A two-stage collaborative optimization scheduling method for distribution network with distributed pumped storage

Abstract

The application provides a kind of two-stage coordinated optimization scheduling method of distribution network containing distributed pumped storage, and relates to the technical field of coordinated operation of distribution network source network load storage.Through S1: considering the water-electric coupling characteristics of DPHS, constructing its equivalent state of charge SOC and dynamic mathematical model of reservoir capacity, and the multi-source joint scheduling model of wind power, photovoltaic, small hydropower output multi-source output model;S2: multi-objective optimization system of day-ahead-intra-day stage scheduling is adopted, multi-dimensional constraint conditions are comprehensively considered, and the feasibility of scheduling scheme in physics and economy is improved;S3: the stage scheduling model is solved by alternating direction multiplier method ADMM, the coordinated operation of source-network-load-storage is realized, the operation cost is reduced, the new energy curtailment rate is reduced, the new energy consumption capacity is improved, the system regulation flexibility is improved, the power supply reliability is improved, and the total system operation cost is reduced.

Description

Technical Field

[0001] This invention relates to the field of coordinated operation technology of power generation, grid, load and storage in distribution networks, and in particular to a two-stage collaborative optimization scheduling method for distribution networks including distributed pumped storage. Background Technology

[0002] With the deepening of the global energy transition strategy and the gradual implementation of dual-carbon targets, the penetration rate of renewable energy sources such as wind and solar power in the power distribution network continues to increase, while also posing severe challenges to the safe and stable operation of the power system. The significant increase in uncertainty on both the source and load sides has led to insufficient system regulation resources, frequent wind and solar curtailment, and continuously rising operating costs. At the distribution network level, traditional regulation methods are no longer sufficient to cope with the operational pressure brought about by the high proportion of renewable energy integration.

[0003] In existing technologies, energy storage technology serves as an effective means to mitigate fluctuations and improve system flexibility. This includes: 1. Traditional pumped storage power plants, which offer large capacity, long lifespan, and good economic efficiency. However, traditional centralized pumped storage power plants have long construction cycles, high investment costs, and are limited by geographical conditions, making it difficult to fully cover the distributed regulation needs at the distribution network level. 2. Distributed pumped storage offers flexible operation and rapid start-up and shutdown, serving as an important auxiliary regulation means. Coordinating distributed pumped storage with small hydropower to construct a multi-energy complementary and coordinated optimized operating system at the distribution network level can enhance the system's ability to accommodate high proportions of renewable energy and improve its economic efficiency.

[0004] However, current scheduling methods have significant shortcomings in time-scale coordination: First, most studies only conduct day-ahead or real-time scheduling independently, lacking an effective day-ahead-intra-day coordination mechanism. The day-ahead plan and real-time operation are poorly integrated, failing to address real-time fluctuations in wind, solar, and load conditions in a closed-loop manner. Second, they fail to accurately characterize the bidirectional water-electricity coupling characteristics of distributed pumped storage hydroelectric power generation, the dynamic changes in reservoir capacity, and the evolution of the equivalent state of charge (SOC), resulting in significant deviations between the scheduling model and physical reality. Third, they do not incorporate rolling optimization and distributed solution algorithms, making it difficult to handle high-dimensional variables, multiple constraints, and strong uncertainties, leading to poor feasibility of scheduling schemes and delayed regulatory responses. Fourth, there is insufficient coordination among multiple objectives: the goals of minimizing operating costs, maximizing renewable energy consumption, and ensuring system stability are not comprehensively considered, making it difficult to balance economic efficiency with clean energy utilization. Summary of the Invention

[0005] The main objective of this invention is to provide a two-stage collaborative optimization scheduling method for power distribution networks containing distributed pumped storage, which solves the technical problems of fragmented time scales, insufficient coupled modeling, low regulation and operation efficiency, and limited renewable energy absorption capacity.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is: a two-stage collaborative optimization scheduling method for a distribution network including distributed pumped storage, comprising the following steps: S1. Model Construction: Construct a multi-source joint scheduling model that includes wind power, photovoltaic power, small hydropower and distributed pumped storage. The multi-source joint scheduling model includes: DPHS water-electricity coupling characteristics, equivalent state of charge (SOC) and reservoir capacity dynamic mathematical model of distributed pumped storage, and multi-source output model of wind power, photovoltaic power and small hydropower output. S2, Phased Scheduling: The first phase aims to minimize system operating costs and generates a baseline scheduling scheme based on predictive data; the second phase adopts Model Predictive Control (MPC) rolling optimization, divides the control cycle according to a preset time granularity, and executes rolling closed-loop control with the goal of maximizing system power generation. S3. Model Solving: The Alternating Directional Multiplier Method (ADMM) is used to solve the staged scheduling model in a distributed manner, satisfying power balance, reservoir capacity constraints, output constraints and grid transmission constraints, and realizing the coordinated operation of source-grid-load-storage.

