovoltaic output time series data and construct the discrete probability distribution of wind power and photovoltaic output. Step 2): Based on the approximate load duration curve and discrete probability distribution of wind and solar power output constructed in Step 1), conduct annual or multi-year time-domain production simulations with a weekly time resolution. The objective function is to minimize the sum of thermal power operating costs, energy storage operating costs, energy storage investment costs, wind and solar curtailment penalties, and load shedding penalties. The constraints adopted include power balance constraints, unit maintenance constraints, thermal power unit output constraints, new energy unit constraints, and seasonal hydrogen storage operation constraints. By solving the annual or multi-year time-domain production simulations, the maintenance schedule of thermal power units, seasonal hydrogen storage configuration schemes, energy allocation plans, wind curtailment rates, and solar curtailment rates are determined. Step 3): Based on the load time-series curve in Step 1) and the seasonal hydrogen storage configuration scheme, energy allocation plan, wind curtailment rate, and solar curtailment rate calculated in Step 2), a preprocessing module is constructed as boundary conditions. The preprocessing module takes minimizing the difference between the net load curves as the objective function and uses the net load curve calculation constraints, new energy unit related constraints, and seasonal hydrogen storage energy allocation constraints as constraints to optimize and calculate a smoothed net load curve. The net load curve is given by the net load curve calculation constraints. Step 4): Based on the maintenance schedule of the thermal power units described in Step 2), calculate the overall available capacity and operating cost of the thermal power units on a time-period basis with a preset time resolution, and the smoothed net load curve described in Step 3), construct a frequency domain multi-type energy storage collaborative configuration model. The frequency domain multi-type energy storage collaborative configuration model takes minimizing the sum of thermal power operating cost, energy storage operating cost, and energy storage investment cost as its objective function. The constraints adopted include spectrum distribution coefficient constraints, system power balance constraints, overall thermal power unit operation constraints, energy storage operation constraints, and energy storage configuration constraints. By solving the frequency domain multi-type energy storage collaborative configuration model, a collaborative configuration scheme for multiple types of energy storage is determined.

[0009] Furthermore, in step 1), the approximate load duration curve with a weekly time resolution is constructed using a three-segment approximate load duration curve; the annual or multi-year time-domain production simulation employs a stochastic optimization framework to handle load and new energy uncertainties, specifically including: The formula for constructing the three-segment approximate load duration curve is as follows: ; In the formula, s is the index of the load segment of the approximate load duration curve, and k is the index of the load segment of the load duration curve; This represents the load value corresponding to load segment k in the load duration curve; This represents the load value corresponding to the load segment index s in the approximate load duration curve; , These represent the start and end indices of the load segment of the approximate load duration curve, denoted by load segment index s. Then, based on the time-series mapping information of load time-series curve data and wind power and photovoltaic output time-series data, the wind power and photovoltaic output values ​​corresponding to the load segments of each load duration curve are obtained, and the discrete probability distribution of wind power and photovoltaic output corresponding to the load segments of each approximate load duration curve is calculated. The discrete probability distribution of wind power and photovoltaic output can retain the time-series correlation between load and wind power and photovoltaic output, as well as the randomness of wind power and photovoltaic output. The formulas for calculating the discrete probability distribution of wind and solar power output are as follows: ; In the formula, This indicates that the wind power output value corresponding to the load segment index s of the approximate load duration curve is... The probability of time, which is given by Give; This indicates that the photovoltaic output value corresponding to the load segment index s of the approximate load duration curve is... The probability of time, which is given by Give; , These are the probability values ​​of wind power and photovoltaic power under the load segment index s and load segment index k of the approximate load duration curve, respectively, and they satisfy the normalization condition.

[0010] Furthermore, in step 2), firstly, the objective function for the annual or multi-year time-domain production simulation is determined: ; In the formula, The total investment, operation, and maintenance costs for a year- or multi-year time-domain production simulation; This represents the total number of load segments on the approximate load duration curve. This represents the total number of thermal power units. The duration in hours of the load segment index s of the approximate load duration curve; Let g be the coal consumption function of the thermal power unit. The load segment index of the approximate load duration curve represents the generating power of thermal power unit g; , and These are the curtailed wind power, curtailed solar power, and unloaded power, respectively, for the load segment index s of the approximate load duration curve. , and These are the cost coefficients for wind curtailment, solar curtailment, and load shedding; Investment and operation and maintenance costs for seasonal hydrogen storage; The total investment and operation cost of seasonal hydrogen storage includes investment cost and operation and maintenance cost. Since each energy storage system has a different operating life, the investment cost and operation and maintenance cost of energy storage need to be converted into an equivalent annual value. The constraints for annual or multi-year time-domain production simulations include: (1) Power balance constraints: ; In the formula, The load segment index of the approximate load duration curve represents the generating power of thermal power unit g; , The indexes s of the approximate load duration curve load segments represent the power generation capacity of wind turbines and photovoltaic units, respectively. , These are the charging and discharging power of seasonal hydrogen storage, respectively, based on the load segment index of the approximate load duration curve; The unload power is the index s of the load segment of the approximate load duration curve; This represents the load value corresponding to the load segment index s in the approximate load duration curve; (2) Unit maintenance constraints: Unit maintenance constraints include maintenance frequency constraints, maintenance time constraints, and maintenance continuity constraints; The unit maintenance status variables and thermal power unit start-up and shutdown status of the load segment index s of the approximate load duration curve should be consistent with the weekly index w corresponding to the load segment index s. (3) Output constraints of thermal power units: The output constraints of thermal power units include upper and lower limits for thermal power output. (4) Constraints on new energy units: Constraints on new energy units include upper limits on the output of wind and solar power units, as well as constraints related to wind and solar curtailment. (5) Seasonal constraints on hydrogen storage operation: Seasonal hydrogen storage operation constraints include seasonal hydrogen storage charge and discharge power constraints, mutual exclusion constraints of charge and discharge states, continuous constraints of energy state changes, and constraints of equal energy states at the beginning and end of seasonal hydrogen storage. The aforementioned annual or multi-year time-domain production simulation is a mixed-integer linear programming problem. By solving it through an operations research optimization solver, we can obtain the maintenance schedule of thermal power units, seasonal hydrogen storage capacity configuration and seasonal allocation scheme, as well as the wind curtailment rate and solar curtailment rate thresholds. This will provide important boundary conditions for the smoothing of the net load curve in step 3) and the collaborative configuration model of multiple types of energy storage in step 4), ensuring the coordination and consistency of the collaborative configuration of multiple types of energy storage in the time and frequency domains.

[0011] Furthermore, in step 3), the mathematical expression of the objective function of the preprocessing module is: ; In the formula, , As an auxiliary variable, it represents the maximum and minimum values ​​of the net load curve; The constraints include: (1) Constraints for net load curve calculation: Net load curve calculation constraints include net load curve calculation definition constraints, net load curve maximum value constraints, and net load curve minimum value constraints; (2) Constraints related to new energy units: Constraints related to new energy units include upper limits for wind power output, upper limits for photovoltaic power output, wind curtailment, solar curtailment, and total curtailment rate. (3) Seasonal hydrogen storage energy allocation constraints: Seasonal hydrogen storage energy allocation constraints include seasonal hydrogen storage charge and discharge power constraints, charge and discharge state mutual exclusion constraints, energy state change continuity constraints, and weekly energy transport constraints. The above preprocessing module is a mixed integer linear programming problem. It is solved by an operations research optimization solver to obtain a smoothed net load curve, which provides important boundary conditions for the frequency domain multi-type energy storage collaborative configuration model in step 4).

