Double-layer optimization scheduling method for participation of wind-solar-thermal-storage multi-source combined system in peak regulation and frequency modulation of power grid

Through the double-layer optimization scheduling method, combined with the power grid fluctuation and energy storage energy transfer characteristics, the wind and light grid connection power and thermal power unit output are optimized, which solves the problem of insufficient guiding energy storage charging and discharging strategies in peak and frequency regulation of wind and light heat storage systems. , achieving the effect of minimum total system cost and highest scenic absorption rate.

CN119093401APending Publication Date: 2024-12-06CHINA THREE GORGES UNIV
View PDF 0 Cites 5 Cited by

Patent Information

Application Number
CN202411101588.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-12
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

The existing multi-source combined system of wind and light fire storage is difficult to accurately adjust the charging and discharging power of energy storage in peak and frequency regulation, resulting in high total system cost and insufficient wind and light absorption rate, and failing to effectively utilize energy storage in different charge areas and Combination strategy under load range.

Method used

A two-layer optimization scheduling method is proposed. By constructing a mathematical model with the minimum net load fluctuation of the power grid as the upper layer objective function, combining the energy transfer characteristics of energy storage, the wind and light grid connection power is optimized, and the wind and light grid connection power is optimized in different load intervals and energy storage charge areas The charging and discharging power is optimized under the combination, and the energy storage output is coordinated, and the thermal power unit output is optimized with the minimum system operation cost as the lower level.

Benefits of technology

The double-layer optimized scheduling with the smallest total system cost and the highest wind and light absorption rate is achieved, reducing the peak and frequency regulation pressure of the thermal power unit, and improving the frequency stability of the system and the utilization rate of energy storage.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119093401A_ABST
    Figure CN119093401A_ABST
Patent Text Reader

Abstract

A double-layer optimization scheduling method for a wind-light-fire-storage multi-source combined system to participate in peak regulation and frequency modulation comprises the following steps: constructing an upper-layer optimization scheduling model by taking the minimum net load fluctuation of a power grid as a target function; in the upper-layer optimization scheduling model, establishing a charge-discharge power optimization mathematical model for energy storage participating in peak regulation and frequency modulation under the combination of different load intervals and different energy storage charge areas; optimizing the output of the thermal power generating unit at each moment according to the energy storage charging and discharging power obtained by the upper-layer optimization scheduling model, the optimal wind-solar grid-connected power and a net load curve, optimizing the thermal power generating unit by taking the minimum system operation cost as a target, and constructing a lower-layer optimization scheduling model; and through optimization solution, determining a thermal power generating unit scheduling strategy with optimal economical efficiency. The method provided by the invention can minimize the total cost of the system and maximize the wind and light absorption rate while participating in peak regulation and frequency modulation of the power grid.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of optimized dispatching of a wind, solar, thermal and storage multi-source combined system participating in peak load and frequency regulation, and in particular to a double-layer optimized dispatching method for a wind, solar, thermal and storage multi-source combined system participating in peak load and frequency regulation of a power grid. Background Art

[0002] Due to the anti-peaking and uncertain characteristics of wind power and the increase in load, the dual pressure of source and load has aggravated the peak-to-valley difference of load. Large-scale grid-connected renewable energy not only cannot provide sufficient frequency regulation reserve capacity, but also causes the system's frequency regulation capability to continue to decline, which has brought great pressure to the peak and frequency regulation of the power grid. If the grid-connected output of wind and solar power is restricted, a large amount of wind and solar power will be abandoned, resulting in a waste of renewable resources. While configuring energy storage to meet the needs of renewable energy consumption, using energy storage for peak and frequency regulation can also reduce system load fluctuations and enhance the system's frequency stability.

[0003] The two-layer optimization scheduling method based on the participation of wind, solar, thermal and storage multi-energy combined system in peak load regulation and frequency regulation can be divided into two parts, one is the power optimization of energy storage participating in peak load regulation and frequency regulation, and the other is the two-layer optimization scheduling of wind, solar, thermal and storage multi-energy combined system. Based on the former, reference [1]: Li Junhui, Hou Tao, Yan Gangui, et al. Two-layer optimization of frequency regulation power of multi-energy storage system considering frequency regulation cost and state of charge recovery [J]. Proceedings of the CSEE, 2021, 41(23): 8020-8033. A two-layer optimization strategy for frequency regulation power of energy storage system considering frequency regulation cost is proposed, which includes a frequency regulation power optimization layer and allocates frequency regulation power in different SOC areas of energy storage. This document only considers frequency regulation in the multi-functional reuse of energy storage, and does not consider peak regulation and frequency regulation to increase the utilization rate of energy storage at the same time.

[0004] Reference [2]: Li Junhui, Zhang Jiahui, Li Cuiping, et al. Configuration scheme and economic analysis of energy storage system participating in peak load regulation [J]. Transactions of China Electrotechnical Society, 2021, 36(19): 4148-4160. A configuration scheme for energy storage to participate in grid peak load regulation is proposed. In order to maximize the consumption of wind power, the dispatching model comprehensively considers the peak load regulation effect of the energy storage system and the deep peak load regulation effect of thermal power units, and optimizes the operating status of energy storage and thermal power units with the goal of minimizing the total peak load regulation cost of the system. This document only considers peak load regulation in the multifunctional reuse of energy storage, and does not consider peak load regulation and frequency regulation to increase the utilization rate of energy storage.

[0005] Reference [3]: Chen Changqing, Li Xinran, Zhang Bingyu, et al. Coordinated control strategy of energy storage peak-shaving and frequency regulation based on multiple time scales [J]. Power System Protection and Control, 2022, 50(05): 94-105. A method for dividing the peak-shaving and frequency regulation working area based on energy storage SOC is proposed to realize the switching of energy storage between peak-shaving and frequency regulation control. On this basis, a coordinated control strategy of peak-shaving and frequency regulation is further proposed to improve the utilization rate of energy storage. This scheme only considers the role of energy storage, which is relatively simple, and does not consider the role of new energy and thermal power units in peak-shaving and frequency regulation.

