A method for collaborative capacity configuration of pumped storage and electrochemical energy storage based on multi-dimensional value evaluation

Abstract

The application provides a kind of based on multi-dimension value evaluation pumped storage and electrochemical energy storage collaborative capacity configuration method, belongs to the technical field of power system planning and energy storage optimization configuration, including: constructing evaluation index system and standardization;Establish pumped storage and electrochemical energy storage operation constraint model, divide multi-time scale application scene and form value weight matrix;Establish double-layer optimization configuration model, through upper multi-objective capacity configuration and lower time sequence operation simulation closed loop iteration obtains Pareto solution set;Introduce dynamic weight adjustment mechanism, according to policy and technical cost factor adaptive correction weight;Solve by using multi-objective intelligent optimization algorithm, combine TOPSIS to select optimal scheme;The application comprehensively evaluates the multi-dimensional value of energy storage, realizes the complementary advantages of pumped storage and electrochemical energy storage, avoids capacity redundancy, can improve new energy consumption rate, reduces carbon emission, scene adaptability is strong, provides decision support for high proportion new energy power system energy storage planning.

Description

Technical Field

[0001] This invention relates to the field of power system planning and energy storage technology, specifically to a method for synergistic capacity configuration of pumped hydro storage and electrochemical energy storage based on multi-dimensional value assessment. Background Technology

[0002] With the deepening implementation of the "dual carbon" goals, my country is accelerating the construction of a new power system dominated by new energy sources. By the end of 2025, the cumulative installed power generation capacity nationwide reached 3.89 billion kilowatts, a year-on-year increase of 16.1%. Among them, the cumulative installed capacity of wind power and photovoltaic power exceeded 1.8 billion kilowatts, reaching 1.84 billion kilowatts, accounting for 47.3% of the total installed power generation capacity. In some provinces, the penetration rate of new energy has exceeded 50%. The strong randomness, intermittency, and anti-peak-shaving characteristics of new energy output have led to a sharp increase in the demand for flexible adjustment of the power system in terms of peak shaving, frequency regulation, voltage support, and rotational inertia. Traditional regulation methods, mainly based on thermal power and hydropower, face constraints such as insufficient deep peak-shaving capacity, slow response speed, and high carbon emission pressure. There is an urgent need for large-scale, multi-type energy storage technologies as the core supporting resource for flexible system regulation.

[0003] Currently, major energy storage technologies can be divided into two main categories: pumped hydro storage and new energy storage represented by electrochemical energy storage. These two types of energy storage have significantly different characteristics. Pumped hydro storage offers large single-station capacity, continuous discharge time of 6-12 hours, a lifespan exceeding 50 years, and low cost per kilowatt-hour, making it suitable for long-cycle energy transfer tasks. However, its site selection is constrained by geographical conditions, construction period is long (6-8 years), and its rapid response capabilities, such as primary frequency regulation, are relatively weak. Electrochemical energy storage (such as lithium-ion batteries) offers millisecond-level response speed, fast ramp-up rate, flexible deployment, and short construction period (several months), demonstrating significant advantages in scenarios such as rapid frequency regulation and instantaneous power support. However, the continuous discharge time of a single unit is typically 1-4 hours, and its calendar life and cycle life are limited, making it difficult to independently undertake long-term peak-shaving tasks. The two technologies naturally complement each other in terms of time scale (short-term rapid response vs. long-term large capacity) and spatially, they represent a potential for distributed and centralized synergy, necessitating unified planning.

[0004] However, existing energy storage capacity configuration methods often focus on a single energy storage type or a single economic objective, failing to comprehensively reflect the integrated value of energy storage in areas such as renewable energy consumption, power supply reliability, carbon emission reduction, and industrial chain development. This approach tends to underestimate the strategic role of long-cycle projects like pumped hydro storage. Furthermore, pumped hydro storage and electrochemical energy storage are often planned separately, failing to fully consider their complementary relationship in terms of regulation rate, charge / discharge duration, and response capability. This may result in redundant configuration of some regulation capacity, leaving regulation gaps at key time scales. In addition, energy storage investment and operation cycles are long, and factors such as renewable energy scale, technology costs, market mechanisms, and carbon prices are constantly changing. Existing static configuration methods do not adequately consider output uncertainty and external environmental evolution, limiting the robustness and long-term applicability of the solutions.

[0005] Therefore, there is an urgent need for a planning method that can comprehensively assess the multidimensional value of pumped storage and electrochemical energy storage, achieve deep synergistic configuration between the two, and have the ability to adapt to multiple scenarios, so as to provide systematic support for large-scale scientific investment in energy storage. Summary of the Invention

[0006] The purpose of this invention is to address the problems mentioned in the background art, such as insufficient multi-dimensional value assessment, inadequate synergistic configuration of pumped hydro storage and electrochemical energy storage, and insufficient adaptability to long-term scenarios. This invention provides the following technical solution: A method for synergistic capacity configuration of pumped hydro storage and electrochemical energy storage based on multi-dimensional value assessment includes the following steps: S1. Obtain basic data on energy output, load, electricity price, ancillary service price, carbon price, energy storage investment and operation and maintenance cost, site resources, planning constraints and policy parameters of the power system to be planned, and construct a multi-dimensional value assessment index system, which includes technical value index, economic value index, environmental value index and social value index. The technical value indicators, economic value indicators, environmental value indicators, and social value indicators are quantified and standardized respectively. The standardization process adopts the range standardization method to map the technical value indicators, economic value indicators, environmental value indicators, and social value indicators to a unified dimension range to obtain standardized value indicators. S2. Based on the operating characteristics of pumped storage and electrochemical energy storage, establish operating constraint models for pumped storage and electrochemical energy storage respectively. Divide different application scenarios according to the duration of regulation demand, response speed and power change amplitude. Calculate the adaptability of pumped storage and electrochemical energy storage in different application scenarios. Arrange the adaptability according to two dimensions: energy storage type and application scenario to form a differentiated value contribution matrix. Based on the differentiated value contribution matrix, use the analytic hierarchy process (AHP) to form the value weight matrix for different application scenarios. S3. Establish a two-layer optimization configuration model with the goal of maximizing the comprehensive value of the standardized technical value indicators, economic value indicators, environmental value indicators and social value indicators obtained in step S1. The two-layer optimization configuration model includes an upper-layer multi-objective capacity configuration optimization model and a lower-layer time-series operation simulation model. The upper-layer multi-objective capacity configuration optimization model generates an initial capacity configuration scheme and sends it to the lower-layer time-series simulation model. The lower-layer time-series simulation model calculates the actual value contribution and feeds it back to the upper-layer multi-objective capacity configuration optimization model. The upper-layer multi-objective capacity configuration optimization model updates the capacity configuration scheme. Through the closed-loop iteration of multi-objective optimization by the upper-layer multi-objective capacity configuration optimization model and simulation verification by the lower-layer time-series simulation model, the capacity configuration scheme is corrected and verified. S4. Introduce a dynamic weight adjustment mechanism to adaptively adjust the weights of the four dimensions of technology value indicators, economic value indicators, environmental value indicators and social value indicators based on policy factors and technology cost factors, so as to obtain the dynamic weights corresponding to the four dimensions. S5. The upper-level multi-objective capacity configuration optimization model is solved using a multi-objective intelligent optimization algorithm. After obtaining the Pareto non-dominated solution set, the optimal capacity configuration scheme with the best comprehensive value is selected from the Pareto non-dominated solution set by combining the adaptively corrected dynamic weights and using the TOPSIS method. The scheme is then output as a synergistic capacity configuration scheme of pumped storage and electrochemical energy storage.

[0007] Further preferably, in step S1, the technical value indicators include the contribution of frequency regulation response speed, the increase in new energy consumption, and the improvement in power supply reliability. The contribution of frequency regulation response speed is quantified based on the ratio of the actual power regulation provided by energy storage to the power regulation required by the system. The increase in new energy consumption is quantified based on the difference in power curtailment before and after energy storage configuration. The improvement in power supply reliability is quantified based on the difference in system load shedding power before and after energy storage configuration, combined with the unit outage loss cost. The economic value indicators include net benefits over the entire life cycle. The environmental value indicators include carbon emission reduction value. The social value indicators include the industrial chain driving effect and the employment driving effect.