[0012] Furthermore, in step 4), the objective function of the frequency domain multi-type energy storage collaborative configuration model is: ; In the formula, The overall operating cost of thermal power units; The investment and operating costs of pumped storage hydroelectric power. The investment and operating costs of battery energy storage; The investment and operating costs for flywheel energy storage; Similar to annual or multi-year time-domain production simulations, the total investment and operating cost of an energy storage system includes the initial purchase cost and operation and maintenance costs. Since each energy storage system has a different life cycle, the total investment and operating cost of energy storage needs to be converted into an equivalent annual value. During the coordinated operation of thermal power and energy storage, energy storage devices will generate energy losses during the charging and discharging process. The energy storage power curve obtained by discrete Fourier transform analysis only contains periodic components, and its power integral value is zero in the complete cycle. However, due to energy losses in actual operation, the energy state of energy storage may shift at the end of the cycle. In order to maintain the consistency between the initial and final energy state of energy storage, thermal power units need to provide additional power compensation to make up for the energy losses generated by energy storage devices during the charging and discharging process. The resulting power generation cost is defined as the compensation cost. Operating cost of thermal power units C g This includes basic output costs and compensation costs; Constraints: (1) Spectral distribution coefficient constraint: The net load curve is decomposed into components of different frequencies using Fourier transform technology. Spectral clustering is performed according to the energy magnitude of each frequency, and allocation is carried out using the spectral distribution coefficient. The following frequency-based division of labor is implemented: for low-frequency components with high energy proportions and slow fluctuations, both thermal power units and pumped storage with strong regulation capabilities share the load, with the DC component (zero frequency component) entirely handled by thermal power units; for medium-frequency fluctuations with moderate variation characteristics, pumped storage with excellent charging and discharging efficiency and high energy density, and battery energy storage are used for balancing; for high-frequency fluctuations with drastic changes, flywheel energy storage with fast response speed and outstanding cycle life is responsible for smoothing them. This frequency-based division of labor fully leverages the technical and economic advantages of different types of energy storage equipment. (2) Overall operational constraints of thermal power units: The overall operating constraints of thermal power units include upper and lower limits of output, ramp rate limit and actual available capacity constraint. The overall actual available capacity of thermal power units needs to be dynamically adjusted according to the maintenance schedule of thermal power units determined in step 2). That is, the available capacity of the corresponding unit is zero during the maintenance period and the rated capacity is used during non-maintenance periods. (3) Energy storage operation constraints: Energy storage operation constraints include energy storage charging and discharging power constraints and energy storage energy state change constraints; (4) Constraints on energy storage configuration: Energy storage configuration constraints include energy storage power configuration constraints and energy storage capacity configuration constraints; The frequency domain multi-type energy storage collaborative configuration model transforms the original mixed integer programming optimization problem involving the charging and discharging state variables of multiple types of energy storage into a linear programming problem by linearizing the constraints. This allows the model to be solved by an operations research optimization solver, yielding the collaborative configuration results of multiple types of energy storage.

[0013] The second objective of this invention is achieved through the following technical solution: a time-frequency domain combined multi-type energy storage collaborative configuration system, used to implement the above-mentioned time-frequency domain combined multi-type energy storage collaborative configuration method, comprising: The data acquisition module is responsible for reading load time-series curve data and constructing an approximate load duration curve with a weekly time resolution; it also reads wind power and solar power output time-series data and constructs discrete probability distributions for wind power and solar power output. The annual or multi-year time-domain production simulation module is used to perform annual or multi-year time-domain production simulations with a weekly time resolution. It comprehensively considers the operating costs of thermal power plants, energy storage operating costs, energy storage investment costs, wind and solar curtailment penalties, and load shedding penalties. Under the conditions of satisfying power balance constraints, unit maintenance constraints, thermal power unit output constraints, new energy unit constraints, and seasonal hydrogen storage operation constraints, it outputs the maintenance schedule of thermal power units, seasonal hydrogen storage configuration schemes and energy allocation plans, as well as wind curtailment rates and solar curtailment rates. The preprocessing module, based on the seasonal hydrogen storage configuration scheme, energy allocation plan, and wind and solar curtailment rates output by the annual or multi-year time-domain production simulation module, optimizes and smooths the original net load curve with the goal of minimizing the difference in the net load curve, and obtains the smoothed net load curve; based on the maintenance schedule of thermal power units, it calculates the overall available capacity and operating cost of thermal power units in time periods with a preset time resolution. The frequency domain multi-type energy storage collaborative configuration module, based on the smoothed net load curve output by the preprocessing module and the overall available capacity and operating cost of the thermal power unit, constructs a frequency domain multi-type energy storage collaborative configuration model with a preset time resolution to determine the collaborative configuration scheme of multiple types of energy storage.

[0014] The third objective of this invention is achieved through the following technical solution: a storage medium storing a program, which, when executed by a processor, implements the above-mentioned time-frequency domain combined multi-type energy storage collaborative configuration method.

[0015] The fourth objective of this invention is achieved through the following technical solution: a computing device, including a processor and a memory for storing processor-executable programs, wherein when the processor executes the program stored in the memory, it implements the above-mentioned time-frequency domain combined multi-type energy storage collaborative configuration method.

[0016] Compared with the prior art, the present invention has the following advantages and beneficial effects: 1. This invention couples annual or multi-year time-domain production simulations with a weekly time resolution with the coordinated configuration of multiple types of energy storage in the frequency domain. It then uses the maintenance schedules of thermal power units, seasonal hydrogen storage configuration schemes, and energy allocation plans as boundary conditions to propagate to subsequent optimization steps. This design effectively solves the problem of connecting decisions at different time scales, enabling the coordinated configuration scheme of multiple types of energy storage to fully consider the actual system operation constraints, significantly improving the reliability and practicality of the planning scheme.

[0017] 2. This invention uses minimizing the net load curve difference as the objective function to collaboratively optimize the allocation of wind and solar curtailment and seasonal hydrogen storage energy. While meeting the wind and solar curtailment rate thresholds, it effectively smooths the net load curve, thereby directly reducing the power and capacity of various types of energy storage required to balance power fluctuations in the subsequent frequency domain multi-type energy storage collaborative configuration model, thus improving the economics of the solution from the source.

[0018] 3. This invention constructs a frequency-domain multi-type energy storage collaborative configuration model, decomposes the smoothed net load curve, and allocates it to the overall thermal power unit and multiple types of energy storage according to the spectral distribution coefficient, while linearizing the energy storage configuration constraints. This method transforms a complex problem into a linear programming problem, and uses operations research and optimization solvers such as Gurobi and CPLEX for efficient solution, significantly reducing computational complexity and making it suitable for the optimization planning of large-scale power systems.

[0019] 4. This invention has good adaptability and scalability, taking into account multiple factors such as energy storage investment cost, operating cost, wind curtailment, solar curtailment penalty and load shedding penalty, and can be dynamically adjusted according to actual operating conditions. It is suitable for multi-type energy storage collaborative configuration with a high proportion of renewable energy access to the power system.