[0006] Based on the latter, reference [4]: ​​Zhao Shuqiang, Liu Dazheng, Xie Yuqi, et al. Energy storage dispatch strategy for wind-solar-storage combined power generation system based on related opportunity target planning [J]. Automation of Electric Power Systems, 2015, 39(14): 30-36+53. In order to improve the uncertainty of wind power and photovoltaic power generation output and make the wind-solar-storage combined power generation system dispatchable, it is proposed to establish an energy storage dispatch model for the wind-solar-storage combined power generation system with the goal of maximizing the matching degree between the wind-solar-storage dispatch output curve and the planned output curve, the most gentle charging and discharging power of the energy storage battery, and the best sustainable working performance of the energy storage battery. This scheme takes wind-solar-storage into consideration when constructing the model, but the output of the wind-solar-storage model is non-persistent.

[0007] Reference [5]: Ye Ze, Li Xiangqi, Jiang Fei, et al. Hierarchical Optimal Economic Dispatch of Wind, Solar, Thermal and Storage Combined System Considering Optimal Abandonment Rate [J]. Power System Technology, 2021, 45(06): 2270-2280. In order to achieve the most economical consumption of renewable energy such as wind and solar, a dispatch strategy that maximizes the economic benefits of the wind, solar, thermal and storage combined system is proposed. A two-layer optimization dispatch model is established with the minimum grid net load variance as the upper optimization target and the minimum grid operation cost as the lower optimization target. In order to reduce the impact of renewable energy output on grid volatility, this scheme chooses to abandon part of the wind and solar power.

[0008] At present, the two-layer optimization problem of wind, solar, thermal and storage participating in peak and frequency regulation has been widely studied at home and abroad. At present, the research at home and abroad has not yet involved the control of energy storage charging and discharging power under the combination of different energy storage charge areas and different load power intervals, and the guidance for the precise regulation of energy storage participation is not strong. Therefore, it is necessary to formulate accurate charging and discharging strategies to optimize the charging and discharging power of energy storage. Summary of the invention

[0009] In order to solve the problem of wind, solar and other renewable energy sources being connected to the grid and participating in the peak load and frequency regulation of the power grid, the present invention proposes a two-layer optimization scheduling method for a wind, solar, thermal and storage multi-source joint system to participate in peak load and frequency regulation, including zoning the energy storage charge state and load range, constructing a mathematical model of energy storage peak load and frequency regulation charging and discharging power under different load ranges and different energy storage charge area combinations, and coordinating energy storage to output at different charging and discharging powers. A two-layer optimization method is adopted to optimize the energy storage output with the minimum net load fluctuation of the power grid as the upper objective function and to optimize the output of thermal power units with the minimum system operating cost as the lower objective function. The two-layer model proposed in the present invention can minimize the total cost of the system and maximize the wind and solar power consumption rate while participating in the peak load and frequency regulation of the power grid.

[0010] The technical solution adopted by the present invention is:

[0011] A double-layer optimization dispatching method for a wind, solar, thermal and energy storage multi-source combined system to participate in peak and frequency regulation includes the following steps:

[0012] Step 1: Taking the minimum net load fluctuation of the power grid as the objective function, construct the upper-level optimization scheduling model;

[0013] Step 2: In the upper-level optimization scheduling model, establish a mathematical model for optimizing the charging and discharging power of energy storage participating in peak load regulation and frequency regulation under different load intervals and different energy storage charging area combinations;

[0014] Step 3: Based on the energy storage charging and discharging power and the optimal wind and solar grid-connected power obtained by the upper-level optimization scheduling model, the output of the thermal power units at each moment is optimized in combination with the net load curve. The thermal power units are optimized with the goal of minimizing the system operation cost, and the lower-level optimization scheduling model is constructed.

[0015] Step 4: Determine the most economical dispatching strategy for thermal power units through optimization solution.

[0016] In step 1, the minimum fluctuation of net load of the power grid is taken as the objective function of the upper optimization dispatching model, and the energy transfer characteristics of energy storage are used to optimize the wind and solar output and energy storage charging and discharging output at each moment, so that the output of the wind, solar and storage coordination system tracks the load curve to the best effect, and reduce the peak and frequency regulation pressure of thermal power units on the remaining load. And the energy storage charging and discharging power and the optimal wind and solar grid-connected power in each period are transmitted to the lower optimization dispatching model.

[0017] The objective function of the upper-level optimization scheduling model is expressed as follows:

[0018]

[0019] P net,t =P load,t -(1-α w,t ) w,t -(1-α pv,t) pv,t -P dis,t +P ch,t ;

[0020]

[0021] Where: F1 is the net load fluctuation of the system; minF1 is the minimum net load fluctuation of the system, which is the objective function of the upper model; P net,t is the net load at time t; P net,av is the average value of the net load in a scheduling cycle; T is a scheduling cycle; t is a time in a scheduling cycle; P load,t is the initial load at time t; α w,t is the wind abandonment rate at time t; P w,t is the output power of wind power at time t; α pv,t is the abandoned light rate at time t; P pv,t is the photovoltaic output power at time t; P ch,t and P dis,t are the charging and discharging power of the energy storage at time t respectively.

[0022] In step 2, energy storage is used to perform peak load regulation and frequency modulation in the upper optimization scheduling model to minimize the net load fluctuation of the power grid. The charging and discharging power optimization mathematical model of energy storage participating in peak load regulation and frequency modulation is firstly as follows Figure 2 As shown in , the load forecast curve is divided into three regions, namely the upper peak shaving region, the middle frequency regulation region and the lower valley filling region; then, as Figure 3 As shown, the energy storage SOC is divided into three state intervals according to the different energy storage SOC sizes, namely the disabled area, the restricted area and the normal area, and the energy storage working mode is analyzed under different load intervals and different energy storage charge area combinations;

[0023] Construct a mathematical model for optimizing the charging and discharging power of energy storage frequency regulation when the load is in the middle frequency regulation area and the frequency deviation exceeds the dead zone range and is low. The details are as follows:

[0024] When the load is in the middle frequency regulation area, the frequency deviation exceeds the dead zone range and is low, the frequency deviation partition diagram is as follows Figure 5 As shown;

[0025] When the grid frequency When the frequency is in the dead zone, the frequency of the node does not need to be adjusted, and the energy storage does not need to be output;