[0007] In the preferred embodiment, the DPHS water-electric coupling characteristics include: The pumping channel is for water storage in the upper reservoir, and the pump is driven by an electric motor. The expression is: (1); The power generation channel involves releasing water from the upper reservoir to drive a turbine to generate electricity. The expression is as follows: (2); in: (3); (4); In the formula: The density of water; It is the acceleration due to gravity; This refers to the inflow rate of the upper reservoir; This refers to the outflow rate from the upper reservoir. Losses along the route; For motor efficiency; The pumping efficiency of the water pump; This is the friction loss factor. This refers to the length of the pipe. The diameter of the water pipe; It is the Reynolds number; This is the roughness coefficient of the pipe.

[0008] In the preferred scheme, constraints are set based on the water-electricity coupling characteristics, including: 1) Capacity constraints of upstream and downstream reservoirs: (13); In the formula: For distributed pumped storage in a given time period Internal pumping power; For distributed pumped storage in a given time period Power generation capacity within the unit; and These are the intraday scheduling time slot number and the total number of intraday scheduling time slots, respectively. For the day before Long duration; 2) Global power balance constraints: (14); In the formula: For distributed pumped storage in a given time period Internal pumping power; For the time period The small hydropower output within; for Photovoltaic power output during the time period; for Wind power output during a given time period; For distributed pumped storage in a given time period Power generation capacity within the unit; For the system in time period The load power borne by the internal components.

[0009] In the preferred embodiment, the distributed pumped storage SOC model includes: The equivalent state of charge (SOC) model is as follows: (5); In the formula: for The effective storage capacity of the reservoir should be maintained at all times. , These are the minimum and maximum operating capacity limits, respectively. The dynamic storage capacity satisfies the mass conservation equation as follows: (6); In the formula: This refers to the inflow rate of the upper reservoir; This refers to the outflow rate of the upper reservoir.

[0010] In the preferred embodiment, the multi-source power output model includes: The wind power generation model is expressed as follows: (7); In the formula: To cut in wind speed; To cut off the wind speed; This refers to the rated power of the fan; Rated wind speed; The photovoltaic power generation model is expressed as follows: (8); In the formula: This represents the rated power of the photovoltaic system under reference conditions. This represents the actual solar irradiance. For reference irradiance; This refers to the actual temperature of the photovoltaic module. This is the temperature coefficient (usually taken as -0.4% / ℃). For reference temperature; The output model for small hydropower is expressed as follows: (9); In the formula: For the power generation efficiency of the water turbine; For the upper reservoir Constant water flow rate; For water purification.

[0011] In the preferred scheme, in step S2, the first stage, with minimizing the system operating cost as the objective function, is expressed as: (10); in, (11); In the formula: This represents the minimum operating cost during the day-ahead scheduling period. The price is for pumping electricity; and These are the intraday scheduling period number and the total number of intraday scheduling periods, respectively, where T = 24 hours. For the day before The period is long, 1 hour; For distributed pumped storage in a given time period Internal pumping power; For small hydropower grid connection tariffs; For the time period The small hydropower output within; The cost of penalizing the abandonment of wind and solar power; The stipulated wind and solar energy absorption ratio; The actual proportion of new energy consumed within one day; for Photovoltaic power output during the time period; for Wind power output during a given time period; For distributed pumped storage in a given time period Power generation capacity within the unit; For the system in time period The load power borne by the internal components.

[0012] In the preferred scheme, the constraints set in the first stage include the upper limit of grid transmission power, the upper limit of wind and solar power output, the upper limit of small hydropower output, and DPHS reservoir capacity and power constraints, specifically: 1) Power transmission limit constraints in the power grid: (15); In the formula: Photovoltaic output power; This refers to the output power of wind power. This represents the maximum capacity of wind-solar hybrid power generation that the power grid can accept. 2) Constraints on wind and solar power output: (16); In the formula: for Maximum wind power output during a given time period; for Maximum photovoltaic output during the specified time period; 3) Output constraints of small hydropower stations: (17); In the formula: for The maximum output of small hydropower stations during certain periods.

[0013] In the preferred embodiment, step S2, the first stage, further includes: setting a threshold for the absorption ratio, and introducing a penalty cost when the actual absorption ratio is lower than the threshold.