[0020] In summary, this invention analyzes historical load and renewable energy time-series data to construct a weekly-resolution approximate load continuity curve, forming discrete probability distributions of wind and solar power output and identifying typical operating scenario characteristics. Based on this, annual or multi-year time-domain production simulations with a weekly time resolution are conducted to optimize and determine thermal power unit maintenance schedules, seasonal hydrogen storage configuration schemes, energy allocation plans, and wind and solar curtailment rate thresholds. Furthermore, using the annual or multi-year time-domain production simulation results as boundary conditions, a frequency-domain multi-type energy storage collaborative configuration model is employed to achieve collaborative planning of multiple energy storage types. By coupling time-domain production simulation with frequency-domain multi-type energy storage collaborative configuration, and comprehensively considering factors such as thermal power unit maintenance capacity and seasonal hydrogen storage energy allocation plans, the economic efficiency and reliability of energy storage configuration are improved. This approach is suitable for high-proportion renewable energy systems and provides key technical support for the collaborative planning of multi-type energy storage in new power systems. Attached Figure Description

[0021] Figure 1 This is a flowchart of the overall process of the method of the present invention.

[0022] Figure 2 A schematic diagram for constructing an approximate load duration curve.

[0023] Figure 3 The diagram shows the configuration results of the frequency domain multi-type energy storage collaborative configuration model in the example.

[0024] Figure 4 The above is a stacked diagram of the spectral distribution coefficients for an example.

[0025] Figure 5 This is a system architecture diagram of the present invention. Detailed Implementation

[0026] The present invention will be further described in detail below with reference to the embodiments and accompanying drawings, but the embodiments of the present invention are not limited thereto.

[0027] Example 1 This embodiment discloses a time-frequency domain combined multi-type energy storage collaborative configuration method, such as... Figure 1 As shown, the specific details are as follows: Step 1): Read the load time-series curve data and construct an approximate load duration curve with a weekly time resolution; read the wind power and solar power output time-series data and construct the discrete probability distributions of wind power and solar power output, as follows: An approximate load duration curve with a weekly time resolution is constructed using a three-segment approximate load duration curve. The annual or multi-year time-domain production simulation employs a stochastic optimization framework to handle load and new energy uncertainties, specifically including: The formula for constructing the three-segment approximate load duration curve is as follows: (1); In the formula, s is the index of the load segment of the approximate load duration curve, and k is the index of the load segment of the load duration curve; This represents the load value corresponding to load segment k in the load duration curve; This represents the load value corresponding to the load segment index s in the approximate load duration curve; , These represent the start and end indices of the load segment of the approximate load duration curve, denoted by load segment index s. Then, based on the time-series mapping information of load time-series curve data and wind power and photovoltaic output time-series data, the wind power and photovoltaic output values ​​corresponding to the load segments of each load duration curve are obtained, and the discrete probability distribution of wind power and photovoltaic output corresponding to the load segments of each approximate load duration curve is calculated. The discrete probability distribution of wind power and photovoltaic output can retain the time-series correlation between load and wind power and photovoltaic output, as well as the randomness of wind power and photovoltaic output. The formulas for calculating the discrete probability distribution of wind and solar power output are as follows: (2); In the formula, This indicates that the wind power output value corresponding to the load segment index s of the approximate load duration curve is... The probability of time, which is given by Give; This indicates that the photovoltaic output value corresponding to the load segment index s of the approximate load duration curve is... The probability of time, which is given by Give; , These are the probability values ​​of wind power and photovoltaic power under the load segment index s and load segment index k of the approximate load duration curve, respectively, and they satisfy the normalization condition.

[0028] Step 2): Based on the approximate load duration curve and discrete probability distributions of wind and solar power output constructed in Step 1), conduct annual or multi-year time-domain production simulations with a weekly time resolution. The objective function is to minimize the sum of thermal power operating costs, energy storage operating costs, energy storage investment costs, wind and solar curtailment penalties, and load shedding penalties. The constraints used include power balance constraints, unit maintenance constraints, thermal power unit output constraints, new energy unit constraints, and seasonal hydrogen storage operation constraints. The annual or multi-year time-domain production simulations are solved using operations research and optimization solvers such as Gurobi and CPLEX to determine the maintenance schedule for thermal power units, seasonal hydrogen storage configuration schemes, energy allocation plans, and wind and solar curtailment rates. Details are as follows: First, determine the objective function for annual or multi-year time-domain production simulation: (3); In the formula, The total investment, operation, and maintenance costs for a year- or multi-year time-domain production simulation; This represents the total number of load segments on the approximate load duration curve. This represents the total number of thermal power units. The duration in hours of the load segment index s of the approximate load duration curve; Let g be the coal consumption function of the thermal power unit. The load segment index of the approximate load duration curve represents the generating power of thermal power unit g; , and These are the curtailed wind power, curtailed solar power, and unloaded power, respectively, for the load segment index s of the approximate load duration curve. , and These are the cost coefficients for wind curtailment, solar curtailment, and load shedding; Investment and operation and maintenance costs for seasonal hydrogen storage; The total investment and operating cost of seasonal hydrogen storage includes investment costs and operation and maintenance costs. Since different energy storage systems have different operational lifespans, the investment costs and operation and maintenance costs need to be converted to annual values, expressed as follows: (4); (5); In the formula, The annual interest rate; The service life of seasonal hydrogen storage; , These represent the unit capacity cost and unit power cost of seasonal hydrogen storage, respectively. , These are the rated capacity and rated power of seasonal hydrogen storage, respectively; Annual operating and maintenance costs for seasonal hydrogen storage; , These are the operation and maintenance costs per unit power and per unit capacity of seasonal hydrogen storage, respectively. The constraints for annual or multi-year time-domain production simulations include: (1) Power balance constraints: (6); In the formula, The load segment index of the approximate load duration curve represents the generating power of thermal power unit g; , The indexes s of the approximate load duration curve load segments represent the power generation capacity of wind turbines and photovoltaic units, respectively. , These are the charging and discharging power of seasonal hydrogen storage, respectively, based on the load segment index of the approximate load duration curve; (2) Unit maintenance constraints: Unit maintenance constraints include maintenance frequency constraints, maintenance time constraints, and maintenance continuity constraints, which are mathematically expressed as follows: (7); In the formula, W and w represent the total number of weeks and the week index of the annual or multi-year time-domain production simulation, respectively. Under the three-segment approximate load duration curve, each week index corresponds to three load segment indices. As an intermediate variable; This represents the maintenance status transition variable for thermal power unit g at time w. It is set to 1 when maintenance is initiated at time w, and 0 otherwise. This indicates the required number of maintenance operations for thermal power unit g during a year or multiple years of time-domain production simulation. For the maintenance status variable of Zhouw thermal power unit, take 1 when it is under maintenance, otherwise take 0; For thermal power unit g, the maintenance time is; This represents the start-up and shutdown status of the Zhouw thermal power unit. If it is in the start-up state, use 1; otherwise, use 0. The unit maintenance status variables and thermal power unit start-up and shutdown status of the load segment index s of the approximate load duration curve should be consistent with the weekly index w corresponding to the load segment index s. The mathematical expression is: (8); In the formula, , For load segment index The unit maintenance status variables and the start-up and shutdown status variables of the thermal power unit; (3) Output constraints of thermal power units: The output constraints of thermal power units include upper and lower limits for thermal power output, and the mathematical expressions are as follows: (9); In the formula, , These are the minimum and maximum technical outputs of thermal power unit g, respectively. (4) Constraints on new energy units: Constraints on new energy units include upper output limits for wind and solar power units, as well as constraints related to wind and solar curtailment. The mathematical expression is: (10); (11); In the formula, , The wind power and photovoltaic power generation of the approximate load duration curve load segment index s are respectively obtained by sampling from the discrete probability distribution of wind power and photovoltaic power output described in step 2). (5) Seasonal constraints on hydrogen storage operation: Seasonal hydrogen storage operation constraints include seasonal hydrogen storage charge / discharge power constraints, mutual exclusion constraints of charge / discharge states, continuity constraints of energy state changes, and constraints on the equality of the initial and final energy states of seasonal hydrogen storage. The mathematical expressions are as follows: (12); In the formula, , These represent the charging and discharging states of seasonal hydrogen storage at index s of the load segment of the approximate load duration curve, respectively. , These represent the maximum and minimum energy states of seasonal hydrogen storage, respectively. This indicates the energy state of seasonal hydrogen storage at index s of the approximate load duration curve. This indicates the energy state of seasonal hydrogen storage at index s-1 of the approximate load duration curve. Indicates the charge / discharge efficiency of seasonal hydrogen storage; The aforementioned annual or multi-year time-domain production simulation is a mixed-integer linear programming problem, which can be solved by operations research and optimization solvers such as Gurobi and CPLEX. This yields the maintenance schedule of thermal power units, seasonal hydrogen storage capacity configuration and seasonal allocation scheme, as well as wind curtailment rate and solar curtailment rate thresholds. These outputs will provide important boundary conditions for the smoothing of the net load curve in step 3) and the collaborative configuration model of multiple types of energy storage in step 4), ensuring the coordination and consistency of the collaborative configuration of multiple types of energy storage in the time and frequency domains.