[0026] When the grid frequency △f∈( f1 , f2 ),△f∈( f1 , f2 ) Figure 5As noted in the figure, the current system frequency is low below the dead zone, and the current system frequency is high above the dead zone. When the frequency is in the normal frequency modulation zone (negative), the energy storage is discharged according to mode 1 in Table 2. The energy storage output mode is shown in Table 2. At this time, the energy storage discharge SOC range is SOC∈(SOC high ,SOC max ), when the energy storage provides power in the frequency modulation stage, it is constrained by its power capacity. The power supplied during the entire frequency modulation stage cannot exceed its remaining power capacity. Its output power is determined according to the system frequency deviation and the energy storage power capacity limit. Its discharge power is

[0027]

[0028] When the frequency △f∈( f2 , f3 ), the frequency is in the warning frequency modulation area (negative), and the energy storage is discharged according to mode 2 in Table 2. At this time, the energy storage SOC range is SoC∈(SOC low ,SOC max ), SOC max , SOC high , SOC low , SOC min It is the charge limit of the energy storage, which are the maximum value of the energy storage charge state, the upper value of the energy storage charge state, the lower value of the energy storage charge state, and the minimum value of the energy storage charge state; its discharge power is:

[0029]

[0030] When frequency △f< f3 When the frequency is in the low extreme frequency modulation area, the energy storage is discharged according to mode 3 in Table 2. At this time, the energy storage SOC range is SOC∈(SOC min ,SOC max ), SOC min Indicates the minimum value of the energy storage charge state;

[0031] Since SOC has a lower limit constraint in this SOC range, the discharge power constraint factor λ1 should be added for variable power discharge. The relationship between the energy storage constraint factor and SOC is as follows: Figure 4 As shown, 0<λ1<1 and decreases with the decrease of SOC, and its discharge power is:

[0032]

[0033] Where: P d1 , P d2 and P d3 are the real-time discharge powers of modes 1, 2 and 3 respectively; P d,max The maximum output discharge power of energy storage frequency modulation; and are the AGC power instructions in the normal frequency modulation area (negative), the warning frequency modulation area (negative) and the emergency frequency modulation area (negative); P g is the real-time output power of the thermal power unit; P e is the rated power of energy storage, S i is the real-time SOC value of energy storage;

[0034] △f is the frequency deviation, which is the difference between the current real-time frequency of the system and the grid frequency of 50Hz; f1 It is the boundary between the dead zone and the normal frequency modulation zone (negative value); f2 It is the boundary between the normal frequency modulation area (negative value) and the warning frequency modulation area (positive value); f3 is the lower limit of frequency deviation; S min , S low , S high , S max They are the minimum value, lower value, higher value and maximum value of energy storage respectively; a mathematical model for optimizing the energy storage frequency regulation charging and discharging power is constructed when the load is in the middle frequency regulation area and the frequency deviation exceeds the dead zone range and is high, as follows:

[0035] When the load is in the middle frequency regulation area, the frequency deviation exceeds the dead zone range and is high. When the grid frequency is high, energy storage charging regulation is required. The frequency deviation partition diagram is as follows Figure 5 As shown;

[0036] When the frequency deviation In the normal frequency modulation area (positive), the energy storage is charged according to mode 4 in Table 2. The energy storage output mode is shown in Table 2. At this time, the energy storage SOC charging range is SOC∈(SOC min ,SOC low ); its charging power is:

[0037]

[0038] When the frequency deviation When it is in the warning frequency modulation area (positive), the energy storage is charged according to mode 5 in Table 2. At this time, the energy storage SOC charging range is SOC∈(SOC min ,SOC high ); its charging power is:

[0039]

[0040] When the frequency deviation When it is in the emergency frequency modulation area (positive), it is charged according to mode 6 in Table 2. At this time, the energy storage SOC range is SOC∈(SOC min ,SOC max), within this range, the energy storage has an upper limit constraint, and the charging power constraint factor λ2 is required for variable power charging. The relationship between the energy storage constraint factor and SOC is as follows: Figure 4 As shown, 0<λ2<1 and decreases as SOC increases. Its charging power is:

[0041]

[0042] Where: P c4 , P c5 and P c6 are the real-time charging power of Mode 4, Mode 5 and Mode 6 respectively; P c,max The maximum charging power for energy storage frequency regulation; and They are the AGC power instructions in the normal frequency modulation area (positive), the warning frequency modulation area (positive) and the emergency frequency modulation area (positive);

[0043] It is the boundary between the dead zone and the normal frequency modulation zone (positive value); It is the boundary between the normal frequency modulation area (positive value) and the warning frequency modulation area (positive value); is the upper limit of frequency deviation;

[0044] Construct a mathematical model for optimizing the charging and discharging power of energy storage peak load regulation when the load is in the upper peak load shaving area, as follows:

[0045] When the load is in the upper peak shaving area, energy storage discharge is required to supplement the lack of power in the grid. As shown in Table 2, the energy storage is discharged in mode 7, mode 8 and mode 9 in sequence.

[0046] In mode 7, SOC∈(SOC high ,SOC max ), at this time the energy storage discharge power is:

[0047] P d7 =[P t -P1-P g ,(S i -S high ) e ]

[0048] Then, it is determined whether the energy storage has completed peak shaving. If mode 8 is not executed, SOC∈(SOC low ,SOC max ), the energy storage has a larger capacity to participate in peak shaving, and its discharge power is:

[0049] P d8 =min{P t -P1-P g ,(S i -S low) e}

[0050] Determine whether the energy storage has completed peak shaving. If mode 9 is not executed, then SOC∈(SOC min ,SOC max ), at which point the energy storage is at the lower boundary of charge SoC min There is a charge limit, and it is necessary to add a discharge power constraint factor λ3 when the energy storage is discharged, 0<λ3<1, and λ3 decreases as the energy storage SOC decreases, so that the energy storage can take into account the state of the energy storage SOC during the valley filling period.

[0051] Its discharge power is:

[0052] P d9 =λ3(P t -P1-P g )+(1-λ3)min(S i -S min ) e

[0053] Where: P d7 , P d8 and P d9 are the real-time discharge powers of mode 7, mode 8 and mode 9 respectively; P t is the load power; P1 is the upper limit power that needs to be peak cut; P g is the real-time output power of the thermal power unit; P e is the rated power of energy storage, S i is the real-time SOC value of energy storage; S min , S low , S high , S max are the minimum, lower, upper and maximum energy storage values ​​respectively; λ3 is the constraint factor ( 0<λ3<1 ), the relationship between energy storage constraint factor and SOC is as follows Figure 4 shown.