[0008] More preferably, the contribution of the frequency modulation response speed is quantified by the following formula:

[0009] in For the first The actual power regulation provided by energy storage during each time period This represents the power regulation required by the system. The value coefficient per unit frequency modulation. This represents the total number of time periods within the scheduling cycle. The increase in new energy consumption is quantified by the following formula:

[0010] in and These are the first and second energy storage configurations before and after. The amount of power abandoned during each time period For time step; The power supply reliability improvement index is quantified by the following formula:

[0011] In the formula and These represent the system load shedding power before and after configuring energy storage, respectively. Cost of power outage losses per unit.

[0012] More preferably, the net income over the entire life cycle of the economic value indicator is calculated using the following formula:

[0013] in For the first Annual peak-valley arbitrage profit is calculated based on the price difference of energy storage charging and discharging under time-of-use electricity pricing. For ancillary service revenue, it is calculated based on frequency regulation mileage and capacity compensation; To compensate for the capacity market; To delay the return on investment in power transmission and distribution; Annual maintenance costs; This refers to the initial investment cost; The discount rate; The project's lifespan; The carbon emission reduction value of the environmental value indicator is calculated using the following formula:

[0014] in For the first Energy storage replaces thermal power for peak shaving during specific time periods. Carbon emission factors from thermal power plants For carbon trading prices, This represents the total number of time periods within the scheduling cycle. The industrial chain pulling effect in the aforementioned social value indicators is calculated using the following formula:

[0015] In the formula For the first The total added value factor per unit investment for energy storage-like products. For the first The initial total investment for energy storage-like systems; The employment-driving effect in the aforementioned social value indicators is calculated using the following formula:

[0016] In the formula The social value conversion coefficient of a unit of employment position. For the first The employment coefficient during the construction period of energy storage-type projects. For the first The employment coefficient during the operation period of energy storage-like systems. For the first Installed capacity of energy storage systems For the first The operational lifespan of energy storage systems.

[0017] More preferably, step S2 specifically includes: S21. Establish a pumped storage operation constraint model, which includes power generation constraints, power generation ramping constraints, pumping power constraints, mutually exclusive operating conditions constraints, equivalent energy storage capacity constraints of the reservoir, energy storage recursive balance equations, reservoir capacity boundary constraints at the end of the scheduling cycle, and start-stop frequency constraints. S22. Establish an electrochemical energy storage operation constraint model, which includes charge and discharge power constraints, state of charge constraints, state of charge recursive equations, charge and discharge state-power correlation constraints, cycle lifetime constraints, and ramp rate constraints. S23. Based on the duration of the adjustment demand, the response speed, and the magnitude of power change, different application scenarios are divided into four time scales: millisecond to second, minute to hour, hour to day, and day to week. S24. Calculate the compatibility of pumped storage and electrochemical energy storage under different application scenarios to form a contribution matrix of energy storage type and subdivided value dimensions. The contribution matrix serves as the basis for the quantitative calculation of the value weight matrix under different application scenarios.

[0018] More preferably, the pumped storage operation constraint model in step S21 includes the following constraints: Power generation constraint: and power generation ramp-up constraints : ,in 、 These are the rated maximum power output and the technical minimum power output, respectively. 、 Limits for upward and downward ramp rates during power generation; Pumping power constraint: ,in , These are the rated maximum and technical minimum pumping power, respectively. Mutually exclusive operating conditions: Constrained by formula: ,in , A 0 / 1 state variable, Indicates the first The unit is in power generation mode during each period. This indicates that the system is not in power generation mode. Indicates the first During certain periods, the unit is in pumping mode. This indicates that the unit is not in pumping mode; the mutual exclusion constraint condition for the operating conditions is used to ensure that the pumped storage unit can only be in one of the following states at any given time: power generation, pumping, or shutdown, and simultaneous power generation and pumping are prohibited. ; Equivalent energy storage capacity of reservoir constraint: And the recursive equilibrium equation for energy storage: ,in, For the first The equivalent energy storage capacity of the upper reservoir within a certain time period For the first The equivalent energy storage capacity of the upper reservoir within a certain time period For maximum energy storage capacity, For minimum storage capacity, , These are pumping efficiency and power generation efficiency, respectively. Their product is the overall cycle efficiency of the pumped storage power station. Storage capacity boundary constraints at the end of the scheduling cycle: ,in For the scheduling period, This is the allowable deviation amount; Start-stop count constraint formula: ,in This represents the maximum number of state transitions allowed within the scheduling period. , Representing the first The time period and the first The power generation operating status for each time period. , Representing the first The time period and the first The pumping operation status during each time period. and These represent the initial power generation and initial pumping operation status, respectively; The electrochemical energy storage operation model described in S22 includes the following constraints: definition The net output power of energy storage is represented by positive values ​​indicating charging and negative values ​​indicating discharging; a non-negative auxiliary variable is introduced. , ,satisfy ,and , ,in , This is the maximum charge / discharge power; State of charge constraint: ,in For the first State of charge at each time period, This is the limit of the state of charge. This represents the upper limit of the state of charge. State of charge Recurrence equation: ,in For the first State of charge at each time period, For the first State of charge at each time period, , These represent charging and discharging efficiencies, respectively. , The first Actual charging and discharging power during each time period Rated energy capacity, For time step; Charge / discharge state and power correlation constraints: , , ,in , A 0 / 1 state variable, Indicates the first During each period, the energy storage is in charging mode. This indicates that the device is not in charging mode. Indicates the first During the period when the energy storage is in discharge mode, This indicates that the device is not in a discharge condition. , The first The actual charging power and discharging power within a time period; the charging and discharging state and power correlation constraints are used to ensure that the energy storage unit cannot charge and discharge simultaneously in the same time period, and the charging and discharging power are controlled by the corresponding state variables respectively. Cycle lifetime constraint: The charge-discharge cycles in the state-of-charge curve are extracted using the rainflow counting method to obtain the discharge depth of each cycle. Then the equivalent total loop count constraint is: ,in As the reference discharge depth, γ The fitting index is... This represents the maximum equivalent number of cycles at the reference discharge depth. Climbing speed constraint: ,in This is the climbing speed limit.

[0019] Further preferably, in step S23, different application scenarios are divided according to four time scales: millisecond to second, minute to hour, hour to day, and day to week. These correspond to primary frequency regulation and inertia support, secondary frequency regulation and rapid ramping, day-ahead peak shaving and renewable energy consumption, and long-term power balance under continuous extreme weather conditions, respectively. Based on the energy storage type and subdivided value dimension contribution matrix calculated in S24, electrochemical energy storage is given priority to support primary frequency regulation and inertia, secondary frequency regulation and rapid ramping in the millisecond to second and minute to hour application scenarios, while pumped hydro storage is given priority to support day-ahead peak shaving and renewable energy consumption, and long-term power balance under continuous extreme weather conditions in the hour to day and day to week application scenarios.

[0020] Further preferably, in step S3, the upper-level multi-objective capacity configuration optimization model uses pumped storage installed capacity, pumped storage reservoir capacity, various electrochemical energy storage power capacity and energy capacity as decision variables, and takes maximizing standardized technical value indicators, economic value indicators, environmental value indicators and social value indicators as multiple objective functions, and sets total investment constraints, site resource constraints and system regulation demand constraints. The lower-level time-series operation simulation model is based on the capacity configuration scheme given by the upper-level multi-objective capacity configuration optimization model. With the minimization of system operating cost as the objective function, it simulates the time-series operation of multiple scenarios throughout the year, calculates the actual value contribution of four dimensions: technical value indicators, economic value indicators, environmental value indicators, and social value indicators, and feeds them back to the upper-level multi-objective capacity configuration optimization model. Step S3 specifically includes: S31. Pumped storage installed capacity Cps, pumped storage capacity Eps, various electrochemical energy storage power capacity Cb,i and energy capacity Eb,i are the upper-level decision variables. S32. Establish a multi-objective capacity allocation optimization model at the upper level, with the objective function being to maximize four standardized values: , , ,

[0021] in , , , These are standardized technical value indicators, economic value indicators, environmental value indicators, and social value indicators. The constraints include: Total investment constraints: ,in This represents the total investment cost of pumped storage hydroelectric power. For the first Total investment cost of electrochemical energy storage This represents the upper limit of the total investment budget for energy storage during the planning period; Site resource constraints: , ,in The maximum installed capacity allowed by the site resources for pumped storage hydroelectric power stations. For the first The maximum allowable power capacity of electrochemical energy storage site resources; System peak shaving demand constraints: ,in For the first The maximum discharge power of this type of energy storage Adjust the system's maximum peak-to-valley difference to meet demand capacity requirements; System frequency regulation requirement constraints: ,in For the first Energy storage-like systems provide a capacity conversion factor for frequency regulation. To meet the system's frequency regulation capacity requirements; Lower limit constraint on renewable energy consumption rate: ,in The total amount of electricity wasted annually, This represents the theoretically achievable power generation from new energy sources. S33. Establish a lower-level time-series operation simulation model. Based on the capacity configuration scheme given by the upper level, the objective function is to minimize the system operating cost.