[0029] Step 3): Based on the load time-series curve from Step 1) and the seasonal hydrogen storage configuration scheme, energy allocation plan, and wind and solar curtailment rates calculated in Step 2), a preprocessing module is constructed using these as boundary conditions. The preprocessing module uses minimizing the net load curve difference as its objective function, and net load curve calculation constraints, new energy unit-related constraints, and seasonal hydrogen storage energy allocation constraints as constraints. It optimizes the calculation using operations research solvers such as Gurobi and CPLEX to obtain a smoothed net load curve, which is given by the net load curve calculation constraints. Specifically, the mathematical expression of the objective function of the preprocessing module is: (13); In the formula, , As an auxiliary variable, it represents the maximum and minimum values ​​of the net load curve; Specifically, the constraints include: (1) Constraints for net load curve calculation: The constraints for net load curve calculation include the net load curve calculation definition constraints, the maximum net load curve constraints, and the minimum net load curve constraints, and their mathematical expressions are as follows: (14); In the formula, t represents the time period index of the net load curve; , These represent the load values ​​for time period t in the net load curve and the load time series curve, respectively. , These represent the power generation of the wind turbine and the photovoltaic unit during time period t, respectively. , These represent the charging power and discharging power of seasonal hydrogen storage during time period t, respectively. (2) Constraints related to new energy units: The constraints related to new energy units include upper limits for wind power output, upper limits for photovoltaic power output, wind curtailment, photovoltaic curtailment, and total curtailment rate. The mathematical expressions are as follows: (15); In the formula, α is the curtailment rate of wind and solar power determined by annual or multi-year time-domain production simulation, and represents the curtailment rate threshold for minimizing the difference in net load curves in step 3); T represents the time period index of the load time-series curve and the net load time-series curve; , These represent the time-series data for wind power and solar power output, respectively. (3) Seasonal hydrogen storage energy allocation constraints: Seasonal hydrogen storage energy allocation constraints include seasonal hydrogen storage charge / discharge power constraints, charge / discharge state mutual exclusion constraints, energy state change continuity constraints, and weekly energy transport constraints, the mathematical expressions of which are: (16); In the formula, Indicates the time resolution of the net load curve; , These represent the charging and discharging states of seasonal hydrogen storage during time period t, respectively. , These represent the maximum and minimum energy states of seasonal hydrogen storage, respectively. This indicates the energy state of seasonally stored hydrogen during time period t; This indicates the energy state of seasonally stored hydrogen during period t-1; This indicates that when the time resolution of the net load curve is... Number of time slots per week; This represents the weekly energy state of seasonal hydrogen storage in annual or multi-year time-domain production simulations, serving as a boundary condition. The above preprocessing module is a mixed integer linear programming problem, which can be solved by operations research and optimization solvers such as Gurobi and CPLEX to obtain a smoothed net load curve, providing important boundary conditions for the frequency domain multi-type energy storage collaborative configuration model in step 4).