[0054] The optimized mathematical model of energy storage peak load charging and discharging power when the load is in the lower valley filling area is constructed as follows:

[0055] When the load is in the lower valley filling area, energy storage is needed to be charged to absorb the excess power in the grid. As shown in Table 2, the energy storage is charged in mode 10, mode 11 and mode 12. When in mode 10, SOC∈(SOC min ,SOC low ), at this time the energy storage charging power is:

[0056] P c10 =[P2-P t -P g ,(S low-S i )]

[0057] Then determine whether the energy storage has completed valley filling. If not, continue to execute mode 11 charging.

[0058] SOC∈(SOC min ,SOC high ), the energy storage has a larger capacity to participate in valley filling; its charging power is:

[0059] P c11 =min{P2-P t -P g ,(S high -S i )}

[0060] Determine whether the energy storage has completed valley filling at this time. If it has not completed execution mode 12, then SOC∈(SOC min ,SOC max ), since the upper limit of energy storage charge is SOC max In order to ensure that the energy storage capacity does not exceed the set upper limit during charging, a charging power constraint factor λ4 needs to be added during energy storage charging, 0<λ4<1, and λ4 decreases as the energy storage SOC increases. The relationship between the energy storage constraint factor and SOC is as follows: Figure 4 As shown, this allows the energy storage to take into account the state of the energy storage SOC during the valley filling period. Its charging power is:

[0061] P c12 =λ4(P2-P t -P g )+(1-λ4)min(S max -S i ) e

[0062] Where: P c10 , P c11 and P c12 are the real-time discharge powers of mode 10, mode 11 and mode 12 respectively; P t is the load power; P2 is the lower limit power required to fill the valley; P g is the real-time output power of the thermal power unit; P e is the rated power of energy storage, S i is the real-time SOC value of energy storage; S min , S low , S high , S max They are the minimum, lower, upper and maximum values ​​of energy storage respectively; λ4 is the constraint factor (0<λ4<1)

[0063] In step 3, the objective function of the lower-level optimization scheduling model is to minimize the system operation cost. The objective function of the lower-level optimization scheduling model is expressed as follows:

[0064] minF3=C1+C2+C3

[0065] In the formula: F3 is the operating cost of thermal power units, wind power, photovoltaics and energy storage participating in the optimized scheduling; C1 is the operating cost of thermal power units, including the coal consumption cost and start-up and shutdown costs of thermal power units; C2 is the operating cost of wind power, photovoltaics and energy storage; C3 is the cost of wind and solar power abandonment.

[0066] Operating costs of thermal power units:

[0067]

[0068] Where: f mh 、f qt are the coal consumption cost and start-up and shutdown cost of thermal power units respectively; a i 、b i and c i are the consumption coefficients of thermal power unit i; P i,t is the output of thermal power unit i at time t; S i,t is the start-up and shutdown cost of thermal power unit i at time t; u i,t is a 0-1 variable representing the operating status of thermal power unit i at time t; T is a scheduling cycle; I is the set of all thermal power units; u i,(t-1) is a 0-1 variable representing the operating status of thermal power unit i at time t-1;

[0069] Wind and solar energy storage operating costs:

[0070]

[0071] Where: w , δ p , δ s are the operating cost coefficients of wind power, photovoltaic power and energy storage respectively; are the wind power and photovoltaic power consumption at time t; P ch,t is the charging power of the energy storage at time t; P dis,t is the discharge power of the energy storage at time t.

[0072] Cost of curtailing wind and solar power:

[0073] The absorption capacity of renewable energy is represented by the sum of abandoned wind power and abandoned photovoltaic power during the dispatch period. The smaller the value, the stronger the absorption capacity, which is:

[0074]

[0075] Where: w, δ pv are the cost of wind and solar power abandonment; P w,cut,t is the wind power abandoned at time t; P pv,cut,t is the abandoned optical power at time t; △t is a time period.

[0076] In step 4, the output of the thermal power units at each moment is optimized based on the energy storage charging and discharging power, the optimal wind and solar grid-connected power obtained by the upper-level optimization scheduling model and the net load curve, and the most economical thermal power unit scheduling strategy is determined through optimization solution.

[0077] The optimization solution specifically includes the following steps:

[0078] step4.1: Design implementation example: The present invention adopts the improved IEEE-30 node example for analysis. The system includes 4 thermal power units, 1 wind farm with a total capacity of 600MW, 1 photovoltaic power station with a total capacity of 250MW and 1 energy storage power station. The energy storage is installed on the power generation side of the system. The lithium iron phosphate battery modules that have been widely used are selected for assembly. The configuration result of the energy storage power station does not consider the investment cost. The CPLEX solver is called by the YALMIP toolbox in the Matlab environment to solve the model in this paper.

[0079] Step 4.2: Four models were designed for comparative analysis, namely:

[0080] Model 1: Optimal dispatch of wind, solar and thermal multi-source systems without considering energy storage participation;

[0081] Model 2: Two-layer optimal dispatch of wind, solar, thermal and storage multi-source system considering the participation of energy storage in peak regulation;

[0082] Model 3: Single-layer optimal dispatch of a wind, solar, thermal and storage multi-source system considering the coordinated participation of energy storage in peak and frequency regulation;

[0083] Model 4: Two-layer optimal scheduling of a wind, solar, thermal and storage multi-source system considering the coordinated participation of energy storage in peak and frequency regulation.

[0084] Among them, model 4 is the model adopted by the present invention.

[0085] step4.3: Obtain the scheduling results of each model.

[0086] The optimal economic dispatch strategy for thermal power units is as follows:

[0087] Figure 7 The output diagram of the dispatch result of model 4 shows the optimal output of thermal power units, energy storage and wind and solar power. The dispatch strategy of thermal power units with the best economic efficiency refers to the output of thermal power units, energy storage and wind and solar power with the best economic efficiency obtained by simulation dispatch after designing a suitable mathematical model.

[0088] The present invention provides a double-layer optimization dispatching method for a wind, solar, thermal and energy storage multi-source combined system to participate in peak load and frequency regulation, and the technical effects are as follows:

[0089] 1) In step 1 of the present invention, after renewable energy sources such as wind and solar power are added to the system, the intermittent nature and uncertainty of wind and solar power will increase the volatility of the system and affect the stability of the system. Therefore, it is necessary to take the minimum fluctuation of the net load of the power grid as the objective function. This can ensure that after wind and solar power are added to the power grid, the impact of the addition of wind and solar power on the stability of the power grid operation can be minimized.