[0022] in For fuel costs, For start-up costs, To account for the cost of power curtailment penalties, This refers to the power that has been abandoned. To incur penalties for load loss, The load shedding power is defined by the constraints, which include power balance constraints, conventional unit output constraints, unit ramping constraints, spinning reserve constraints, line power flow constraints, and constraints of the electrochemical energy storage operation model. S34. The lower-level time-series operation simulation model optimizes the annual energy storage scheduling output based on the scheme of the upper-level multi-objective capacity configuration optimization model, and returns the actual value contribution of each dimension. The four objective function values ​​F1 to F4 are calculated and fed back to the upper-level multi-objective capacity configuration optimization model. The capacity configuration scheme is updated accordingly until the convergence condition or the maximum number of iterations is reached.

[0023] More preferably, the dynamic weight adjustment mechanism in step S4 corrects the weights using the following formula:

[0024] in Based on weights, As a factor of policy change, As a factor of change in technology costs, , To adjust the sensitivity coefficient, These correspond to the weights of technological, economic, environmental, and social value, respectively; the revised weights are used for the TOPSIS comprehensive decision in step S5.

[0025] More preferably, the multi-objective intelligent optimization algorithm in step S5 includes the multi-objective particle swarm optimization algorithm MOPSO or the non-dominated sorting genetic algorithm NSGA-II with an elitist strategy; after obtaining the Pareto non-dominated solution set, the TOPSIS method is used, with the weights adaptively corrected in step S4 or the objective weights determined by the information entropy weight method as the importance of the value index dimension, to select the capacity configuration scheme with the best comprehensive value from the Pareto solution set.

[0026] Compared to existing technologies, the present invention provides a method for synergistic capacity configuration of pumped hydro storage and electrochemical energy storage based on multi-dimensional value assessment, which can bring the following significant benefits: (1) Comprehensive value assessment dimensions, resulting in superior overall benefits from the configuration scheme. This invention breaks through the limitations of traditional methods that only focus on economic costs or single ancillary service benefits, and innovatively incorporates technical performance, economic returns, environmental carbon reduction, and social benefits into a unified assessment framework. The capacity configuration scheme obtained based on this method can improve the renewable energy absorption rate by 5%-15% and reduce the system's carbon emission intensity by 10%-20% while ensuring economic feasibility, thereby maximizing the overall benefits of energy storage investment.

[0027] (2) Pumped storage and electrochemical energy storage complement each other to avoid redundant configuration and resource waste. This invention clarifies the advantages of pumped storage in long-term energy migration and electrochemical energy storage in rapid power support through refined operation modeling and multi-timescale value analysis. It achieves automatic optimal complementarity between the two in the time dimension by using a two-layer optimization model. Compared with separate planning, coordinated configuration can reduce total capacity redundancy by 10%-25%, save investment costs, and improve the overall regulation efficiency of the system. (3) Strong adaptability to various scenarios and long-term robustness. The introduction of a dynamic weight adaptive adjustment mechanism based on policy and technology trends enables the configuration scheme to not only meet current needs, but also adapt to future changes in multiple scenarios such as the increase in the penetration rate of new energy, the decrease in energy storage costs, and the development of the carbon market, providing solid support for long-term investment decisions.

[0028] (4) The present invention has strong versatility and scalability. The method of the present invention is not limited to specific energy storage technology types and power system scales. It can be flexibly extended to the planning of mixed energy storage capacity containing multiple electrochemical energy storage (lithium batteries, flow batteries, sodium-ion batteries, compressed air energy storage, etc.) and is applicable to various scenarios such as provincial and regional power systems. Attached Figure Description

[0029] Figure 1 This is a flowchart illustrating the method of the present invention; Figure 2 This is a diagram illustrating the framework of the multi-dimensional value assessment index system of this invention; Figure 3 This is a structural diagram of the two-layer optimized configuration model of the present invention; Figure 4 This is a flowchart of the solution algorithm based on MOPSO-TOPSIS; Figure 5 This is a comparison chart of the configuration results of a local power grid example in an embodiment of the present invention. Detailed Implementation

[0030] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0031] Example 1: This invention provides a method for synergistic capacity configuration of pumped hydro storage and electrochemical energy storage based on multi-dimensional value assessment. The specific implementation of this invention is described in detail below using actual data from a local power grid. In the 2030 planning scenario for this power grid, the proportion of renewable energy installed capacity reaches 45%, including 12 million kilowatts of wind power and 18 million kilowatts of photovoltaic power, with a maximum system load of 35 million kilowatts. Existing regulation resources are insufficient to support the high proportion of renewable energy consumption, necessitating the configuration of pumped hydro storage and electrochemical energy storage. (See relevant references.) Figure 1 The flowchart of the method of the present invention is shown in the figure. The method for synergistic capacity configuration of pumped storage and electrochemical energy storage based on multi-dimensional value assessment in this embodiment includes the following steps.

[0032] S1. Obtain basic data for the power system to be planned. This data includes renewable energy output, load, electricity price, ancillary service prices, carbon price, energy storage investment and operation and maintenance costs, site resources, planning constraints, and policy parameters. Construct a multi-dimensional value assessment indicator system, including technical value indicators, economic value indicators, environmental value indicators, and social value indicators. Quantify and standardize the indicators in the multi-dimensional value assessment indicator system. Standardization uses the range standardization method to map the four indicators to a unified dimension range, resulting in standardized value indicators.

[0033] In this embodiment, basic data such as renewable energy output, load, electricity price, ancillary service price, carbon price, energy storage investment and operation and maintenance costs, site resources, planning constraints, and policy parameters of the area to be planned are obtained. An energy storage value assessment index system is constructed from four dimensions: technology, economy, environment, and society, with specific quantitative indicators under each dimension. The framework diagram of the multi-dimensional value assessment index system is shown below. Figure 2 As shown.

[0034] (1) Technical value indicators Technical value indicators include frequency regulation response speed contribution, renewable energy absorption improvement, and power supply reliability improvement. Frequency regulation response speed contribution is quantified based on the ratio of the actual power regulation provided by energy storage to the power regulation required by the system. Renewable energy absorption improvement is quantified based on the difference in power curtailment before and after energy storage configuration. Power supply reliability improvement is quantified based on the difference in system load shedding power before and after energy storage configuration, combined with the unit outage loss cost.

[0035] ① Frequency regulation response speed contribution: Measured by the amount of power regulation that the energy storage system can provide within a specified time, and calculated using the following formula:

[0036] in For the first The actual power regulation provided by energy storage during each time period This represents the power regulation required by the system. The value coefficient per unit frequency modulation. This represents the total number of time periods within the scheduling cycle.

[0037] ② Increase in renewable energy consumption: This reflects the reduction in curtailed wind and solar power after energy storage is integrated, and is calculated using the following formula:

[0038] in and These are the first and second energy storage configurations before and after. The amount of power abandoned during each time period For time step.

[0039] ③ Power Supply Reliability Improvement Index: Reflects the contribution of energy storage systems to reducing the risk of power outages in extreme scenarios or during peak load periods. It is quantified by combining the reduction in expected power shortage with the unit outage loss cost, and calculated using the following formula:

[0040] In the formula and These represent the system load shedding power before and after configuring energy storage, respectively. Cost of power outage losses per unit.

[0041] In this embodiment, the specific parameters of the above-mentioned technical value indicators are set as follows: frequency regulation response speed contribution weight 0.3, new energy consumption improvement weight 0.5, and power supply reliability improvement weight 0.2. (2) Economic value indicator – net income over the entire life cycle Calculate using the following formula:

[0042] in For the first Annual peak-valley arbitrage profit is calculated based on the price difference of energy storage charging and discharging under time-of-use electricity pricing. For ancillary service revenue, it is calculated based on frequency regulation mileage and capacity compensation; To compensate for the capacity market; To delay the return on investment in power transmission and distribution; Annual maintenance costs; This refers to the initial investment cost; The discount rate; This refers to the project's lifespan.

[0043] In this embodiment, the total life cycle is set to 25 years, the discount rate is 6%, the peak-valley price difference is 0.5 yuan / kWh, the frequency regulation service price is 8 yuan / MW, and the capacity compensation is 120 yuan / kW·year.