[0030] Step 4): Based on the maintenance schedule of the thermal power units described in Step 2), calculate the overall available capacity and operating cost of the thermal power units on a time-period basis with a preset time resolution, and the smoothed net load curve described in Step 3), construct a frequency domain multi-type energy storage collaborative configuration model. The objective function of this model is to minimize the sum of thermal power operating costs, energy storage operating costs, and energy storage investment costs. The constraints include spectral distribution coefficient constraints, system power balance constraints, overall thermal power unit operating constraints, energy storage operating constraints, and energy storage configuration constraints. Solve the frequency domain multi-type energy storage collaborative configuration model using operations research solvers such as Gurobi and CPLEX to determine the collaborative configuration scheme for multiple types of energy storage. Specifically, the objective function of the frequency domain multi-type energy storage collaborative configuration model is: (17); In the formula, The overall operating cost of thermal power units; The investment and operating costs of pumped storage hydroelectric power. The investment and operating costs of battery energy storage; The investment and operating costs for flywheel energy storage; Similar to annual or multi-year time-domain production simulations, the total investment and operating cost of an energy storage system includes initial purchase costs and operation and maintenance costs. Since each energy storage system has a different lifespan, the total investment and operating cost needs to be converted to an equivalent annual value, expressed as: (18); (19); In the formula, , , These refer to the service life of pumped hydro storage, battery energy storage, and flywheel energy storage, respectively. , , These are the unit capacity costs for pumped hydro storage, battery energy storage, and flywheel energy storage, respectively. , , These are the unit power costs for pumped hydro storage, battery storage, and flywheel storage, respectively. , , These are the rated capacities of pumped hydro storage, battery storage, and flywheel storage, respectively. , , These are the rated power of pumped hydro storage, battery energy storage, and flywheel energy storage, respectively. , , These are the annual operation and maintenance costs for pumped hydro storage, battery energy storage, and flywheel energy storage, respectively. , , These are the operation and maintenance costs per unit capacity for pumped hydro storage, battery energy storage, and flywheel energy storage, respectively. , , These are the operation and maintenance costs per unit power for pumped hydro storage, battery energy storage, and flywheel energy storage, respectively. During the coordinated operation of thermal power and energy storage, energy storage devices will generate energy losses during the charging and discharging process. The energy storage power curve obtained by discrete Fourier transform analysis only contains periodic components, and its power integral value is zero in the complete cycle. However, due to energy losses in actual operation, the energy state of energy storage may shift at the end of the cycle. In order to maintain the consistency between the initial and final energy state of energy storage, thermal power units need to provide additional power compensation to make up for the energy losses generated by energy storage devices during the charging and discharging process. The resulting power generation cost is defined as the compensation cost. Operating cost of thermal power units C g Including basic output cost and compensation cost, the mathematical expression is: (20); (twenty one); In the formula, The overall coal consumption function of the thermal power unit is obtained by weighting the thermal power units that have not undergone maintenance; This variable represents the maintenance status of thermal power unit g during time period t. It is 1 if the unit is under maintenance, and 0 otherwise. This value is given by the maintenance schedule of thermal power unit determined by the annual or multi-year time-domain production simulation results in step 2). , These refer to the overall basic output and compensation output of the thermal power unit, respectively. , , These represent the overall compensation power of the generator unit for pumped storage, battery storage, and flywheel storage during time period t, respectively, satisfying the following: (twenty two); In the formula, , , These represent the charging power of pumped hydro storage, battery storage, and flywheel storage, respectively. , , These represent the discharge power of pumped hydro storage, battery storage, and flywheel storage, respectively. , , These represent the charge and discharge efficiencies of pumped hydro storage, battery storage, and flywheel storage, respectively. Constraints: (1) Spectral distribution coefficient constraint: The net load curve is decomposed into components of different frequencies using Fourier transform technology. Spectral clustering is performed according to the energy magnitude of each frequency, and allocation is carried out using the spectral distribution coefficient. The following frequency-based division of labor is implemented: For low-frequency components with high energy proportions and slow fluctuations, both thermal power units and pumped storage with strong regulation capabilities share the load, with the DC component (zero frequency component) entirely handled by thermal power units; for medium-frequency fluctuation components with moderate variation characteristics, pumped storage with excellent charging and discharging efficiency and high energy density, and battery energy storage are used for balancing; for high-frequency fluctuation components with drastic changes, flywheel energy storage with fast response speed and outstanding cycle life is responsible for smoothing them. Through this frequency-based division of labor, the technical and economic advantages of different types of energy storage equipment can be fully utilized. The mathematical expression is: (twenty three); In the formula, n is the index of the spectral component group of the cluster; f is the frequency value; , , , These represent the spectral allocation coefficients of the overall thermal power unit, pumped storage, battery energy storage, and flywheel energy storage in the spectral component group n, respectively. This is the highest frequency that pumped storage can handle; , These are the lowest and highest frequencies that battery energy storage can handle, respectively. This is the lowest frequency that flywheel energy storage can handle; (2) Overall operational constraints of thermal power units: The overall operational constraints of thermal power units include upper and lower output limits, ramp rate limits, and actual available capacity constraints. The overall actual available capacity of thermal power units needs to be dynamically adjusted according to the maintenance schedule determined in step 2). That is, the available capacity of the corresponding unit is zero during maintenance and the rated capacity is used during non-maintenance periods. The mathematical expression is: (twenty four); (25); In the formula, , These are the lower and upper limits of the overall output of the thermal power unit in time period t, taking into account the unit's maintenance. and These represent the lower and upper limits of the overall ramp power of the thermal power unit during time period t, respectively. This represents the total power generation of the thermal power unit during time period t. This represents the overall power generation of the thermal power unit during time period t-1; Let be the overall maintenance capacity of the thermal power unit during time period t, and let be the sum of the maintenance capacities of the thermal power unit in each time period. Its mathematical expression is: (26); (3) Energy storage operation constraints: Energy storage operation constraints include energy storage charging and discharging power constraints and energy storage energy state change constraints, the mathematical expressions of which are: (27); In the formula, , , These represent the net charging and discharging power of pumped hydro storage, battery storage, and flywheel storage during time period t, respectively. , , The energy states of pumped hydro storage, battery energy storage, and flywheel energy storage are respectively defined for time period t. , , These represent the energy states of pumped hydro storage, battery energy storage, and flywheel energy storage during time period t-1. (4) Constraints on energy storage configuration: Energy storage configuration constraints include energy storage power configuration constraints and energy storage capacity configuration constraints; The linearized expression for the energy storage power configuration constraint is: (28); The linearized expression for the energy storage capacity configuration constraint is: (29); (30); In the formula, , These represent the maximum and minimum energy states of pumped storage, respectively. , These are the maximum and minimum SOC values ​​for pumped storage, respectively. , These represent the maximum and minimum states of energy for battery storage, respectively. , These represent the maximum and minimum SOC values ​​for battery energy storage, respectively. , These represent the maximum and minimum energy states of the flywheel energy storage, respectively. , These represent the maximum and minimum SOC values ​​for flywheel energy storage, respectively. The frequency domain multi-type energy storage collaborative configuration model transforms the original mixed integer programming optimization problem involving the charging and discharging state variables of multiple types of energy storage into a linear programming problem by linearizing the constraints. This allows for efficient solution using operations research optimization solvers such as Gurobi and CPLEX, yielding collaborative configuration results for multiple types of energy storage.

[0031] Below, we verify the application effect of our method in a high-proportion renewable energy scenario based on a case study of a provincial power grid improvement project. In this case study, the total installed capacity of thermal power units is approximately 50,600 MW, the peak load is approximately 58,000 MW, and the maximum output of wind power and photovoltaic units is approximately 17,500 MW and 16,000 MW, respectively. We assume that no energy storage is currently configured in the system, but consider the configuration of pumped hydro storage, battery storage, flywheel storage, and seasonal hydrogen storage. Parameters of some thermal power units and technical and economic parameters of various energy storage types are shown in Tables 1 and 2, respectively.

[0032] Table 1. Partial generator set data

[0033] Table 2 Technical and economic parameters for some energy storage configurations

[0034] Taking annual energy storage configuration as an example, the time resolution is set to 15 minutes (35,040 time periods throughout the year), and the annual interest rate is 5%. Regarding spectrum allocation, the maximum number of adaptive spectrum clusters is 60, and the number of frequency points within a group does not exceed 2% of the total number of frequency points. In the spectrum distribution coefficient constraints, the following frequencies are set: pumped storage allocation cycles greater than 12 hours; battery energy storage allocation cycles greater than 1 hour to less than 24 hours; and flywheel energy storage allocation cycles less than 2 hours. Thermal power units can participate in the regulation of the DC component and all cycle components greater than 1 hour within the allowable ramp rate range. Frequencies between 24 and 12 hours can be shared by multiple types of energy storage.

[0035] According to step 1) of the method of the present invention, based on the annual load and new energy time series data, an approximate load duration curve with a weekly time resolution and the discrete probability distribution of wind power and photovoltaic output are constructed according to equations (1) and (2) to obtain the typical operating characteristics of 52 weeks. Figure 2 The construction effect of the approximate load duration curve is shown using week 1 as an example. (Step 4) of the method according to the present invention. Figure 3 The configuration of the frequency domain multi-type energy storage collaborative configuration model of the embodiment is shown. Figure 4 The distribution ratio of the spectral distribution coefficients for each type of energy storage in the embodiments is shown. Based on the combined results of seasonal hydrogen storage configuration from annual or multi-year time-domain production simulations, the time-frequency domain combined results of multi-type energy storage collaborative configuration are shown in Table 3.

[0036] Table 3 Results of Coordinated Configuration of Multiple Energy Storage Types

[0037] In summary, this invention demonstrates that the proposed model utilizes reasonable boundary conditions provided by annual or multi-year time-domain production simulations, effectively smooths the net load curve by minimizing the net load curve difference, and employs spectrum splitting technology to achieve coordinated configuration of multiple types of energy storage in the frequency domain, providing a reliable technical solution for energy storage planning in high-proportion renewable energy systems.