[0014] In the preferred scheme, in step S2, the second-stage scheduling takes maximizing the total power generation of the system as the objective function, satisfying the real-time power balance constraint. The expression of the objective function is: (12); In the formula: This represents the maximum power generation during the daily dispatch period. and These are the intraday scheduling period number and the total number of intraday scheduling periods, respectively, where T = 24 hours. Within the day The duration is long, 15 minutes; and The first day Measured output values ​​of wind power and photovoltaic power during the time period; For the day within Small hydropower output value within a time period; For the day within Distributed pumped storage power generation capacity within a given time period.

[0015] In the preferred embodiment, the real-time power balance constraint is expressed as: (18); In the formula: For the day within Distributed pumped storage pumping power within a time period; For the day within Actual electricity load during the time period.

[0016] This invention provides a two-stage collaborative optimization scheduling method for a distribution network including distributed pumped storage (DPHS). It considers the water-electricity coupling characteristics of DPHS, constructs a dynamic mathematical model of its equivalent state of charge (SOC) and reservoir capacity, and establishes a multi-source joint scheduling model based on the output models of wind power, photovoltaic power, and small hydropower. This model achieves source-storage collaborative optimization through the spatiotemporal complementarity of multiple sources. Furthermore, it employs a multi-objective optimization system for day-ahead to intraday staged scheduling, comprehensively considering multi-dimensional constraints to improve the physical and economic feasibility of the scheduling scheme. Finally, it uses the Alternating Direction Multiplier Method (ADMM) to solve the staged scheduling model in a distributed manner, achieving coordinated operation of source-grid-load-storage, reducing operating costs, decreasing renewable energy curtailment, improving renewable energy absorption capacity, enhancing system regulation flexibility, improving power supply reliability, and lowering the total system operating cost. Attached Figure Description

[0017] The present invention will be further described below with reference to the accompanying drawings and embodiments: Figure 1 This is a schematic diagram of the overall system operation architecture of the present invention; Figure 2 This is a basic framework diagram of the two-stage coordinated regulation of the present invention, from day to day. Figure 3 This is the improved IEEE 30-node topology diagram of this invention; Figure 4 This is a time-series diagram showing the changes in reservoir water level and DPHS output according to the present invention; Figure 5 This invention provides day-ahead forecast curves for wind power, photovoltaic power, small hydropower, and load. Figure 6 This invention is a time-series distribution diagram of wind and solar power, small hydropower, load power, and time-of-use electricity pricing; Figure 7 This is a diagram illustrating the intraday rolling control ADMM convergence process of this invention; Figure 8 This is a graph showing the intraday rolling control results of the wind-solar-hydro-storage synergy of the present invention; Figure 9 This is a comparison chart of new energy consumption under scenarios A and C of this invention; Figure 10This is a comparison diagram of DPHS rolling control under scenario B and scenario C of the present invention. Detailed Implementation

[0018] Example 1 like Figures 1-10 As shown, a two-stage collaborative optimization scheduling method for a distribution network including distributed pumped storage includes the following steps: S1. Model Construction: Construct a multi-source joint scheduling model that includes wind power, photovoltaic, small hydropower and distributed pumped storage. The multi-source joint scheduling model includes: DPHS water-electricity coupling characteristics, equivalent state of charge (SOC) and reservoir capacity dynamic mathematical model, and multi-source output model of wind power, photovoltaic and small hydropower output.

[0019] S2. Phased scheduling: During the daytime phase, the goal is to minimize system operating costs, and a baseline scheduling scheme is generated based on forecast data. During the intraday phase, model predictive control (MPC) is used for rolling optimization. The control cycle is divided according to a preset time granularity, and rolling closed-loop control is executed with the goal of maximizing system power generation.

[0020] S3. Model Solution: The Alternating Direction Method of Multipliers (ADMM) is used to solve the staged scheduling model in a distributed manner, satisfying power balance, reservoir capacity constraints, output constraints and grid transmission constraints, and realizing coordinated operation of source-grid-load-storage.

[0021] This embodiment analyzes the bidirectional coupling and conversion law of distributed pumped storage (DPHS) from the perspective of physical mechanisms. Considering the water-electricity coupling characteristics of DPHS, it constructs a dynamic mathematical model of its equivalent state of charge (SOC) and reservoir capacity, as well as a multi-source joint scheduling model of multiple power sources including wind power, photovoltaic power, and small hydropower. The source-storage synergistic optimization is achieved through the spatiotemporal complementarity of multiple power sources. Then, a multi-objective optimization system of day-ahead-intraday staged scheduling is adopted, which comprehensively considers multi-dimensional constraints and improves the physical and economic feasibility of the scheduling scheme. Finally, the Alternating Direction Multiplier Method (ADMM) is used to solve the staged scheduling model in a distributed manner, realizing the coordinated operation of source-grid-load-storage, reducing operating costs, reducing the curtailment rate of new energy, improving the absorption capacity of new energy, improving the system regulation flexibility, improving power supply reliability, and reducing the total system operating cost.