[0090] 2) Currently, many documents optimize and adjust the peak and frequency modulation power by partitioning the energy storage SoC, but have not yet involved the energy storage charging and discharging power control under different SoC areas and different load power combinations, and the guidance for the energy storage to participate in precise regulation is not strong. Step 2 of the present invention is improved on this basis, and the energy storage charging and discharging power is controlled under different SoC areas and different load power combinations.

[0091] 3) In step 3 of the present invention, the charging and discharging power of the energy storage can be obtained according to the upper model, and the net load curve can be obtained by combining the wind and solar grid-connected power and the load curve. When the peak and frequency regulation is required on the net load curve, the energy storage is used first, and the rest is left to the thermal power unit, which can reduce the operating cost of the thermal power unit and improve the economy of the system.

[0092] 4) In step 4 of the present invention, the essence of optimization scheduling is to optimize the economic efficiency of the system by designing reasonable solutions and mathematical models. Because the upper optimization model optimizes the output of energy storage, the lower optimization model optimizes the output of thermal power units. The specific content is shown in step 4.3 above. BRIEF DESCRIPTION OF THE DRAWINGS

[0093] The present invention will be further described below in conjunction with the accompanying drawings and embodiments:

[0094] Figure 1 Schematic diagram of the two-layer optimization scheduling model.

[0095] Figure 2 Schematic diagram of load interval division.

[0096] Figure 3 Schematic diagram of energy storage partition.

[0097] Figure 4 Schematic diagram of the relationship between energy storage constraint factor and SOC.

[0098] Figure 5 This is a schematic diagram of frequency deviation partitioning.

[0099] Figure 6 It is the charging and discharging power and SOC of energy storage in each period.

[0100] Figure 7 The double-layer optimization dispatch results considering the participation of wind, solar, thermal and storage multi-energy combined system in peak and frequency regulation are presented.

[0101] Figure 8 Frequency deviation before and after optimization under continuous load disturbance.

[0102] Fig. 9 This is the output curve of energy storage during the frequency regulation period. DETAILED DESCRIPTION

[0103] A double-layer optimization dispatching method for a wind, solar, thermal and energy storage multi-source combined system participating in peak and frequency regulation, comprising:

[0104] Establish a two-layer optimization dispatch model for peak and frequency regulation including wind, solar, thermal and energy storage;

[0105] The two-layer optimization scheduling model is divided into two parts, namely, an upper-layer optimization scheduling model in which wind, solar, and thermal storage participate in peak-shaving and frequency regulation, and a lower-layer optimization model in which wind, solar, and thermal storage participate in peak-shaving and frequency regulation;

[0106] In the upper-level optimization dispatching model, peak-shaving and frequency regulation are optimized through energy storage to minimize the net load fluctuation of the power grid. In the lower-level optimization model, the output of thermal power units is controlled with the goal of minimizing the total system cost.

[0107] In addition, a mathematical model of the charging and discharging power of energy storage participating in peak and frequency regulation is constructed to optimize the charging and discharging power of energy storage. Then, a two-layer optimization scheduling model is constructed to optimize the output of energy storage with the minimum net load fluctuation of the power grid as the upper goal and to optimize the output of thermal power units with the minimum system operating cost as the lower goal.

[0108] Figure 1 It is a two-layer optimization scheduling model. The specific steps to establish the two-layer optimization scheduling model are as follows:

[0109] (1): Taking the minimum fluctuation of net load of the power grid as the objective function of the upper optimization model, the energy transfer characteristics of energy storage are used to optimize the wind and solar power output and the energy storage charging and discharging output at each moment, so that the output of the wind, solar and energy storage coordinated system can track the load curve to the best effect, reducing the peak and frequency regulation pressure of thermal power units on the remaining load.

[0110]

[0111] P net,t =P load,t -(1-α w,t ) w,t -(1-α pv,t ) pv,t -P dis,t +P ch,t

[0112]

[0113] Where: P net,t is the net load at time t; P net,av is the average value of the net load within a dispatching cycle; P load,t is the initial load at time t; α w,t is the wind abandonment rate at time t; P w,t is the output power of wind power at time t; α pv,t is the abandoned light rate at time t; P pv,t is the photovoltaic output power at time t; P ch,t and P dis,t are the charging and discharging power of the energy storage at time t respectively.

[0114] P net,t =[353.89521, 310.45131, 292.6653, 120.51694, 291.36909, 460.1852, 476.14943, 656.7057, 836.77601, 882.32851, 732.82103, 753.37252, 775.54077, 500.57374, 416.10311, 198.96529, 260.88476, 365.16259, 525.29731, 625.05615, 654.55535, 730.18695, 802.99031, 470.08636]

[0115] P net,av =520.5266225.

[0116] (2): Taking the minimum system operation cost as the objective function of the lower optimization model, the output of thermal power units at each moment is optimized based on the energy storage charging and discharging power, the optimal wind and solar grid-connected power obtained in the upper layer, and the net load curve. The dispatching strategy of thermal power units with the best economic efficiency is determined through optimization solution. The lower objective function is expressed as follows:

[0117] minF3=C1+C2+C3

[0118] In the formula: F3 is the operating cost of thermal power units, wind power, photovoltaics and energy storage participating in the optimized scheduling; C1 is the operating cost of thermal power units, including the coal consumption cost and start-up and shutdown costs of thermal power units; C2 is the operating cost of wind power, photovoltaics and energy storage; C3 is the cost of wind and solar power abandonment.

[0119] Thermal power unit operating costs

[0120]

[0121] Where: fmh 、f qt are the coal consumption cost and start-up and shutdown cost of thermal power units respectively; a i , b i and c i are the consumption coefficients of thermal power unit i; P i,t is the output of thermal power unit i at time t; S i,t is the start-up and shutdown cost of thermal power unit i at time t; u i,t is the operating status of thermal power unit i at time t.