[0038] Example 2 This embodiment discloses a time-frequency domain combined multi-type energy storage collaborative configuration system, used to implement the time-frequency domain combined multi-type energy storage collaborative configuration method described in Embodiment 1, such as... Figure 5 As shown, it includes the following functional modules: The data acquisition module is responsible for reading load time-series curve data and constructing an approximate load duration curve with a weekly time resolution; it also reads wind power and solar power output time-series data and constructs discrete probability distributions for wind power and solar power output. The annual or multi-year time-domain production simulation module is used to perform annual or multi-year time-domain production simulations with a weekly time resolution. It comprehensively considers the operating costs of thermal power plants, energy storage operating costs, energy storage investment costs, wind and solar curtailment penalties, and load shedding penalties. Under the conditions of satisfying power balance constraints, unit maintenance constraints, thermal power unit output constraints, new energy unit constraints, and seasonal hydrogen storage operation constraints, it outputs the maintenance schedule of thermal power units, seasonal hydrogen storage configuration schemes and energy allocation plans, as well as wind curtailment rates and solar curtailment rates. The preprocessing module, based on the seasonal hydrogen storage configuration scheme, energy allocation plan, and wind and solar curtailment rates output by the annual or multi-year time-domain production simulation module, optimizes and smooths the original net load curve with the goal of minimizing the difference in the net load curve, and obtains the smoothed net load curve; based on the maintenance schedule of thermal power units, it calculates the overall available capacity and operating cost of thermal power units in time periods with a preset time resolution. The frequency domain multi-type energy storage collaborative configuration module, based on the smoothed net load curve output by the preprocessing module and the overall available capacity and operating cost of the thermal power unit, constructs a frequency domain multi-type energy storage collaborative configuration model with a preset time resolution to determine the collaborative configuration scheme of multiple types of energy storage.

[0039] Example 3 This embodiment discloses a storage medium storing a program. When the program is executed by a processor, it implements the time-frequency domain combined multi-type energy storage collaborative configuration method described in Embodiment 1.

[0040] The storage medium in this embodiment can be a disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), USB flash drive, portable hard drive, etc.

[0041] Example 4 This embodiment discloses a computing device, including a processor and a memory for storing processor-executable programs. When the processor executes the program stored in the memory, it implements the time-frequency domain combined multi-type energy storage collaborative configuration method described in Embodiment 1.

[0042] The computing device described in this embodiment may be a desktop computer, laptop computer, smartphone, PDA handheld terminal, tablet computer, programmable logic controller (PLC), or other terminal device with processor function.

[0043] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

Claims

1. A time-frequency domain combined multi-type energy storage collaborative configuration method, characterized in that, Includes the following steps: Step 1): Read the load time series curve and construct an approximate load duration curve with a weekly time resolution; read the wind power and photovoltaic output time series data and construct the discrete probability distribution of wind power and photovoltaic output. Step 2): Based on the approximate load duration curve and discrete probability distribution of wind and solar power output constructed in Step 1), conduct annual or multi-year time-domain production simulations with a weekly time resolution. The objective function is to minimize the sum of thermal power operating costs, energy storage operating costs, energy storage investment costs, wind and solar curtailment penalties, and load shedding penalties. The constraints adopted include power balance constraints, unit maintenance constraints, thermal power unit output constraints, new energy unit constraints, and seasonal hydrogen storage operation constraints. By solving the annual or multi-year time-domain production simulations, the maintenance schedule of thermal power units, seasonal hydrogen storage configuration schemes, energy allocation plans, wind curtailment rates, and solar curtailment rates are determined. Step 3): Based on the load time-series curve in Step 1) and the seasonal hydrogen storage configuration scheme, energy allocation plan, wind curtailment rate, and solar curtailment rate calculated in Step 2), a preprocessing module is constructed as boundary conditions. The preprocessing module takes minimizing the difference between the net load curves as the objective function and uses the net load curve calculation constraints, new energy unit related constraints, and seasonal hydrogen storage energy allocation constraints as constraints to optimize and calculate a smoothed net load curve. The net load curve is given by the net load curve calculation constraints. Step 4): Based on the maintenance schedule of the thermal power units described in Step 2), calculate the overall available capacity and operating cost of the thermal power units on a time-period basis with a preset time resolution, and the smoothed net load curve described in Step 3), construct a frequency domain multi-type energy storage collaborative configuration model. The frequency domain multi-type energy storage collaborative configuration model takes minimizing the sum of thermal power operating cost, energy storage operating cost, and energy storage investment cost as its objective function. The constraints adopted include spectrum distribution coefficient constraints, system power balance constraints, overall thermal power unit operation constraints, energy storage operation constraints, and energy storage configuration constraints. By solving the frequency domain multi-type energy storage collaborative configuration model, a collaborative configuration scheme for multiple types of energy storage is determined.

2. The time-frequency domain combined multi-type energy storage collaborative configuration method according to claim 1, characterized in that, In step 1), the approximate load duration curve with a weekly time resolution is constructed using a three-segment approximate load duration curve; the annual or multi-year time-domain production simulation employs a stochastic optimization framework to handle load and new energy uncertainties, specifically including: The formula for constructing the three-segment approximate load duration curve is as follows: （1）； In the formula, s is the index of the load segment of the approximate load duration curve, and k is the index of the load segment of the load duration curve; This represents the load value corresponding to load segment k in the load duration curve; This represents the load value corresponding to the load segment index s in the approximate load duration curve; , These represent the start and end indices of the load segment of the approximate load duration curve, denoted by load segment index s. Then, based on the time-series mapping information of load time-series curve data and wind power and photovoltaic output time-series data, the wind power and photovoltaic output values ​​corresponding to the load segments of each load duration curve are obtained, and the discrete probability distribution of wind power and photovoltaic output corresponding to the load segments of each approximate load duration curve is calculated. The discrete probability distribution of wind power and photovoltaic output can retain the time-series correlation between load and wind power and photovoltaic output, as well as the randomness of wind power and photovoltaic output. The formulas for calculating the discrete probability distribution of wind and solar power output are as follows: （2）； In the formula, This indicates that the wind power output value corresponding to the load segment index s of the approximate load duration curve is... The probability of time, which is given by Give; This indicates that the photovoltaic output value corresponding to the load segment index s of the approximate load duration curve is... The probability of time, which is given by Give; , These are the probability values ​​of wind power and photovoltaic power under the load segment index s and load segment index k of the approximate load duration curve, respectively, and they satisfy the normalization condition.