[0044] (3) Environmental value indicators – carbon emission reduction value Calculate using the following formula:

[0045] in For the first The amount of thermal power that can be replaced by energy storage during a given period. Carbon emission factors from thermal power plants For carbon trading prices, This represents the total number of time periods within the scheduling cycle.

[0046] In this embodiment, the carbon price is set at 80 yuan / ton, and the carbon emission factor for thermal power is set at 0.8 kg / kWh.

[0047] (4) Social value indicators ① Supply chain pulling effect: Based on the input-output model, it is measured by multiplying the initial investment in energy storage by the industry's added value multiplier.

[0048] In the formula For the first The total added value factor per unit investment for energy storage-like products. For the first The initial total investment for energy storage-like systems.

[0049] ②Employment multiplication effect: Measured by the equivalent number of jobs created during the project construction and operation periods, combined with the social value coefficient per job.

[0050] In the formula The social value conversion coefficient of a unit of employment position. For the first The employment coefficient during the construction period of energy storage-type projects. For the first The employment coefficient during the operation period of energy storage-like systems. For the first Installed capacity of energy storage systems For the first The operational lifespan of energy storage systems.

[0051] In this embodiment, an input-output model is used to calculate that the added value rate of the energy storage industry is 25%, and every 100 million yuan of investment creates 120 jobs.

[0052] The total value across the aforementioned technological, economic, environmental, and social dimensions , , , All values ​​are mapped to the [0,1] interval through range standardization to form standard values. :

[0053] in and This refers to the lower and upper bounds of the value of this dimension, which are pre-calculated using multiple typical configuration schemes.

[0054] S2. Based on the operating characteristics of pumped storage and electrochemical energy storage, establish operating constraint models for pumped storage and electrochemical energy storage respectively. Divide different application scenarios according to the duration of regulation demand, response speed and power change amplitude. Calculate the adaptability of pumped storage and electrochemical energy storage under the different application scenarios. Arrange the adaptability according to the two dimensions of energy storage type and application scenario to form a differentiated value contribution matrix. Based on the differentiated value contribution matrix, use the analytic hierarchy process (AHP) to form the value weight matrix under the different application scenarios.

[0055] This step aims to establish refined operation models for pumped hydro storage and various types of electrochemical energy storage, characterize their technical differences in regulation rate, continuous discharge time, response time constant, etc., and analyze their differentiated value contribution in the power system based on multiple time scales, providing technical parameter inputs and decision weight references for subsequent multi-objective two-level optimization configuration.

[0056] S21. Establish a refined operation model for pumped storage. The pumped storage operation constraint model includes power generation constraints, power generation ramping constraints, pumping power constraints, mutually exclusive operating conditions constraints, equivalent energy storage capacity constraints of the reservoir, energy storage recursive balance equations, reservoir capacity boundary constraints at the end of the scheduling cycle, and start-stop frequency constraints.

[0057] In this embodiment, the pumped storage power station has two basic operating conditions: power generation and pumping. Its operating characteristics are constrained by factors such as reservoir capacity, water head changes, and the unit's climbing ability.

[0058] (1) Power generation operating condition model When the pumped-storage unit is in operation, it converts the potential energy of the upper reservoir into electrical energy to meet the power shortage during peak load periods or periods of low output from new energy sources. ①Power generation constraints :

[0059] ②Power generation ramp rate constraint :

[0060] in , These are the rated maximum power output and the technical minimum power output, respectively. , The limits for the upward and downward ramp rates under power generation conditions are typically 30% to 50% of the rated power.

[0061] (2) Pumping operation model Pumped storage units consume system electrical energy to lift water from the lower reservoir to the upper reservoir during pumping operations, achieving a spatial and temporal transfer of energy. ① Pumping power constraint:

[0062] in , These are the rated maximum and technical minimum pumping power, respectively. ② Pumping power adjustability: The pumping power of some variable-speed pumped-storage units can be continuously adjusted within a certain range; fixed-speed units usually have a fixed power, in which case... It can only be started and stopped.

[0063] in , These are the rated maximum and technical minimum pumping power, respectively.

[0064] (3) Mutual exclusion constraints of operating conditions Pumped storage units can only be in one of three states at any given time: generating electricity, pumping water, or shut down. ,in , A 0 / 1 state variable, Indicates the first The unit is in power generation mode during each period. This indicates that the system is not in power generation mode. Indicates the first During certain periods, the unit is in pumping mode. This indicates that the unit is not in pumping mode. The mutually exclusive constraint condition is used to ensure that the pumped storage unit can only be in one of the following states at the same time: power generation, pumping, or shutdown. Simultaneous power generation and pumping are prohibited.

[0065] (4) Reservoir energy storage and capacity balance model ① Equivalent state of charge of the upper reservoir (energy storage capacity) )constraint:

[0066] in For the first The equivalent energy storage capacity of the upper reservoir within a certain time period This is the maximum energy storage capacity (determined by the reservoir capacity). This is the minimum storage capacity (usually the energy storage corresponding to dead storage capacity).

[0067] ② Recursive equilibrium equation for reservoir energy storage:

[0068] in For the first The equivalent energy storage capacity of the upper reservoir within a certain time period For the first The equivalent energy storage capacity of the upper reservoir within a certain time period For maximum energy storage capacity, For minimum storage capacity, , These are pumping efficiency and power generation efficiency, respectively. Their product is the overall cycle efficiency of the pumped storage power station, which is usually 75% to 80%.

[0069] ③ Daily / Weekly Scheduling Cycle End-of-Cycle Storage Capacity Boundary Constraints: To ensure continuous operation, storage capacity is typically required to return to its initial state or a set range at the end of the scheduling cycle.

[0070] in For the scheduling period, This is the allowable deviation.

[0071] (5) Start-stop count and state transition constraints Frequent start-ups and shutdowns of pumped-storage units accelerate equipment wear and tear, necessitating the setting of a start-up / shutdown frequency limit. The formula for this limit is as follows:

[0072] in This represents the maximum number of state transitions allowed within the scheduling period. , Representing the first The time period and the first The power generation operating status for each time period. , Representing the first The time period and the first The pumping operation status during each time period. and These represent the initial power generation and initial pumping operation status, respectively.

[0073] S22, Refined Operation Model of Electrochemical Energy Storage An electrochemical energy storage operation constraint model is established, which includes charge and discharge power constraints, state of charge constraints, state of charge recursive equations, charge and discharge state-power correlation constraints, cycle life constraints, and ramp rate constraints.

[0074] Electrochemical energy storage primarily utilizes lithium-ion batteries and flow batteries, but also includes compressed air energy storage. The following model uses lithium-ion batteries as a representative example to establish a general model; different types of electrochemical energy storage can be adapted by adjusting parameters. The electrochemical energy storage operation model includes the following constraints: (1) Definition and constraints of charging and discharging power definition This represents the net output power of energy storage; positive values ​​indicate charging, and negative values ​​indicate discharging. A non-negative auxiliary variable is introduced. , ,satisfy: ,and , ,in , The maximum charge / discharge power is usually equal to the rated power of the converter. For lithium batteries, the charge / discharge rate is typically 1C to 4C (i.e., the ratio of charge / discharge power to energy capacity). Flow batteries, due to their decoupled power and capacity design, can have their rate flexibly configured in the range of 0.5C to 2C.

[0075] (2) State of charge constraint

[0076] in For the first The state of charge (between 0 and 1) for each time period. This is the limit of the state of charge. This represents the upper limit of the state of charge. It is usually taken as 1.0. To avoid over-discharge, a value of 0.1~0.2 is typically used. For flow batteries, since the electrolyte can be recycled, It can be as low as 0.05.

[0077] (3) State of charge recurrence equation

[0078] For the first State of charge at each time period, For the first State of charge at each time period, , These refer to charging and discharging efficiencies, with lithium batteries having a combined efficiency of approximately 88% to 95% and flow batteries approximately 75% to 85%. , The first Actual charging and discharging power during each time period Rated energy capacity, For time step.

[0079] (4) Constraints on the relationship between charge / discharge state and power Individual energy storage units cannot be charged and discharged simultaneously at the same time, and their power is limited by state variables: , , ,in , For 0-1 state variables, Indicates the first During each period, the energy storage is in charging mode. This indicates that the device is not in charging mode. Indicates the first During the period when the energy storage is in discharge mode, This indicates that the device is not in a discharge condition. , The first The actual charging and discharging power within a time period; the charging and discharging state and power correlation constraint is used to ensure that the energy storage unit cannot charge and discharge simultaneously in the same time period, and the charging and discharging power are controlled by the corresponding state variables respectively. If the energy storage system consists of multiple units, it can be aggregated and modeled, and this constraint can be relaxed to a total power constraint.