[0122] Among them, a of unit 1 i , b i 、c i and S i,t They are 0.0211, 20.86, 1324.3, and 32500 respectively; the upper limit of output is 280MW, and the lower limit of output is 140MW; the a of unit 2 i , b i 、c i and S i,t They are 0.086, 21.62, 478.29, and 32410 respectively, with an upper limit of 243MW and a lower limit of 120MW; a of unit 3 i , b i 、c i and S i,t They are 0.045, 23.54, 637.6, and 3850 respectively, with an upper limit of 140MW and a lower limit of 70MW. i , b i 、c i and S i,t They are 0.068, 16.31, 504.3 and 3955 respectively. The upper limit of output is 130MW and the lower limit of output is 65MW.

[0123] Wind and solar energy storage operating costs:

[0124]

[0125] Where: w , δ p , δ s are the operating cost coefficients of wind power, photovoltaic power and energy storage respectively; are the wind power and photovoltaic power consumption at time t. w , δ p , δ s They are 375 yuan / MW, 570 yuan / MW and 571 yuan / MW respectively.

[0126] Cost of curtailing wind and solar power:

[0127] The absorption capacity of renewable energy is represented by the sum of abandoned wind power and abandoned photovoltaic power during the dispatch period. The smaller the value, the stronger the absorption capacity, which is:

[0128]

[0129] Where: w , δ pv are the cost of wind and solar power abandonment; P w,cut,t is the wind power abandoned at time t; P pv,cut,t is the abandoned optical power at time t. w , δ pv They are 134 yuan / MW and 134 yuan / MW respectively.

[0130] In the upper optimization model, a mathematical model of the charging and discharging power of energy storage participating in peak load and frequency regulation is constructed to optimize the charging and discharging power of energy storage, as follows:

[0131] Table 2 Energy storage mode classification and corresponding charging and discharging power

[0132]

[0133] 1) Construct a mathematical model for optimizing the charging and discharging power of energy storage frequency regulation when the load is in the middle frequency regulation zone and the frequency deviation exceeds the dead zone range and is low. The details are as follows: The schematic diagram of frequency deviation partition is as follows Figure 5 As shown in the figure, the dead zone of primary frequency compensation is 50±0.033Hz, so a dead zone of ±0.033Hz is set around the rated frequency. f1 is -0.033HZ; f2 is -0.0915HZ; f3 -0.15HZ. Rated capacity P e The maximum charge and discharge power is 360MW·h. d,max The upper limit of charge is 0.9, the lower limit of charge is 0.1, and the min , S low , S high , S max <0.9.

[0134] When the grid frequency When the frequency is in the dead zone, the node frequency does not need to be adjusted, and the energy storage does not need to be output; when the grid frequency △f∈( f1 , f2 ), the frequency is in the normal frequency modulation area (negative), the energy storage is discharged according to mode 1 in Table 2, and the energy storage output mode is shown in Table 2. At this time, the energy storage discharge SOC range is SOC∈(SOC high ,SOC max ​), when the energy storage provides power in the frequency modulation stage, it is constrained by its power capacity. The power supplied during the entire frequency modulation stage cannot exceed its remaining power capacity. Its output power is determined according to the system frequency deviation and the energy storage power capacity limit. Its discharge power is:

[0135]

[0136] When the frequency △f∈( f2 , f3 ), the frequency is in the warning frequency modulation area (negative), and the energy storage is discharged according to mode 2 in Table 2. At this time, the energy storage SOC range is SoC∈(SOC low ,SOC max ), its discharge power is:

[0137]

[0138] When frequency △f< f3 When the frequency is in the low extreme frequency modulation area, the energy storage is discharged according to mode 3 in Table 2. At this time, the energy storage SOC range is SOC∈(SOC min ,SOC max ), since the SOC has a lower limit constraint in this SOC range, the discharge power constraint factor λ1 should be added for variable power discharge, 0<λ1<1 and decreases with the decrease of SOC, and its discharge power is:

[0139]

[0140] Where: P d1 , P d2 and P d3 are the real-time discharge powers of modes 1, 2 and 3 respectively; P d,max The maximum output discharge power of energy storage frequency modulation; and are the AGC power instructions in the normal frequency modulation area (negative), the warning frequency modulation area (negative) and the emergency frequency modulation area (negative); P g is the real-time output power of the thermal power unit; P e is the rated power of energy storage, S i is the real-time SOC value of energy storage; f1 It is the boundary between the dead zone and the normal frequency modulation zone (negative value); f2 It is the boundary between the normal frequency modulation area (negative value) and the warning frequency modulation area (positive value); f3 is the lower limit of frequency deviation; S min (0.1), S low , S high , S max (0.9) are the minimum, lower, upper and maximum values ​​of energy storage respectively.

[0141] 2) Construct a mathematical model for optimizing the charging and discharging power of energy storage frequency regulation when the load is in the middle frequency regulation area and the frequency deviation exceeds the dead zone range and is high. The details are as follows:

[0142] At this time, when the grid frequency is high, energy storage charging adjustment is required. The frequency deviation partition diagram is as follows Figure 5 As shown, when the frequency deviation In the normal frequency modulation area (positive), the energy storage is charged according to mode 4 in Table 2. The energy storage output mode is shown in Table 2. At this time, the energy storage SOC charging range is SOC∈(SOC min ,SOC low ), its charging power is:

[0143]

[0144] When the frequency deviation When it is in the warning frequency modulation area (positive), the energy storage is charged according to mode 5 in Table 2. At this time, the energy storage SOC charging range is SOC∈(SOC min ,SOC high ), its charging power is:

[0145]

[0146] When the frequency deviation When it is in the emergency frequency modulation area (positive), it is charged according to mode 6 in Table 2. At this time, the energy storage SOC range is SOC∈(SOC min ,SOC max ), within this range, the energy storage has an upper limit constraint, and the charging power constraint factor λ2 is required for variable power charging, 0<λ2<1 and decreases with the increase of SOC, and its charging power is:

[0147]

[0148] Where: P c4 , P c5 and P c6 are the real-time charging power of modes 4, 5 and 6 respectively; P c,max The maximum charging power for energy storage frequency regulation; and They are the AGC power instructions in the normal frequency modulation area (positive), the warning frequency modulation area (positive) and the emergency frequency modulation area (positive); It is the boundary between the dead zone and the normal frequency modulation zone (positive value); It is the boundary between the normal frequency modulation area (positive value) and the warning frequency modulation area (positive value); is the upper limit of frequency deviation; S min , S low , S high , S maxThey are the minimum, lower, upper and maximum energy storage values ​​respectively.