3. The time-frequency domain combined multi-type energy storage collaborative configuration method according to claim 2, characterized in that, In step 2), firstly, the objective function for the annual or multi-year time-domain production simulation is determined: （3）； In the formula, The total investment, operation, and maintenance costs for a year- or multi-year time-domain production simulation; This represents the total number of load segments on the approximate load duration curve. This represents the total number of thermal power units. The duration in hours of the load segment index s of the approximate load duration curve; Let g be the coal consumption function of the thermal power unit. The load segment index of the approximate load duration curve represents the generating power of thermal power unit g; , and These are the curtailed wind power, curtailed solar power, and unloaded power, respectively, for the load segment index s of the approximate load duration curve. , and These are the cost coefficients for wind curtailment, solar curtailment, and load shedding; Investment and operation and maintenance costs for seasonal hydrogen storage; The total investment and operating cost of seasonal hydrogen storage includes investment costs and operation and maintenance costs. Since different energy storage systems have different operational lifespans, the investment costs and operation and maintenance costs need to be converted to annual values, expressed as follows: （4）； （5）； In the formula, The annual interest rate; The service life of seasonal hydrogen storage; , These represent the unit capacity cost and unit power cost of seasonal hydrogen storage, respectively. , These are the rated capacity and rated power of seasonal hydrogen storage, respectively; Annual operating and maintenance costs for seasonal hydrogen storage; , These are the operation and maintenance costs per unit power and per unit capacity of seasonal hydrogen storage, respectively. The constraints for annual or multi-year time-domain production simulations include: (1) Power balance constraints: （6）； In the formula, The load segment index of the approximate load duration curve represents the generating power of thermal power unit g; , The indexes s of the approximate load duration curve load segments represent the power generation capacity of wind turbines and photovoltaic units, respectively. , These are the charging and discharging power of seasonal hydrogen storage, respectively, based on the load segment index of the approximate load duration curve; (2) Unit maintenance constraints: Unit maintenance constraints include maintenance frequency constraints, maintenance time constraints, and maintenance continuity constraints, which are mathematically expressed as follows: （7）； In the formula, W and w represent the total number of weeks and the week index of the annual or multi-year time-domain production simulation, respectively. Under the three-segment approximate load duration curve, each week index corresponds to three load segment indices. As an intermediate variable; This represents the maintenance status transition variable for thermal power unit g at time w. It is set to 1 when maintenance is initiated at time w, and 0 otherwise. This indicates the required number of maintenance operations for thermal power unit g during a year or multiple years of time-domain production simulation. For the maintenance status variable of Zhouw thermal power unit, take 1 when it is under maintenance, otherwise take 0; For thermal power unit g, the maintenance time is; This represents the start-up and shutdown status of the Zhouw thermal power unit. If it is in the start-up state, use 1; otherwise, use 0. The unit maintenance status variables and thermal power unit start-up and shutdown status of the load segment index s of the approximate load duration curve should be consistent with the weekly index w corresponding to the load segment index s. The mathematical expression is: （8）； In the formula, , For load segment index The unit maintenance status variables and the start-up and shutdown status variables of the thermal power unit; (3) Output constraints of thermal power units: The output constraints of thermal power units include upper and lower limits for thermal power output, and the mathematical expressions are as follows: （9）； In the formula, , These are the minimum and maximum technical outputs of thermal power unit g, respectively. (4) Constraints on new energy units: Constraints on new energy units include upper output limits for wind and solar power units, as well as constraints related to wind and solar curtailment. The mathematical expression is: （10）； （11）； In the formula, , The wind power and photovoltaic power generation of the approximate load duration curve load segment index s are respectively obtained by sampling from the discrete probability distribution of wind power and photovoltaic power output described in step 2). (5) Seasonal constraints on hydrogen storage operation: Seasonal hydrogen storage operation constraints include seasonal hydrogen storage charge / discharge power constraints, mutual exclusion constraints of charge / discharge states, continuity constraints of energy state changes, and constraints on the equality of the initial and final energy states of seasonal hydrogen storage. The mathematical expressions are as follows: （12）； In the formula, , These represent the charging and discharging states of seasonal hydrogen storage at index s of the load segment of the approximate load duration curve, respectively. , These represent the maximum and minimum energy states of seasonal hydrogen storage, respectively. This indicates the energy state of seasonal hydrogen storage at index s of the approximate load duration curve. This indicates the energy state of seasonal hydrogen storage at index s-1 of the approximate load duration curve. Indicates the charge / discharge efficiency of seasonal hydrogen storage; The aforementioned annual or multi-year time-domain production simulation is a mixed-integer linear programming problem. By solving it through an operations research optimization solver, we can obtain the maintenance schedule of thermal power units, seasonal hydrogen storage capacity configuration and seasonal allocation scheme, as well as the wind curtailment rate and solar curtailment rate thresholds. This will provide important boundary conditions for the smoothing of the net load curve in step 3) and the collaborative configuration model of multiple types of energy storage in step 4), ensuring the coordination and consistency of the collaborative configuration of multiple types of energy storage in the time and frequency domains.

4. The time-frequency domain combined multi-type energy storage collaborative configuration method according to claim 3, characterized in that, In step 3), the mathematical expression of the objective function of the preprocessing module is: （13）； In the formula, , As an auxiliary variable, it represents the maximum and minimum values ​​of the net load curve; The constraints include: (1) Constraints for net load curve calculation: The constraints for net load curve calculation include the net load curve calculation definition constraints, the maximum net load curve constraints, and the minimum net load curve constraints, and their mathematical expressions are as follows: （14）； In the formula, t represents the time period index of the net load curve; , These represent the load values ​​for time period t in the net load curve and the load time series curve, respectively. , These represent the power generation of the wind turbine and the photovoltaic unit during time period t, respectively. , These represent the charging power and discharging power of seasonal hydrogen storage during time period t, respectively. (2) Constraints related to new energy units: The constraints related to new energy units include upper limits for wind power output, upper limits for photovoltaic power output, wind curtailment, photovoltaic curtailment, and total curtailment rate. The mathematical expressions are as follows: （15）； In the formula, α is the curtailment rate of wind and solar power determined by annual or multi-year time-domain production simulation, and represents the curtailment rate threshold for minimizing the difference in net load curves in step 3); T represents the time period index of the load time-series curve and the net load time-series curve; , These represent the time-series data for wind power and solar power output, respectively. (3) Seasonal hydrogen storage energy allocation constraints: Seasonal hydrogen storage energy allocation constraints include seasonal hydrogen storage charge / discharge power constraints, charge / discharge state mutual exclusion constraints, energy state change continuity constraints, and weekly energy transport constraints, the mathematical expressions of which are: （16）； In the formula, Indicates the time resolution of the net load curve; , These represent the charging and discharging states of seasonal hydrogen storage during time period t, respectively. , These represent the maximum and minimum energy states of seasonal hydrogen storage, respectively. This indicates the energy state of seasonally stored hydrogen during time period t; This indicates the energy state of seasonally stored hydrogen during period t-1; This indicates that when the time resolution of the net load curve is... Number of time slots per week; This represents the weekly energy state of seasonal hydrogen storage in annual or multi-year time-domain production simulations, serving as a boundary condition. The above preprocessing module is a mixed integer linear programming problem. It is solved by an operations research optimization solver to obtain a smoothed net load curve, which provides important boundary conditions for the frequency domain multi-type energy storage collaborative configuration model in step 4).