[0022] This embodiment is based on distributed pumped storage regulation. Addressing the shortcomings of existing technologies, it analyzes the bidirectional coupling and conversion law of "pumping-power generation" in distributed pumped storage from a physical mechanism perspective. The core architecture is as follows: Figure 1As shown, the multi-power joint dispatch model mainly includes distributed pumped storage, small hydropower, wind power and photovoltaic power. It is interconnected with the distribution network through tie lines to realize the coordination of multiple elements such as source-grid-load-storage.

[0023] I. Construction of Multi-Power Supply Joint Scheduling Model 1) DPHS water-electric coupling characteristics The energy conversion process of the "pumping-power generation" dual-channel mode of distributed pumped storage in this embodiment can be described by the following model.

[0024] In the preferred embodiment, the water-electric coupling characteristics of DPHS include: The pumping channel is for water storage in the upper reservoir, and the pump is driven by an electric motor. The expression is: (1); The power generation channel involves releasing water from the upper reservoir to drive a turbine to generate electricity. The expression is as follows: (2); (3); (4); in: In the formula: The density of water; It is the acceleration due to gravity; This refers to the inflow rate of the upper reservoir; This refers to the outflow rate from the upper reservoir. Losses along the route; For motor efficiency; The pumping efficiency is 0.78 (typical value). This is the friction loss factor. This refers to the length of the pipe. The diameter of the water pipe; It is the Reynolds number; This is the roughness coefficient of the pipe.

[0025] 2) Distributed pumped storage SOC model In the preferred scheme, pumped hydro storage is implemented when renewable energy sources are abundant, and hydropower generation is achieved when renewable energy sources are insufficient. To quantitatively describe the energy storage state of distributed pumped hydro storage, the equivalent state of charge (SOC) model is defined as follows: (5); In the formula: for The effective storage capacity of the reservoir should be maintained at all times. , These are the minimum and maximum operating capacity limits, respectively. The dynamic storage capacity satisfies the mass conservation equation as follows: (6); In the formula: This refers to the inflow rate of the upper reservoir; This refers to the outflow rate of the upper reservoir.

[0026] 3) Multi-source power output model: including power output characteristic models of wind power, photovoltaic power and small hydropower, to characterize the intermittency of wind and solar power output and the regulation capability of small hydropower.

[0027] In the preferred scheme, the output power of wind power generation is mainly affected by wind speed, and the basic relationship is represented by the following piecewise function: (7); In the formula: To cut in wind speed; To cut off the wind speed; This refers to the rated power of the fan; Rated wind speed; The output power of photovoltaic (PV) power generation mainly depends on solar irradiance, ambient temperature, and the characteristics of the PV modules. The PV power generation model is expressed as: (8); In the formula: This represents the rated power of the photovoltaic system under reference conditions. This represents the actual solar irradiance. For reference irradiance; This refers to the actual temperature of the photovoltaic module. This is the temperature coefficient (usually taken as -0.4% / ℃). This is a reference temperature.

[0028] The small hydropower output model, based on the turbine power equation, is used to characterize the instantaneous power generation capacity and operating boundary of small hydropower under runoff drive, providing adjustable capacity support for renewable energy consumption and system regulation. The expression is: (9); In the formula: For the power generation efficiency of the water turbine; For the upper reservoir Constant water flow rate; For water purification.

[0029] The above steps complete the construction of the multi-power source joint scheduling model in this embodiment.

[0030] II. Day-to-Day Phased Scheduling Model like Figure 2As shown, this is the overall optimization and control framework in this embodiment, namely the basic framework of two-stage coordinated control from day-ahead to intraday, presenting a dual-time-scale scheduling logic: The first stage, day-ahead, is a 1-hour-level optimization scheduling, generating a day-ahead baseline control plan in advance based on day-ahead forecast information such as the maximum output and load demand of wind power, photovoltaics, and small hydropower. The second stage is the intraday control stage, which adopts the MPC rolling optimization strategy, dividing the control cycle into multiple time periods with a 15-minute time granularity, and achieving dynamic closed-loop control through a sliding time window. Each optimization executes only the control command for the first time period, and then slides the optimization window forward by 15 minutes, re-acquires measured data and updates forecast information, and starts the next round of optimization calculation.