[0080] (5) Cycle life and capacity decay constraints The charge-discharge cycle data from the state-of-charge (SOC) curve was extracted using the rainflow counting method to obtain the depth of discharge for each cycle. Then the equivalent total loop count constraint is: ,in The reference depth of discharge (e.g., 80%). γ The fitting index is... This represents the maximum equivalent number of cycles at the reference discharge depth.

[0081] (6) Response time and ramp rate Electrochemical energy storage exhibits extremely fast response speeds; lithium-ion battery energy storage response times are in the millisecond range, with ramp rates reaching 100% of rated power per second. For rapid response requirements such as primary frequency regulation, there is virtually no dynamic process delay. This characteristic is typically treated as a non-inertial element in modeling, i.e.: ,in For lithium batteries, the ramp rate limit is... This constraint is usually much greater than the power change demand, so it can be ignored in hourly scheduling, but it needs to be considered in second- or minute-level frequency modulation simulations.

[0082] S23. Analysis of the Value Differences Between Pumped Storage and Electrochemical Energy Storage Across Multiple Time Scales Based on the duration, response speed, and power change amplitude of the regulation demand, different application scenarios are divided into four time scales: millisecond to second, minute to hour, hour to day, and day to week. In step S23, different application scenarios are divided into four time scales: millisecond to second, minute to hour, hour to day, and day to week, corresponding to primary frequency regulation and inertia support, secondary frequency regulation and rapid ramping, day-ahead peak shaving and renewable energy consumption, and long-term power balance under continuous extreme weather conditions, respectively. Based on the energy storage type and subdivided value dimension contribution matrix calculated in S24, electrochemical energy storage is given priority in undertaking primary frequency regulation and inertia support, secondary frequency regulation and rapid ramping in the millisecond to second and minute to hour application scenarios, while pumped hydro storage is given priority in undertaking day-ahead peak shaving and renewable energy consumption, and long-term power balance under continuous extreme weather conditions in the hour to day and day to week application scenarios.

[0083] Based on the above technical characteristic model, the regulation demand of the power system is decomposed into four levels according to the time scale, and the differentiated value contribution capabilities of pumped storage and electrochemical energy storage in each level are analyzed: (1) Millisecond to second level - inertia support and primary frequency modulation At this timescale, the system needs to respond quickly to frequency deviations and provide instantaneous support for active power. Electrochemical energy storage such as lithium batteries, with their millisecond-level response time and extremely high ramp rate, can fully meet this task and can replace part of the primary frequency regulation reserve capacity of thermal power units. Pumped storage units, due to the mechanical inertia of the turbine unit and the governor response delay (usually several seconds to tens of seconds), contribute relatively little value at this scale.

[0084] (2) Minute-level to hour-level - Secondary frequency modulation and rapid ramping The system needs to track fluctuations in load and net output of renewable energy sources within minutes to hours. Electrochemical energy storage, with its flexible charge-discharge switching capability and high ramp rate, performs excellently in minute-level rapid regulation. Although pumped-storage units have a slower ramp rate, they can undertake continuous regulation tasks from minutes to hours, especially providing continuous and stable reverse regulation when renewable energy sources fluctuate significantly.

[0085] (3) Hourly to daily level - day-ahead peak shaving and renewable energy consumption This scale is the main battleground for energy storage value contribution. Pumped storage power stations can discharge continuously for 6-12 hours, fully covering the peak-valley difference cycle of the daily net load, enabling large-scale intraday transfer of new energy sources. Short-term energy storage such as lithium batteries can cooperate with pumped storage for peak shaving and valley filling within 2-4 hours, playing a pioneering role in peak regulation, quickly responding to steep changes in net load, and providing buffer time for the slow ramp-up of pumped storage.

[0086] (4) Daily to weekly level - long-term power balance during continuous cloudy / windless days Under extreme weather conditions, renewable energy sources may experience low output for several consecutive days, necessitating long-term power balance capabilities. Pumped storage power stations, with their reservoir capacity advantage (capable of continuous full power generation for several days), are irreplaceable in this context. Electrochemical energy storage, due to its limited energy capacity, contributes less value in this scenario; however, if configured with large-scale flow batteries (capable of continuous discharge for 4-8 hours), it can also participate in long-term power balance to some extent.

[0087] S24. Generation of the Differentiated Value Weight Matrix The compatibility between pumped hydro storage and electrochemical energy storage under different application scenarios is calculated to form a contribution matrix of energy storage type and subdivided value dimensions. The contribution matrix serves as the basis for the quantitative calculation of the value weight matrix under different application scenarios.

[0088] Based on the above multi-timescale value analysis results, a contribution matrix A∈R^{M×N} of energy storage type and subdivided value dimensions is constructed, where M represents energy storage types such as pumped storage, lithium-ion batteries, and flow hydrants, and N represents various subdivided value indicators (such as frequency regulation response contribution, peak-valley arbitrage capability, and carbon emission reduction contribution). Matrix elements a_{m,n}∈[0,1] represent the relative contribution factor of the m-th type of energy storage on the nth value indicator. This matrix can serve as a reference basis for subsequent multi-objective optimization and weight adjustment in the final decision-making process.

[0089] In this embodiment, the energy storage technology parameters are set as follows: ① Pumped storage: Overall efficiency 75%, power generation / pumping power ratio 1:1.1, continuous discharge time 8 hours, unit investment 5000 yuan / kW, ramp rate 50% of rated power / minute, lifespan 50 years.

[0090] ② Lithium battery energy storage: comprehensive efficiency 88%, continuous discharge time 2 hours, unit investment 1500 yuan / kWh, ramp rate 100% rated power / second, cycle life 8000 times, calendar life 12 years.

[0091] ③ Flow battery energy storage: overall efficiency 75%, continuous discharge time 4 hours, unit investment 2500 yuan / kWh, ramp rate 100% rated power / second, cycle life 15000 times, calendar life 20 years.

[0092] Based on the adaptability of different energy storage types in scenarios such as peak shaving and power balance, frequency regulation and rapid support, backup and emergency support, and new energy consumption, a preliminary differentiated value weight matrix is ​​constructed, as shown in Table 1.

[0093] Table 1. Examples of differentiated value weight matrices for pumped storage and electrochemical energy storage in different application scenarios.

[0094] This weight matrix provides a technical basis for the dynamic adjustment of the value weights of each dimension of the objective function of the upper layer of the subsequent model. The final configuration scheme will be automatically searched by the two-layer optimization model to find the optimal ratio under the premise of meeting the system operation constraints.

[0095] S3. Establish a two-layer optimization configuration model with the goal of maximizing the comprehensive value of the standardized technical value indicators, economic value indicators, environmental value indicators, and social value indicators obtained in step S1. The two-layer optimization configuration model includes an upper-layer multi-objective capacity configuration optimization model and a lower-layer time-series operation simulation model. The upper-layer multi-objective capacity configuration optimization model generates an initial capacity configuration scheme and sends it to the lower-layer time-series operation simulation model. The lower-layer time-series operation simulation model calculates the actual value contribution and feeds it back to the upper-layer multi-objective capacity configuration optimization model. The upper-layer multi-objective capacity configuration optimization model updates the capacity configuration scheme. Through the closed-loop iteration of multi-objective optimization of the upper-layer multi-objective capacity configuration optimization model and operation simulation verification of the lower-layer time-series operation simulation model, the capacity configuration scheme is corrected and verified. The upper-level multi-objective capacity configuration optimization model uses pumped storage installed capacity, pumped storage reservoir capacity, various electrochemical energy storage power capacity and energy capacity as decision variables. It takes maximizing standardized technical value indicators, economic value indicators, environmental value indicators and social value indicators as multiple objective functions, and sets total investment constraints, site resource constraints and system regulation demand constraints. The lower-level time-series operation simulation model is based on the capacity configuration scheme given by the upper-level multi-objective capacity configuration optimization model. It takes minimizing system operating cost as the objective function, simulates the time-series operation of multiple scenarios throughout the year, calculates the actual value contribution of the four dimensions of technical value indicators, economic value indicators, environmental value indicators and social value indicators, and feeds them back to the upper-level multi-objective capacity configuration optimization model.

[0096] S31, with pumped storage installed capacity C ps Pumped storage capacity E ps Various electrochemical energy storage power capacities C b,i and energy capacity E b,i (i represents different types of electrochemical energy storage) is the upper-level decision variable.