5. The time-frequency domain combined multi-type energy storage collaborative configuration method according to claim 4, characterized in that, In step 4), the objective function of the frequency domain multi-type energy storage collaborative configuration model is: （17）； In the formula, The overall operating cost of thermal power units; The investment and operating costs of pumped storage hydroelectric power. The investment and operating costs of battery energy storage; The investment and operating costs for flywheel energy storage; Similar to annual or multi-year time-domain production simulations, the total investment and operating cost of an energy storage system includes initial purchase costs and operation and maintenance costs. Since each energy storage system has a different lifespan, the total investment and operating cost needs to be converted to an equivalent annual value, expressed as: （18）； （19）； In the formula, , , These refer to the service life of pumped hydro storage, battery energy storage, and flywheel energy storage, respectively. , , These are the unit capacity costs for pumped hydro storage, battery energy storage, and flywheel energy storage, respectively. , , These are the unit power costs for pumped hydro storage, battery storage, and flywheel storage, respectively. , , These are the rated capacities of pumped hydro storage, battery storage, and flywheel storage, respectively. , , These are the rated power of pumped hydro storage, battery energy storage, and flywheel energy storage, respectively. , , These are the annual operation and maintenance costs for pumped hydro storage, battery energy storage, and flywheel energy storage, respectively. , , These are the operation and maintenance costs per unit capacity for pumped hydro storage, battery energy storage, and flywheel energy storage, respectively. , , These are the operation and maintenance costs per unit power for pumped hydro storage, battery energy storage, and flywheel energy storage, respectively. During the coordinated operation of thermal power and energy storage, energy storage devices will generate energy losses during the charging and discharging process. The energy storage power curve obtained by discrete Fourier transform analysis only contains periodic components, and its power integral value is zero in the complete cycle. However, due to energy losses in actual operation, the energy state of energy storage may shift at the end of the cycle. In order to maintain the consistency between the initial and final energy state of energy storage, thermal power units need to provide additional power compensation to make up for the energy losses generated by energy storage devices during the charging and discharging process. The resulting power generation cost is defined as the compensation cost. Operating cost of thermal power units C g Including basic output cost and compensation cost, the mathematical expression is: （20）； （21）； In the formula, The overall coal consumption function of the thermal power unit is obtained by weighting the thermal power units that have not undergone maintenance; This variable represents the maintenance status of thermal power unit g during time period t. It is 1 if the unit is under maintenance, and 0 otherwise. This value is given by the maintenance schedule of thermal power unit determined by the annual or multi-year time-domain production simulation results in step 2). , These refer to the overall basic output and compensation output of the thermal power unit, respectively. , , These represent the overall compensation power of the generator unit for pumped storage, battery storage, and flywheel storage during time period t, respectively, satisfying the following: （22）； In the formula, , , These represent the charging power of pumped hydro storage, battery storage, and flywheel storage, respectively. , , These represent the discharge power of pumped hydro storage, battery storage, and flywheel storage, respectively. , , These represent the charge and discharge efficiencies of pumped hydro storage, battery storage, and flywheel storage, respectively. Constraints: (1) Spectral distribution coefficient constraint: The net load curve is decomposed into components of different frequencies using Fourier transform technology. Spectral clustering is performed according to the energy magnitude of each frequency, and allocation is carried out using the spectral distribution coefficient. The following frequency-based division of labor is implemented: For low-frequency components with high energy proportions and slow fluctuations, both thermal power units and pumped storage with strong regulation capabilities share the load, with the DC component (zero frequency component) entirely handled by thermal power units; for medium-frequency fluctuation components with moderate variation characteristics, pumped storage with excellent charging and discharging efficiency and high energy density, and battery energy storage are used for balancing; for high-frequency fluctuation components with drastic changes, flywheel energy storage with fast response speed and outstanding cycle life is responsible for smoothing them. Through this frequency-based division of labor, the technical and economic advantages of different types of energy storage equipment can be fully utilized. The mathematical expression is: （23）； In the formula, n is the index of the spectral component group of the cluster; f is the frequency value; , , , These represent the spectral allocation coefficients of the overall thermal power unit, pumped storage, battery energy storage, and flywheel energy storage in the spectral component group n, respectively. This is the highest frequency that pumped storage can handle; , These are the lowest and highest frequencies that battery energy storage can handle, respectively. This is the lowest frequency that flywheel energy storage can handle; (2) Overall operational constraints of thermal power units: The overall operational constraints of thermal power units include upper and lower output limits, ramp rate limits, and actual available capacity constraints. The overall actual available capacity of thermal power units needs to be dynamically adjusted according to the maintenance schedule determined in step 2). That is, the available capacity of the corresponding unit is zero during maintenance and the rated capacity is used during non-maintenance periods. The mathematical expression is: （24）； （25）； In the formula, , These are the lower and upper limits of the overall output of the thermal power unit in time period t, taking into account the unit's maintenance. and These represent the lower and upper limits of the overall ramp power of the thermal power unit during time period t, respectively. This represents the total power generation of the thermal power unit during time period t. This represents the overall power generation of the thermal power unit during time period t-1; Let be the overall maintenance capacity of the thermal power unit during time period t, and let be the sum of the maintenance capacities of the thermal power unit in each time period. Its mathematical expression is: （26）； (3) Energy storage operation constraints: Energy storage operation constraints include energy storage charging and discharging power constraints and energy storage energy state change constraints, the mathematical expressions of which are: （27）； In the formula, , , These represent the net charging and discharging power of pumped hydro storage, battery storage, and flywheel storage during time period t, respectively. , , The energy states of pumped hydro storage, battery energy storage, and flywheel energy storage are respectively defined for time period t. , , These represent the energy states of pumped hydro storage, battery energy storage, and flywheel energy storage during time period t-1. (4) Constraints on energy storage configuration: Energy storage configuration constraints include energy storage power configuration constraints and energy storage capacity configuration constraints; The linearized expression for the energy storage power configuration constraint is: （28）； The linearized expression for the energy storage capacity configuration constraint is: （29）； （30）； In the formula, , These represent the maximum and minimum energy states of pumped storage, respectively. , These are the maximum and minimum SOC values ​​for pumped storage, respectively. , These represent the maximum and minimum states of energy for battery storage, respectively. , These represent the maximum and minimum SOC values ​​for battery energy storage, respectively. , These represent the maximum and minimum energy states of the flywheel energy storage, respectively. , These represent the maximum and minimum SOC values ​​for flywheel energy storage, respectively. The frequency domain multi-type energy storage collaborative configuration model transforms the original mixed integer programming optimization problem involving the charging and discharging state variables of multiple types of energy storage into a linear programming problem by linearizing the constraints. This allows the model to be solved by an operations research optimization solver, yielding the collaborative configuration results of multiple types of energy storage.

6. A multi-type energy storage collaborative configuration system combining time and frequency domains, characterized in that, The method for collaborative configuration of multiple types of energy storage combining time and frequency domains as described in any one of claims 1 to 5 includes: The data acquisition module is responsible for reading load time-series curve data and constructing an approximate load duration curve with a weekly time resolution; it also reads wind power and solar power output time-series data and constructs discrete probability distributions for wind power and solar power output. The annual or multi-year time-domain production simulation module is used to perform annual or multi-year time-domain production simulations with a weekly time resolution. It comprehensively considers the operating costs of thermal power plants, energy storage operating costs, energy storage investment costs, wind and solar curtailment penalties, and load shedding penalties. Under the conditions of satisfying power balance constraints, unit maintenance constraints, thermal power unit output constraints, new energy unit constraints, and seasonal hydrogen storage operation constraints, it outputs the maintenance schedule of thermal power units, seasonal hydrogen storage configuration schemes and energy allocation plans, as well as wind curtailment rates and solar curtailment rates. The preprocessing module, based on the seasonal hydrogen storage configuration scheme, energy allocation plan, and wind and solar curtailment rates output by the annual or multi-year time-domain production simulation module, optimizes and smooths the original net load curve with the goal of minimizing the difference in the net load curve, and obtains the smoothed net load curve; based on the maintenance schedule of thermal power units, it calculates the overall available capacity and operating cost of thermal power units in time periods with a preset time resolution. The frequency domain multi-type energy storage collaborative configuration module, based on the smoothed net load curve output by the preprocessing module and the overall available capacity and operating cost of the thermal power unit, constructs a frequency domain multi-type energy storage collaborative configuration model with a preset time resolution to determine the collaborative configuration scheme of multiple types of energy storage.

7. A storage medium storing a program, characterized in that, When the program is executed by the processor, it implements the time-frequency domain combined multi-type energy storage collaborative configuration method as described in any one of claims 1 to 5.

8. A computing device, comprising a processor and a memory for storing a processor-executable program, characterized in that, When the processor executes the program stored in the memory, it implements the multi-type energy storage collaborative configuration method combining time and frequency domains as described in any one of claims 1 to 5.

Images