[0031] 1) The current stage: In the preferred scheme, in step S2, the day-ahead scheduling objective function is to minimize the system operating cost. Relevant indicators need to be set to achieve the specified wind-solar integration ratio; if this ratio is not met, corresponding penalty measures will be implemented. The objective function expression is: (10); in, (11); In the formula: This represents the minimum operating cost during the day-ahead scheduling period. The price is for pumping electricity; and These are the intraday scheduling period number and the total number of intraday scheduling periods, respectively, where T = 24 hours. For the day before The period is long, 1 hour; For distributed pumped storage in a given time period Internal pumping power; For small hydropower grid connection tariffs; For the time period The small hydropower output within; The cost of penalizing the abandonment of wind and solar power; The stipulated wind and solar energy absorption ratio; The actual proportion of new energy consumed within one day; for Photovoltaic power output during the time period; for Wind power output during a given time period; For distributed pumped storage in a given time period Power generation capacity within the unit; For the system in time period The load power borne by the internal components.

[0032] 2) Intraday phase: In the preferred scheme, in step S2, during the intraday rolling dispatch phase, considering the uncertainty of wind power and photovoltaic output and the real-time fluctuation of load, the objective function is to maximize the total power generation of the system, while satisfying the real-time power balance constraint. The expression of the objective function is: (12); In the formula: This represents the maximum power generation during the daily dispatch period. and These are the intraday scheduling period number and the total number of intraday scheduling periods, respectively, where T = 24 hours. Within the day The duration is long, 15 minutes; and The first day Measured output values ​​of wind power and photovoltaic power during the time period; For the day within Small hydropower output value within a time period; For the day within Distributed pumped storage power generation capacity within a given time period.

[0033] In this embodiment, it is further necessary to set corresponding constraints for the objective function in the corresponding model, which are consistent with its working principle, specifically as follows: (1) Analysis of water-electric coupling characteristics In distributed pumped storage systems, fluctuations in electricity levels can cause fluctuations in water volume. For example... Figure 4 As shown, the coupling relationship between power fluctuations and water volume changes is clearly demonstrated: increased power output → decreased water level; decreased power output → increased water level.

[0034] Based on the water-electricity coupling characteristics, the constraints include: 1) Capacity constraints of upstream and downstream reservoirs: (13); In the formula: For distributed pumped storage in a given time period Internal pumping power; For distributed pumped storage in a given time period Power generation capacity within the unit; and These are the intraday scheduling time slot number and the total number of intraday scheduling time slots, respectively. For the day before The time period is long; in this embodiment, =1 hour, T=24 hours.

[0035] 2) Global power balance constraints: (14); In the formula: For distributed pumped storage in a given time period Internal pumping power; For the time period The small hydropower output within; for Photovoltaic power output during the time period; for Wind power output during a given time period; For distributed pumped storage in a given time period Power generation capacity within the unit; For the system in time period The load power borne by the internal components.

[0036] 3) Multi-source power output model: including power output characteristic models of wind power, photovoltaic power and small hydropower, to characterize the intermittency of wind and solar power output and the regulation capability of small hydropower.

[0037] (2) Day-ahead dispatch Based on 24-hour time-series data, a distribution network operation scenario with a high proportion of renewable energy penetration is constructed and simulated under the day-ahead dispatch framework. The system incorporates key operating parameters such as electricity price, wind and solar power output, small hydropower output, and load to analyze its dispatch characteristics and operation patterns under multi-source collaborative conditions.

[0038] In the preferred scheme, the day-ahead constraints include upper limits on grid transmission power, wind and solar power output, small hydropower output, and DPHS reservoir capacity and power constraints, specifically: 1) Power transmission limit constraints in the power grid: (15); In the formula: Photovoltaic output power; This refers to the output power of wind power. This represents the maximum capacity for wind-solar hybrid power generation that the power grid can accept.

[0039] 2) Constraints on wind and solar power output: (16); In the formula: for Maximum wind power output during a given time period; for The maximum output of photovoltaic power during a given period.

[0040] 3) Output constraints of small hydropower stations: (17); In the formula: for The maximum output of small hydropower stations during certain periods.

[0041] In the preferred embodiment, step S2, the first stage, further includes: setting a threshold for the absorption ratio, and introducing a penalty cost when the actual absorption ratio is lower than the threshold.

[0042] like Figure 5 As shown, the predicted curves for wind power, solar power, small hydropower, and load under day-ahead dispatch are presented. It can be seen that the load curve shows a trend of first gradually rising, then falling, and then fluctuating throughout the day; wind power fluctuates relatively frequently and with large amplitude; solar power output is significant during the daytime, reaching a high level around noon, and then zero at night; small hydropower output fluctuates relatively little overall, remaining within a certain range. These predicted curves reflect the intraday variation patterns of different energy output characteristics and load demand.

[0043] (3) Intraday scheduling Intraday control uses data sampled every 15 minutes to simulate the operation of the distribution network. The control results obtained on a typical day are used as inputs for training the intraday control. One day is randomly selected from the intraday operation dataset as the intraday time-of-use electricity price, load, wind and solar power output, and small hydropower output used in the simulation.