[0097] S32. Establish a multi-objective capacity allocation optimization model at the upper level, with the objective function being to maximize four standardized values: , , ,

[0098] in , , , These are standardized technical value indicators, economic value indicators, environmental value indicators, and social value indicators.

[0099] The constraints include: ①Total investment constraints:

[0100] in This represents the total investment cost of pumped storage hydroelectric power. For the first Total investment cost of electrochemical energy storage This represents the upper limit of the total investment budget for energy storage during the planning period.

[0101] ②Site resource constraints:

[0102] in The maximum installed capacity allowed by the site resources for pumped storage hydroelectric power stations. For the first The maximum allowable power capacity of electrochemical energy storage site resources.

[0103] ③ System regulation demand constraints: The total regulation capacity provided by all types of energy storage must meet the lower limit of the system's peak shaving and frequency regulation requirements.

[0104] in For the first The maximum discharge power of this type of energy storage The system's maximum peak-to-valley difference adjustment capacity.

[0105] ④ System frequency regulation requirement constraints:

[0106] in Provides a capacity conversion factor for frequency regulation for the m-th type of energy storage. This is to meet the system's frequency regulation capacity requirements.

[0107] ⑤ Lower limit constraint on renewable energy consumption rate:

[0108] in The total amount of electricity wasted annually, This represents the theoretically achievable power generation from new energy sources.

[0109] S33. Establish a lower-level time-series operation simulation model. Based on the capacity configuration scheme given by the upper layer, and with the goal of minimizing system operating costs, this embodiment is based on a multi-scenario time-series operation simulation of the system for 8760 hours a year or a typical day, with the objective function being the minimization of system operating costs:

[0110] in For fuel costs, For start-up costs, To account for the cost of power curtailment penalties, This refers to the power that has been abandoned. To incur penalties for load loss, The load shedding power is defined as follows: constraints include power balance constraints, conventional unit output constraints, unit ramping constraints, spinning reserve constraints, line power flow constraints, and constraints of the electrochemical energy storage operation model.

[0111] S34. Iterative Closed Loop Between Upper and Lower Layers. The lower-layer time-series simulation model optimizes the annual energy storage scheduling output based on the scheme of the upper-layer multi-objective capacity configuration optimization model, and returns the actual value contribution of each dimension. The four objective function values ​​F1 to F4 are calculated and fed back to the upper-layer multi-objective capacity configuration optimization model. The capacity configuration scheme is updated accordingly until the convergence condition or the maximum number of iterations is reached.

[0112] Based on the aforementioned value indicators and energy storage operation model, an upper-level multi-objective capacity configuration optimization model and a lower-level time-series operation simulation model are established. The upper-level model performs multi-objective optimization with four independent objectives: maximizing technological value, economic value, environmental value, and social value. Specifically, the total investment limit for the upper-level model is 20 billion yuan, with site resource constraints limiting pumped storage to a maximum of 2 million kilowatts, lithium batteries to a maximum of 3 million kilowatts / 6 million kilowatt-hours, and flow batteries to a maximum of 1.5 million kilowatts / 6 million kilowatt-hours. The lower-level model uses eight typical days to represent the annual operation process, i.e., selecting one working day and one rest day per quarter, with a time resolution of 1 hour.

[0113] The lower-level model, based on the capacity configuration scheme given by the upper-level model, simulates the charging and discharging processes of pumped hydro storage, lithium battery energy storage, and flow battery energy storage at various time periods. It calculates the renewable energy absorption rate, abandoned power, power shortage, operating costs, and actual value indicators across various dimensions, feeding these calculations back to the upper-level model for iterative correction. Through closed-loop iteration between the upper and lower levels, the overall value can be maximized at the capacity configuration level, while the operational feasibility of the scheme is verified at the operational simulation level. The structure diagram of the two-layer optimized configuration model is shown below. Figure 3 As shown.

[0114] S4. Dynamic weight adaptive adjustment mechanism for decision-making A dynamic weight adjustment mechanism is introduced to adaptively adjust the weights of the four dimensions of technical value indicators, economic value indicators, environmental value indicators, and social value indicators based on policy factors and technical cost factors, thereby obtaining the dynamic weights corresponding to the four dimensions.

[0115] After obtaining the Pareto nondominated solution set for multi-objective optimization, the weights of each value dimension in the final decision can be dynamically adjusted according to current policy orientation and technological development trends. The following dynamic weight adjustment model is adopted:

[0116] in Based on weights, Policy change factors (such as changes in carbon peaking progress and energy storage subsidy levels). This is a factor affecting the change in technology costs (such as the rate of decrease in the cost of electrochemical energy storage). , To adjust the sensitivity coefficient, it can be set to 0.5. The corrected weights will be directly used for TOPSIS integrated decision-making in S5.

[0117] In this embodiment, the upper-level model employs the MOPSO algorithm for multi-objective optimization, aiming to maximize four independent objectives: technological value, economic value, environmental value, and social value. The algorithm parameters are set as follows: population size 50, 100 iterations, inertia weight 0.7, and both individual and global learning factors are 1.5. Through multiple iterations, a uniformly distributed Pareto non-dominated solution set is obtained, containing 35 non-dominated capacity configuration schemes.

[0118] After obtaining the Pareto non-dominated solution set, based on the dynamic weight adjustment mechanism described in S4, and combined with the province's carbon peaking target and energy storage development plan, the basic weight vector is set to [0.25, 0.35, 0.25, 0.15], corresponding to technical value, economic value, environmental value, and social value, respectively. The TOPSIS method is used to select the optimal solution with comprehensive value from the Pareto solution set. In the long-term low-carbon scenario, to reflect the impact of the strengthened carbon emission reduction target, the basic weight of environmental value is increased to 0.40, and the weight vector is adjusted to [0.20, 0.25, 0.40, 0.15]. TOPSIS screening is then performed again to ensure that the capacity allocation results can respond to changes in low-carbon development goals and policy guidance. The flowchart of the solution algorithm based on MOPSO-TOPSIS is shown below. Figure 4 As shown.

[0119] S5. Model Solving and Scheme Optimization Based on Hybrid Intelligent Algorithms Multi-objective intelligent optimization algorithms, including Multi-Objective Particle Swarm Optimization (MOPSO) or Non-Dominated Sorting Genetic Algorithm with Elite Strategy (NSGA-II), are employed to solve the upper-level multi-objective capacity configuration optimization model, yielding a uniformly distributed Pareto non-dominated solution set. Based on this, and incorporating adaptively corrected dynamic weights, the TOPSIS method is used. The weights, dynamically corrected by S4 or objective weights determined by the information entropy weighting method, are used as the importance of each value dimension to select the capacity configuration scheme with the optimal overall value from the Pareto non-dominated solution set. Finally, sensitivity analysis is performed on key parameters such as new energy penetration rate, energy storage cost reduction rate, and carbon price to verify the robustness of the configuration scheme, and a synergistic capacity configuration scheme combining pumped hydro storage and electrochemical energy storage is output.

[0120] In this embodiment, to demonstrate the effectiveness of the multi-dimensional value-optimal capacity configuration scheme (Scheme A) of the present invention, separate planning schemes (Scheme B) and only economically optimal collaborative configuration schemes (Scheme C) are established for comparison. A comparison diagram of the configuration results of a local power grid example in this embodiment is shown below. Figure 5 Among them, Scheme A is the optimal scheme selected from the Pareto non-dominated solution set using the TOPSIS method combined with basic weights.

[0121] Under the basic weighted scenario, the optimal configuration of Scheme A is as follows: pumped hydro storage 1.6 million kW / 12.8 million kWh; lithium battery energy storage 1.2 million kW / 2.4 million kWh; flow battery energy storage 0.6 million kW / 2.4 million kWh. Under this scheme, the technological value score is 0.82, increasing the renewable energy absorption rate from 89% to 96%; the economic value score is 0.78, with a positive net present value over the entire life cycle and an investment payback period of approximately 12 years; the environmental value score is 0.85, with an annual carbon reduction of approximately 2.8 million tons; and the social value score is 0.72, creating approximately 3,500 jobs.

[0122] After obtaining the standardized scores for each scheme across the four dimensions, the TOPSIS method described in S5 was used to calculate the comprehensive proximity of each scheme as the comprehensive value score using the basic weight vector. Scheme A has a comprehensive value score of 0.795. Scheme B is a separate capacity configuration scheme, with a configuration of 1.8 million kW pumped storage, 1.5 million kW lithium-ion batteries, and 0 kW flow hydrant, resulting in a comprehensive value score of 0.713. Scheme C is a scheme with only economically optimal coordinated configuration, with a configuration of 1.5 million kW pumped storage, 1.4 million kW lithium-ion batteries, and 0 kW flow hydrant, resulting in a comprehensive value score of 0.751.