[0149] 3): Construct a mathematical model for optimizing the charging and discharging power of energy storage peak load regulation when the load is in the upper peak load shaving area, as follows:

[0150] At this time, energy storage discharge is required to supplement the lack of power in the grid. As shown in Table 2, the energy storage is discharged in mode 7, mode 8 and mode 9. In mode 7, SOC∈(SOC high ,SOC max ), at this time the energy storage discharge power is:

[0151] P d7 =[P t -P1-P g ,(S i -S high ) e ]

[0152] Then determine whether the energy storage has completed peak shaving. If mode 8 is not executed, SOC∈(SOC low ,SOC max ), the energy storage has a larger capacity to participate in peak shaving, and its discharge power is:

[0153] P d8 =min{P t -P1-P g ,(S i -S low ) e}

[0154] Determine whether the energy storage has completed peak shaving. If mode 9 is not executed, then SOC∈(SOC min ,SOC max ), at which point the energy storage is at the lower boundary of charge SoC min There is a charge limit, and it is necessary to add a discharge power constraint factor λ3 (0<λ3<1) when the energy storage is discharged, and λ3 decreases as the energy storage SOC decreases, so that the energy storage can take into account the state of the energy storage SOC during the valley filling period. Its discharge power is:

[0155] P d9 =λ3(P t -P1-P g )+(1-λ3)min(S i -S min ) e

[0156] Where: P d7 , P d8 and P d9 are the real-time discharge powers of modes 7, 8 and 9 respectively; Pt is the 24-hour load power; P1 is the upper limit power that needs to be peak-cut, which is calculated to be 350.47MW; P g is the real-time output power of the thermal power unit; P e The rated power of the energy storage is 360MW, S i is the real-time SOC value of energy storage; S min (0.1), S low , S high , S max (0.9) are the minimum, lower, upper and maximum energy storage values, respectively, 0.1 min , S low , S high , S max <0.9; λ3 is the constraint factor (0<λ3<1).

[0157] (4) Construct a mathematical model for optimizing the peak-shaving charging and discharging power of energy storage when the load is in the lower valley filling area, as follows:

[0158] At this time, energy storage is needed to charge to absorb the excess power in the grid. As shown in Table 2, the energy storage is charged in mode 10, mode 11 and mode 12. When in mode 10, SOC∈(SOC min ,SOC low ), at this time the energy storage charging power is:

[0159] P c10 =[P2-P t -P g ,(S low -S i )]

[0160] Then determine whether the energy storage has completed valley filling. If not, continue to execute mode 11 charging. At this time, SOC∈(SOC min ,SOC high ), the energy storage has a larger capacity to participate in valley filling and its charging power is:

[0161] P c11 =min{P2-P t -P g ,(S high -S i )}

[0162] Determine whether the energy storage has completed valley filling at this time. If it has not completed execution mode 12, then SOC∈(SOC min ,SOC max ), since the upper limit of energy storage charge is SOC max ​In order to ensure that the energy storage capacity does not exceed the set upper limit during charging, a charging power constraint factor λ4 needs to be added during energy storage charging, 0<λ4<1, and λ4 decreases as the energy storage SOC increases, so that the energy storage can take into account the state of the energy storage SOC during the valley filling period. Its charging power is:

[0163] P c12 =λ4(P2-P t -P g )+(1-λ4)min(S max -S i ) e

[0164] Where: P c10 , P c11 and P c12 are the real-time discharge powers of modes 10, 11 and 12 respectively; P t is the load power; P2 is the lower limit power required to fill the valley, which is calculated to be: 147.62MW; P g is the real-time output power of the thermal power unit; P e The rated power of the energy storage is 360MW, S i is the real-time SOC value of energy storage; S min (0.1), S low , S high , S max (0.9) are the minimum, lower, upper and maximum energy storage values ​​respectively and 0.1 min , S low , S high , S max <0.9; λ4 is the constraint factor (0<λ4<1).

[0165] Example:

[0166] The present invention includes 4 thermal power units, 1 wind farm with a total capacity of 600MW, 1 photovoltaic power station with a total capacity of 250MW and 1 energy storage power station. The effectiveness of the method proposed in the present invention is analyzed and verified based on the actual data provided as an example.

[0167] in Figure 6 It is the charging and discharging power and SOC of energy storage in each period, such as Figure 6 ​As shown, during low-peak hours such as 3:00-5:00 and 16:00-18:00, the energy storage starts charging to "fill" the trough of the load. During peak hours such as 10:00-13:00 and 19:00-22:00, the energy storage takes discharge measures to effectively reduce the peak value of the load, thereby reducing the peak-to-valley difference of the load and alleviating the peak-shaving pressure of the thermal power units. The maximum charging and discharging power of the energy storage at each moment is 60MW, which is within the rated power range. Its state of charge is always maintained between 0.1-0.9 to meet the energy storage operation constraints, indicating that the energy storage scheduling strategy proposed in this patent can meet the constraints of energy storage. Since the present invention adopts a variable parameter charging and discharging strategy, the charging and discharging power of the energy storage at 5:00 and 18:00 at the charging time will decrease as the energy storage SOC increases, and the discharging power of the energy storage at 13:00, 21:00 and 22:00 at the discharging time will decrease as the energy storage SOC decreases. It can be seen from the energy storage SOC curve that the SOC curve of the energy storage at these times will change slowly.

[0168] Figure 7 This is the result of the double-layer optimization dispatch considering the wind, solar, thermal and storage multi-energy combined system participating in peak and frequency regulation. The dispatch results are as follows: Figure 7 As shown in the figure, thermal power units 1, 2, 3, and 4 are always running during the dispatch period. The output of unit 1 is relatively large because the capacity of unit 1 is large and the output is stable. When the four units are running at the same time, the power demand during the dispatch period can be met. Since wind power has the characteristics of anti-peak regulation, the dispatch results show that energy storage is charged to fill the valley at 3:00-5:00 and 16:00-18:00, and discharged to cut the peak at 10:00-13:00 and 19:00-22:00. Due to the addition of frequency regulation, the absorption rate of photovoltaic power in the model is significantly improved, and the wind and light abandonment rate is only 5.446%, which increases the absorption space of wind power and photovoltaic power. Through the reasonable charging and discharging of energy storage during the load valley and peak period, the output curve of thermal power units during the load valley and peak period is smoother, which can make the unit operate more safely and stably, and reduce the regulation pressure of the unit.