[0044] In the preferred scheme, the real-time power balance constraint is expressed as: (18); In the formula: For the day within Distributed pumped storage pumping power within a time period; For the day within Actual electricity load during the time period.

[0045] like Figure 6 As shown, this displays the daily variations in wind and solar power output, small hydropower, load, and electricity prices across 96 time periods (15-minute intervals): 1) Data sampled every 15 minutes throughout the day, showing the temporal changes of wind power generation, photovoltaic power generation, small hydropower, load, and time-of-use electricity prices over 96 time periods. The overall output of wind power generation fluctuates significantly, with significant peaks at hours 4, 8, 12, 15, and 20, reaching a maximum output of approximately 15MW at hour 12.

[0046] 2) Photovoltaic power generation gradually increases during the daytime, while it remains at a low level in the early morning and midnight, reaching a maximum output of about 6MW at the 14th hour; small hydropower maintains a relatively stable output level throughout the day. However, the load is significantly higher than the sum of the output of wind and solar power and small hydropower for most of the time, with demand peaks occurring at the 7th, 10th, 15th, and 17th hours.

[0047] 3) Price levels in different periods exhibit multi-peak variations due to the influence of multiple factors such as load demand and renewable energy output. When the power supply and demand relationship is relatively tight, the electricity price shows a clear upward trend; subsequently, as the demand for electricity or the supply on the generation side changes, the electricity price also decreases or rises again.

[0048] III. Model Solving and Verification In this embodiment, to verify the effectiveness of the proposed two-stage regulation model, a numerical example analysis is performed using the improved IEEE 30-node system. The test system configuration includes: 1. The total capacity of distributed pumped storage is 21MW; 2. A wind farm with a maximum output of 15MW; 3. A photovoltaic power station with a peak output of 6MW; 4. The total installed capacity of small hydropower is 8MW; 5. The total peak load of the system is approximately 35MW, and the load types include residential load, commercial load and some industrial load.

[0049] like Figure 3 As shown, the distribution network topology after the access of multiple energy sources is displayed. By combining the optimization results of typical daytime scenarios with rolling control driven by actual intraday data, the entire process from macro planning to real-time control has been verified.

[0050] In this embodiment, to evaluate the applicability and advantages of the proposed scheduling mechanism in actual operation, three typical operating scenarios are set up for comparative analysis: Scenario A: No DPHS regulation. DPHS regulation is not considered in day-ahead and intraday control; the goal is solely to minimize day-ahead system operating costs and maximize intraday power generation.

[0051] Scenario B: Considering DPHS regulation. A coordinated regulation mechanism involving power generation, grid, load, and storage is introduced during the day-ahead phase, but it is not included in intraday rolling regulation.

[0052] Scenario C: Consider DPHS regulation. Deeply integrate the source-grid-load-storage coordinated regulation mechanism in day-ahead and intraday regulation to achieve coordinated optimization across all time scales.

[0053] A unified comparative analysis of the three scenarios was conducted, and the results in Table 1 verified that the proposed method, through full-time-scale collaborative optimization, fully utilizes the flexibility of the energy storage system during the day-ahead planning and intraday rolling correction process, dynamically responds to changes in new energy output and load, and effectively improves the collaborative efficiency of new energy and energy storage.

[0054] Table 1 Comparison of Results for Three Operating Scenarios

[0055] In this embodiment, ADMM is used for distributed solution to cope with real-time fluctuations in wind, solar and small hydropower output, electricity prices and loads.

[0056] like Figure 7 As shown, the residuals change with the number of iterations during the intraday rolling control ADMM process. Initially, both the original residuals and the dual residuals show a decreasing trend, indicating that the algorithm is in the initial convergence exploration stage. During the iteration process, although the two curves fluctuate, they continue to decrease overall, reflecting the process of the model continuously optimizing parameters and approaching the optimal solution. Finally, the residuals stabilize at a low level, indicating that the ADMM algorithm improves computational efficiency and stability in the intraday rolling control scenario.

[0057] like Figure 8 As shown, the 24-hour period is divided into 96 time periods, illustrating the temporal changes in wind power, solar power, load, and DPHS charging and discharging power. It can be seen that the DPHS can flexibly respond according to the output of new energy sources and load demand: charging when wind and solar power output is sufficient and there is ample room for new energy absorption, promoting the effective utilization of new energy; discharging when load is at its peak or when insufficient wind and solar power leads to difficulties in new energy absorption, achieving reasonable allocation of new energy and improving grid stability. These control strategies, through multi-factor synergistic optimization, effectively improve the system's economy and the level of new energy absorption while addressing the uncertainties of new energy sources.