[0123] The scores for each dimension mentioned above are calculated using the quantitative model constructed in this invention (step one) and the results of the operational simulation (step three, lower level) after normalization. Specifically, the technological value is weighted by the frequency regulation contribution and the absorption improvement rate; the economic value is mapped by the net present value over the entire life cycle according to the ideal value ratio; the environmental value is mapped by the carbon reduction amount according to the maximum emission reduction potential; and the social value is normalized by the input-output multiplier and the employment coefficient. The comparison results show that Scheme A has a higher overall value score than Schemes B and C, indicating that this invention, through multi-dimensional value assessment and synergistic optimization, can achieve a better balance between economic efficiency, technological adaptability, carbon emission reduction, and social benefits. Scheme B scores lower, mainly because separate planning easily leads to capacity redundancy and insufficient synergy between energy storage types; although Scheme C has better economic efficiency, it does not adequately consider environmental value and long-term regulation capacity, resulting in a lower overall performance than Scheme A.

[0124] In the long-term low-carbon scenario, the environmental value weight is increased to 0.40, and TOPSIS screening is conducted again. The calculation results show that the capacity of flow battery energy storage configuration increases by about 25%, the capacity of pumped hydro storage configuration increases by about 10%, and the annual carbon reduction increases to about 3.5 million tons, indicating that the dynamic weighting mechanism can adjust the energy storage configuration structure according to changes in policy objectives.

[0125] Further sensitivity analysis revealed that, under conditions of ±20% change in new energy penetration rate, ±5% change in annual energy storage cost reduction rate, and carbon price variation range of 50-120 yuan / ton, the value fluctuation of each dimension of the configuration scheme obtained by the method of this invention is less than 10%. The results indicate that this capacity configuration method has strong robustness and scenario adaptability.

[0126] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such modifications and substitutions should be considered within the scope of protection of the present invention.

Claims

1. A method for synergistic capacity configuration of pumped hydro storage and electrochemical energy storage based on multi-dimensional value assessment, characterized in that, Includes the following steps S1. Obtain basic data on energy output, load, electricity price, ancillary service price, carbon price, energy storage investment and operation and maintenance cost, site resources, planning constraints and policy parameters of the power system to be planned, and construct a multi-dimensional value assessment index system, which includes technical value index, economic value index, environmental value index and social value index. The technical value indicators, economic value indicators, environmental value indicators, and social value indicators are quantified and standardized respectively. The standardization process adopts the range standardization method to map the technical value indicators, economic value indicators, environmental value indicators, and social value indicators to a unified dimension range to obtain standardized value indicators. S2. Based on the operating characteristics of pumped storage and electrochemical energy storage, establish operating constraint models for pumped storage and electrochemical energy storage respectively. Divide different application scenarios according to the duration of regulation demand, response speed and power change amplitude. Calculate the adaptability of pumped storage and electrochemical energy storage in different application scenarios. Arrange the adaptability according to two dimensions: energy storage type and application scenario to form a differentiated value contribution matrix. Based on the differentiated value contribution matrix, use the analytic hierarchy process (AHP) to form the value weight matrix for different application scenarios. S3. Establish a two-layer optimization configuration model with the goal of maximizing the comprehensive value of the standardized technical value indicators, economic value indicators, environmental value indicators and social value indicators obtained in step S1. The two-layer optimization configuration model includes an upper-layer multi-objective capacity configuration optimization model and a lower-layer time-series operation simulation model. The upper-layer multi-objective capacity configuration optimization model generates an initial capacity configuration scheme and sends it to the lower-layer time-series simulation model. The lower-layer time-series simulation model calculates the actual value contribution and feeds it back to the upper-layer multi-objective capacity configuration optimization model. The upper-layer multi-objective capacity configuration optimization model updates the capacity configuration scheme. Through the closed-loop iteration of multi-objective optimization by the upper-layer multi-objective capacity configuration optimization model and simulation verification by the lower-layer time-series simulation model, the capacity configuration scheme is corrected and verified. S4. Introduce a dynamic weight adjustment mechanism to adaptively adjust the weights of the four dimensions of technology value indicators, economic value indicators, environmental value indicators and social value indicators based on policy factors and technology cost factors, so as to obtain the dynamic weights corresponding to the four dimensions. S5. The upper-level multi-objective capacity configuration optimization model is solved using a multi-objective intelligent optimization algorithm. After obtaining the Pareto non-dominated solution set, the TOPSIS method is used to select the capacity configuration scheme with the best comprehensive value from the Pareto non-dominated solution set, and the pumped storage and electrochemical energy storage synergistic capacity configuration scheme is output.

2. The method for synergistic capacity configuration of pumped hydro storage and electrochemical energy storage based on multi-dimensional value assessment as described in claim 1, characterized in that, In step S1, the technical value indicators include the contribution of frequency regulation response speed, the increase in renewable energy consumption, and the improvement in power supply reliability. The contribution of frequency regulation response speed is quantified based on the ratio of the actual power regulation provided by energy storage to the power regulation required by the system. The increase in renewable energy consumption is quantified based on the difference in power curtailment before and after energy storage configuration. The improvement in power supply reliability is quantified based on the difference in system load shedding power before and after energy storage configuration, combined with the unit outage loss cost. The economic value indicators include net benefits over the entire life cycle. The environmental value indicators include carbon emission reduction value. The social value indicators include the industrial chain driving effect and the employment driving effect.

3. The method for synergistic capacity configuration of pumped hydro storage and electrochemical energy storage based on multi-dimensional value assessment as described in claim 2, characterized in that, The contribution of the frequency modulation response speed is quantified by the following formula: in For the first The actual power regulation provided by energy storage during each time period This represents the power regulation required by the system. The value coefficient per unit frequency modulation. This represents the total number of time periods within the scheduling cycle. The increase in new energy consumption is quantified by the following formula: in and These are the first and second energy storage configurations before and after. The amount of power abandoned during each time period For time step; The power supply reliability improvement index is quantified by the following formula: In the formula and These represent the system load shedding power before and after configuring energy storage, respectively. Cost of power outage losses per unit.

4. The method for synergistic capacity configuration of pumped hydro storage and electrochemical energy storage based on multi-dimensional value assessment according to claim 2, characterized in that, The net income over the entire life cycle of the economic value indicator is calculated using the following formula: in For the first Annual peak-valley arbitrage profit is calculated based on the price difference of energy storage charging and discharging under time-of-use electricity pricing. For ancillary service revenue, compensation is calculated based on frequency regulation mileage and capacity. To compensate for the capacity market; To delay the return on investment in power transmission and distribution; Annual maintenance costs; This refers to the initial investment cost; The discount rate; The project's lifespan; The environmental value indicator, carbon emission reduction value, is calculated using the following formula: in For the first The amount of thermal power that can be replaced by energy storage during a given period. Carbon emission factors from thermal power plants For carbon trading prices, This represents the total number of time periods within the scheduling cycle. The industrial chain pulling effect in the aforementioned social value indicators is calculated using the following formula: In the formula For the first The total added value factor per unit investment for energy storage-like products. For the first The initial total investment for energy storage-like systems; The employment-driving effect in the aforementioned social value indicators is calculated using the following formula: In the formula The social value conversion coefficient of a unit of employment position. For the first The employment coefficient during the construction period of energy storage-like devices. For the first The employment coefficient during the operation period of energy storage-like systems. For the first Installed capacity of energy storage systems For the first The operational lifespan of energy storage systems.

5. The method for synergistic capacity configuration of pumped hydro storage and electrochemical energy storage based on multi-dimensional value assessment according to claim 1, characterized in that, Step S2 specifically includes: S21. Establish a pumped storage operation constraint model, which includes power generation constraints, power generation ramping constraints, pumping power constraints, mutually exclusive operating conditions constraints, equivalent energy storage capacity constraints of the reservoir, energy storage recursive balance equations, reservoir capacity boundary constraints at the end of the scheduling cycle, and start-stop frequency constraints. S22. Establish an electrochemical energy storage operation constraint model, which includes charge and discharge power constraints, state of charge constraints, state of charge recursive equations, charge and discharge state-power correlation constraints, cycle lifetime constraints, and ramp rate constraints. S23. Based on the duration of the adjustment demand, the response speed, and the magnitude of power change, different application scenarios are divided into four time scales: millisecond to second, minute to hour, hour to day, and day to week. S24. Calculate the compatibility of pumped storage and electrochemical energy storage under different application scenarios to form a contribution matrix of energy storage type and subdivided value dimensions. The contribution matrix serves as the basis for the quantitative calculation of the value weight matrix under different application scenarios.