[0058] like Figure 9 As shown in the figure, the comparison diagram of renewable energy consumption under scenario A and scenario C shows that the consumption level is compared with that without DPHS and with DPHS two-stage scheduling. It shows that the method significantly improves the renewable energy consumption rate and enhances the consumption stability.

[0059] like Figure 10 As shown, in actual regulation, since the only difference between scenario B and scenario C is whether intraday regulation is carried out, only these two scenarios will be compared here: 1) When only daytime regulation is considered, the power of each unit is basically stable, the output is more concentrated during peak electricity consumption periods, the units follow the grid dispatch instructions, actively consume new energy sources when they are generating a lot, and respond to the demand-side response mechanism during peak electricity consumption periods, but the regulation flexibility is limited.

[0060] 2) After implementing day-ahead and intraday coordinated control, the average unit power output is smoother, power fluctuations are optimized in most periods, and the system can proactively adjust when renewable energy output fluctuates significantly, thus improving the renewable energy absorption capacity. Data shows that under the coordinated control mode, the amplitude of unit power fluctuations is reduced, system stability is improved, and the phenomenon of renewable energy curtailment is alleviated. Overall, the day-ahead and intraday coordinated control mode enhances the system's adaptability to the uncertainties of renewable energy, and improves system stability and adjustment flexibility.

[0061] This embodiment proposes a two-stage optimization scheduling method based on the hydro-electric coupling characteristics of DPHS (Distributed Pumped Storage Power Systems). It employs a coordinated control strategy that integrates day-ahead and intraday optimization objectives: the day-ahead stage aims to minimize operating costs by constructing an economic scheduling model; the intraday stage focuses on maximizing power generation to achieve efficient absorption of clean energy. A dual-timescale coordinated optimization mechanism is established, and its effectiveness in improving system operating economy and renewable energy acceptance capacity is verified in a multi-source distribution network. In the field of distributed pumped storage system regulation, the combined application of the MPC rolling optimization algorithm and the ADMM distributed solution method effectively addresses the uncertainty of wind and solar power output, real-time fluctuations in electricity prices, and dynamic changes in load demand. By establishing a multi-timescale optimization framework, real-time perception and dynamic adjustment of system operating status are achieved, significantly improving the economy and operational reliability of distributed pumped storage systems.

[0062] The above embodiments are merely preferred technical solutions of the present invention and should not be considered as limitations on the present invention. The scope of protection of the present invention should be limited to the technical solutions described in the claims, including equivalent substitutions of the technical features described in the claims. That is, equivalent substitutions and improvements within this scope are also within the scope of protection of the present invention.

Claims

1. A two-stage collaborative optimization scheduling method for a distribution network including distributed pumped storage, characterized in that, Includes the following steps: S1. Model Construction: Construct a multi-source joint scheduling model that includes wind power, photovoltaic power, small hydropower and distributed pumped storage. The multi-source joint scheduling model includes: DPHS water-electricity coupling characteristics, equivalent state of charge (SOC) and reservoir capacity dynamic mathematical model of distributed pumped storage, and multi-source output model of wind power, photovoltaic power and small hydropower output. S2, Phased Scheduling: The first phase aims to minimize system operating costs and generates a baseline scheduling scheme based on predictive data; The second stage adopts model predictive control (MPC) rolling optimization, divides the control cycle according to the preset time granularity, and executes rolling closed-loop control with the goal of maximizing system power generation. S3. Model Solving: The Alternating Directional Multiplier Method (ADMM) is used to solve the staged scheduling model in a distributed manner, satisfying power balance, reservoir capacity constraints, output constraints and grid transmission constraints, and realizing the coordinated operation of source-grid-load-storage.

2. The two-stage collaborative optimization scheduling method for distributed pumped storage power distribution networks according to claim 1, characterized in that, The water-electric coupling characteristics of the DPHS include: The pumping channel is for water storage in the upper reservoir, and the pump is driven by an electric motor. The expression is: (1)； The power generation channel involves releasing water from the upper reservoir to drive a turbine to generate electricity. The expression is as follows: (2)； in: (3)； (4)； In the formula: The density of water; It is the acceleration due to gravity; This refers to the inflow rate of the upper reservoir; This refers to the outflow rate from the upper reservoir. Losses along the route; For motor efficiency; The pumping efficiency of the water pump; This is the friction loss factor. This refers to the length of the pipe. The diameter of the water pipe; It is the Reynolds number; This is the roughness coefficient of the pipe.