6. The method for synergistic capacity configuration of pumped hydro storage and electrochemical energy storage based on multi-dimensional value assessment according to claim 5, characterized in that, The pumped storage operation constraint model described in step S21 includes the following constraints: Power generation constraint: and power generation ramp-up constraints : ,in 、 These are the rated maximum power output and the technical minimum power output, respectively. 、 Limits for upward and downward ramp rates during power generation; Pumping power constraint: ,in , These are the rated maximum and technical minimum pumping power, respectively. Mutually exclusive operating conditions: Constrained by formula: ,in , A 0 / 1 state variable, Indicates the first The unit is in power generation mode during each period. This indicates that the system is not in power generation mode. Indicates the first During certain periods, the unit is in pumping mode. This indicates that the pumping unit is not in pumping mode; the mutual exclusion constraint condition of the operating conditions is used to ensure that the pumped storage unit can only be in one of the following states at the same time: power generation, pumping, or shutdown, and simultaneous power generation and pumping are prohibited. Equivalent energy storage capacity of reservoir constraint: And the recursive equilibrium equation for energy storage: ,in, For the first The equivalent energy storage capacity of the upper reservoir within a certain time period For the first The equivalent energy storage capacity of the upper reservoir within a certain time period For maximum energy storage capacity, For minimum storage capacity, , These are pumping efficiency and power generation efficiency, respectively. Their product is the overall cycle efficiency of the pumped storage power station. Storage capacity boundary constraints at the end of the scheduling cycle: ,in For the scheduling period, This is the allowable deviation amount; Start-stop count constraint formula: ,in This represents the maximum number of state transitions allowed within the scheduling period. , Representing the first The time period and the first The power generation operating status for each time period. , Representing the first The time period and the first The pumping operation status during each time period. and These represent the initial power generation and initial pumping operation status, respectively; The electrochemical energy storage operation model described in S22 includes the following constraints: definition The net output power of energy storage is represented by positive values ​​indicating charging and negative values ​​indicating discharging; a non-negative auxiliary variable is introduced. , ,satisfy ,and , ,in , This is the maximum charge / discharge power; State of charge constraint: ,in For the first State of charge at each time period, This is the limit of the state of charge. This represents the upper limit of the state of charge. State of charge Recurrence equation: ,in For the first State of charge at each time period, For the first State of charge at each time period, , These represent charging and discharging efficiencies, respectively. , The first Actual charging and discharging power during each time period Rated energy capacity, For time step; Charge / discharge state and power correlation constraints: , , ,in , A 0 / 1 state variable, Indicates the first During each period, the energy storage is in charging mode. This indicates that the device is not in charging mode. Indicates the first During the period when the energy storage is in discharge mode, This indicates that the device is not in a discharge state. , The first The actual charging power and discharging power within a time period; the charging and discharging state and power correlation constraints are used to ensure that the energy storage unit cannot be charged and discharged simultaneously in the same time period, and the charging and discharging power are controlled by the corresponding state variables respectively. Cycle lifetime constraint: The charge-discharge cycles in the state-of-charge curve are extracted using the rainflow counting method to obtain the discharge depth of each cycle. Then the equivalent total loop count constraint is: ,in As the reference discharge depth, γ The fitting index is... This represents the maximum equivalent number of cycles at the reference discharge depth. Climbing speed constraint: ,in This is the limit for the climbing speed.

7. The method for synergistic capacity configuration of pumped hydro storage and electrochemical energy storage based on multi-dimensional value assessment according to claim 5, characterized in that, In step S23, different application scenarios are divided according to four time scales: millisecond to second, minute to hour, hour to day, and day to week. These correspond to primary frequency regulation and inertia support, secondary frequency regulation and rapid ramping, day-ahead peak shaving and renewable energy consumption, and long-term power balance under continuous extreme weather conditions. Based on the energy storage type and subdivided value dimension contribution matrix calculated by S24, electrochemical energy storage takes priority in supporting primary frequency regulation and inertia, secondary frequency regulation and rapid ramping in millisecond to second-level and minute-level to hour-level application scenarios. Pumped hydro storage takes priority in supporting day-ahead peak shaving and new energy consumption, and long-term power balance under continuous extreme weather conditions in hour-level to day-level and day-level to week-level application scenarios.

8. According to claim 1, the method for coordinated capacity configuration of pumped storage and electrochemical energy storage based on multi-dimensional value assessment, in step S3, the upper-level multi-objective capacity configuration optimization model takes pumped storage installed capacity, pumped storage reservoir capacity, various electrochemical energy storage power capacity and energy capacity as decision variables, and takes maximizing standardized technical value indicators, economic value indicators, environmental value indicators and social value indicators as multiple objective functions, and sets total investment constraints, site resource constraints and system regulation demand constraints; The lower-level time-series operation simulation model is based on the capacity configuration scheme given by the upper-level multi-objective capacity configuration optimization model. With the minimization of system operating cost as the objective function, it simulates the time-series operation of multiple scenarios throughout the year, calculates the actual value contribution of four dimensions: technical value indicators, economic value indicators, environmental value indicators, and social value indicators, and feeds them back to the upper-level multi-objective capacity configuration optimization model. Step S3 specifically includes: S31. Pumped storage installed capacity Cps, pumped storage capacity Eps, various electrochemical energy storage power capacity Cb,i and energy capacity Eb,i are the upper-level decision variables. S32. Establish a multi-objective capacity allocation optimization model at the upper level, with the objective function being to maximize four standardized values: ， ， ， in , , , These are standardized technical value indicators, economic value indicators, environmental value indicators, and social value indicators. The constraints include: Total investment constraints: ,in This represents the total investment cost of pumped storage. For the first Total investment cost of electrochemical energy storage This represents the upper limit of the total investment budget for energy storage during the planning period; Site resource constraints: , ,in The maximum installed capacity allowed by the site resources for pumped storage hydroelectric power stations. For the first The maximum allowable power capacity of electrochemical energy storage site resources; System peak shaving demand constraints: ,in For the first The maximum discharge power of this type of energy storage Adjust the system's maximum peak-to-valley difference to meet demand capacity requirements; System frequency regulation requirement constraints: ,in For the first Energy storage-like systems provide a capacity conversion factor for frequency regulation. To meet the system's frequency regulation capacity requirements; Lower limit constraint on renewable energy consumption rate: ,in The total amount of electricity wasted annually, This represents the theoretically achievable power generation of new energy sources. S33. Establish a lower-level time-series operation simulation model. Based on the capacity configuration scheme given by the upper level, the objective function is to minimize the system operating cost. in For fuel costs, For start-up costs, To cover the cost of power curtailment penalties, This refers to the power that has been abandoned. To incur penalties for load loss, The load shedding power is defined by the constraints, which include power balance constraints, conventional unit output constraints, unit ramping constraints, spinning reserve constraints, line power flow constraints, and constraints of the electrochemical energy storage operation model. S34. The lower-level time-series operation simulation model optimizes the annual energy storage scheduling output based on the scheme of the upper-level multi-objective capacity configuration optimization model, and returns the actual value contribution of each dimension. The four objective function values ​​F1 to F4 are calculated and fed back to the upper-level multi-objective capacity configuration optimization model. The capacity configuration scheme is updated accordingly until the convergence condition or the maximum number of iterations is reached.

9. The method for synergistic capacity configuration of pumped hydro storage and electrochemical energy storage based on multi-dimensional value assessment according to claim 1, characterized in that, The dynamic weight adjustment mechanism described in step S4 corrects the weights using the following formula: in Based on weights, As a policy change factor, As a factor of change in technology costs, , To adjust the sensitivity coefficient, These correspond to the weights of technological, economic, environmental, and social value, respectively; the revised weights are used for the TOPSIS comprehensive decision in step S5.

10. The method for synergistic capacity configuration of pumped hydro storage and electrochemical energy storage based on multi-dimensional value assessment according to claim 1, characterized in that, The multi-objective intelligent optimization algorithm mentioned in step S5 includes the multi-objective particle swarm optimization algorithm MOPSO or the non-dominated sorting genetic algorithm NSGA-II with an elite strategy; after obtaining the Pareto non-dominated solution set, the TOPSIS method is used, with the weights adaptively corrected in step S4 or the objective weights determined by the information entropy weight method as the importance of the value index dimension, to select the capacity configuration scheme with the best comprehensive value from the Pareto non-dominated solution set.

Images