3. The two-stage collaborative optimization scheduling method for distributed pumped storage power distribution networks according to claim 2, characterized in that, Based on the water-electricity coupling characteristics, constraints are set, including: 1) Capacity constraints of upstream and downstream reservoirs: (13)； In the formula: For distributed pumped storage in a given time period Internal pumping power; For distributed pumped storage in a given time period Power generation capacity within the region; and These are the intraday scheduling time slot number and the total number of intraday scheduling time slots, respectively. For the day before Long duration; 2) Global power balance constraints: (14)； In the formula: For distributed pumped storage in a given time period Internal pumping power; For the time period The small hydropower output within; for Photovoltaic power output during the time period; for Wind power output during a given time period; For distributed pumped storage in a given time period Power generation capacity within the region; For the system in time period The load power borne by the internal components.

4. The two-stage collaborative optimization scheduling method for distributed pumped storage power distribution networks according to claim 1, characterized in that, The distributed pumped storage SOC model includes: The equivalent state of charge (SOC) model is as follows: (5)； In the formula: for The effective storage capacity of the reservoir should be maintained at all times. , These are the minimum and maximum operating capacity limits, respectively. The dynamic storage capacity satisfies the mass conservation equation as follows: (6)； In the formula: This refers to the inflow rate of the upper reservoir; This refers to the outflow rate of the upper reservoir.

5. The two-stage collaborative optimization scheduling method for distributed pumped storage power distribution networks according to claim 1, characterized in that, The multi-source output model includes: The wind power generation model is expressed as follows: (7)； In the formula: To cut in wind speed; To cut off the wind speed; This refers to the rated power of the fan; Rated wind speed; The photovoltaic power generation model is expressed as follows: (8)； In the formula: This represents the rated power of the photovoltaic system under reference conditions. This represents the actual solar irradiance. For reference irradiance; This refers to the actual temperature of the photovoltaic module. This is the temperature coefficient (usually taken as -0.4% / ℃). For reference temperature; The output model for small hydropower is expressed as follows: (9)； In the formula: For the power generation efficiency of the water turbine; For the upper reservoir Constant water flow rate; For water purification.

6. The two-stage collaborative optimization scheduling method for distributed pumped storage power distribution networks according to claim 1, characterized in that, In step S2, the first stage uses minimizing the system operating cost as the objective function, expressed as: (10)； in, (11)； In the formula: This represents the minimum operating cost during the day-ahead scheduling period. The price is for pumping electricity; and These are the intraday scheduling period number and the total number of intraday scheduling periods, respectively, where T = 24 hours. For the day before The period is long, 1 hour; For distributed pumped storage in a given time period Internal pumping power; For small hydropower grid connection tariffs; For the time period The small hydropower output within; The cost of penalizing the abandonment of wind and solar power; The stipulated wind and solar energy absorption ratio; The actual proportion of new energy consumed within one day; for Photovoltaic power output during the time period; for Wind power output during a given time period; For distributed pumped storage in a given time period Power generation capacity within the region; For the system in time period The load power borne by the internal components.

7. The two-stage collaborative optimization scheduling method for distributed pumped storage power distribution networks according to claim 6, characterized in that, The first phase of constraints includes upper limits on grid transmission power, wind and solar power output, small hydropower output, and DPHS reservoir capacity and power constraints, specifically: 1) Power transmission limit constraints in the power grid: (15)； In the formula: Photovoltaic output power; This refers to the output power of wind power. This represents the maximum capacity of wind-solar hybrid power generation that the power grid can accept. 2) Constraints on wind and solar power output: (16)； In the formula: for Maximum wind power output during a given time period; for Maximum photovoltaic output during the specified time period; 3) Output constraints of small hydropower stations: (17)； In the formula: for The maximum output of small hydropower stations during a given time period.

8. The two-stage collaborative optimization scheduling method for distributed pumped storage power distribution networks according to claim 1, characterized in that, In step S2, the first stage also includes: by setting a threshold for the consumption ratio, a penalty cost is introduced when the actual consumption ratio is lower than the threshold.

9. The two-stage collaborative optimization scheduling method for distributed pumped storage power distribution networks according to claim 1, characterized in that, In step S2, the second-stage scheduling takes maximizing the total power generation of the system as the objective function, satisfying the real-time power balance constraint. The expression of the objective function is: (12)； In the formula: This represents the maximum power generation during the daily dispatch period. and These are the intraday scheduling period number and the total number of intraday scheduling periods, respectively, where T = 24 hours. Within the day The duration is long, 15 minutes; and The first day Measured output values ​​of wind power and photovoltaic power during the time period; For the day within Small hydropower output value within a time period; For the day within Distributed pumped storage power generation capacity within a given time period.

10. The two-stage collaborative optimization scheduling method for distributed pumped storage power distribution networks according to claim 9, characterized in that, The real-time power balance constraint is expressed as follows: (18)； In the formula: For the day within Distributed pumped storage pumping power within a time period; For the day within Actual electricity load during the time period.

Images