[0169] The frequency deviation before and after optimization under continuous load disturbance is as follows: Figure 8 As shown in Figure 2, the output curve of energy storage during the frequency regulation period is as follows: Fig. 9 The frequency modulation sampling period is 1s, and the continuous random load disturbance within the frequency modulation period is obtained. In order to verify the frequency modulation effect of the frequency modulation control strategy of the present invention under continuous load disturbance, the frequency deviation comparison curve before and after adding the control strategy is obtained, as shown in Figure 8 As shown, without the control strategy, the maximum frequency deviation is 0.13295 Hz and -0.12621 Hz. After adding the control strategy, the frequency deviation is within the dead zone range (-0.033, 0.033). Therefore, the frequency modulation strategy of the present invention has a great improvement on the frequency deviation.

[0170] The energy storage charging and discharging efficiency is set to 0.9, the rated capacity is 360MW·h, the maximum charging and discharging power is 60MW, the upper limit of charge is 0.9, and the lower limit of charge is 0.1. The operating cost of the above-mentioned method is obtained by calculation, as shown in Table 1.

[0171] Four models were designed for comparative analysis, namely:

[0172] Model 1: Optimal dispatch of wind, solar and thermal multi-source systems without considering energy storage participation;

[0173] Model 2: Two-layer optimal dispatch of wind, solar, thermal and storage multi-source system considering the participation of energy storage in peak regulation;

[0174] Model 3: Single-layer optimal dispatch of a wind, solar, thermal and storage multi-source system considering the coordinated participation of energy storage in peak and frequency regulation;

[0175] Model 4: Two-layer optimal scheduling of a wind, solar, thermal and storage multi-source system considering the coordinated participation of energy storage in peak and frequency regulation.

[0176] Among them, Model 4 is the model adopted by this patent.

[0177] step4.3: The scheduling results of each model are shown in Table 1 below.

[0178] Table 1 Scheduling results of each model

[0179]

[0180]

[0181] From the dispatching results in Table 1 above, it can be seen that the net load variance of model 4 is the smallest, indicating that the net load fluctuation is the smallest, the total cost is the smallest, and the wind and solar power abandonment rate is the smallest, indicating that the model proposed in the present invention can effectively improve the economy of system operation and the absorption of wind and solar renewable energy.

Claims

1. A two-layer optimization scheduling method for a wind, solar, thermal and storage multi-source joint system to participate in peak and frequency regulation, characterized in that The following steps are involved: Step 1: Taking the minimum net load fluctuation of the power grid as the objective function, construct the upper-level optimization scheduling model; Step 2: In the upper-level optimization scheduling model, establish a mathematical model for optimizing the charging and discharging power of energy storage participating in peak load regulation and frequency regulation under different load intervals and different energy storage charging area combinations; Step 3: Based on the energy storage charging and discharging power and the optimal wind and solar grid-connected power obtained by the upper-level optimization scheduling model, the output of the thermal power units at each moment is optimized in combination with the net load curve. The thermal power units are optimized with the goal of minimizing the system operation cost, and the lower-level optimization scheduling model is constructed. Step 4: Determine the most economical dispatching strategy for thermal power units through optimization solution.

2. According to claim 1, a two-layer optimization scheduling method for a wind, solar, thermal and energy storage multi-source joint system to participate in peak load and frequency regulation is characterized by: In step 1, the minimum net load fluctuation of the power grid is used as the objective function of the upper optimization scheduling model, and the energy transfer characteristics of energy storage are used to optimize the wind and solar power output and energy storage charging and discharging output at each moment, so that the wind, solar and storage coordination system output tracks the load curve to the best effect, and the energy storage charging and discharging power and the optimal wind and solar grid-connected power in each time period are transmitted to the lower optimization scheduling model; The objective function of the upper-level optimization scheduling model is expressed as follows: Where: F1 is the net load fluctuation of the system; minF1 is the minimum net load fluctuation of the system, which is the objective function of the upper model; P net,t is the net load at time t; P net,av is the average value of the net load in a scheduling cycle; T is a scheduling cycle; t is a time in a scheduling cycle; P load,t is the initial load at time t; α w,t is the wind abandonment rate at time t; P w,t is the output power of wind power at time t; α pv,t is the abandoned light rate at time t; P pv,t is the photovoltaic output power at time t; P ch,t and P dis,t are the charging and discharging power of the energy storage at time t respectively.

3. According to claim 2, a double-layer optimization scheduling method for a wind, solar, thermal and energy storage multi-source joint system to participate in peak load and frequency regulation is characterized by: In step 2, energy storage is used for peak-shaving and frequency regulation in the upper-level optimization scheduling model to minimize the net load fluctuation of the power grid. The charge-discharge power optimization mathematical model of energy storage participating in peak-shaving and frequency regulation divides the load forecast curve into three areas, namely, the upper peak-shaving area, the middle frequency regulation area, and the lower valley-filling area; then, the energy storage SOC is divided into three state intervals according to the different sizes of the energy storage SOC, namely, the disabled area, the constrained area, and the normal area, and the energy storage working mode is analyzed under different load intervals and different energy storage charge area combinations; When the load is in the middle frequency regulation area, the frequency deviation exceeds the dead zone range and is low. When the grid frequency When the frequency is in the dead zone, the frequency of the node does not need to be adjusted, and the energy storage does not need to be output; When the grid frequency △f∈( f1 , f2 ), the frequency is in the normal frequency modulation area, and the energy storage is discharged according to mode 1. At this time, the energy storage discharge SOC range is SOC∈(SOC high ,SOC max ), when the energy storage provides power in the frequency modulation stage, it is constrained by its power capacity. The power supplied during the entire frequency modulation stage cannot exceed its remaining power capacity. Its output power is determined according to the system frequency deviation and the energy storage power capacity limit. Its discharge power is: When the frequency △f∈( f2 , f3 ), the frequency is in the warning frequency modulation area, and the energy storage mode 2 is discharged. At this time, the energy storage SOC range is SoC∈(SOC low ,SOC max ), SOC max , SOC high , SOC low , SOC min It is the charge limit of the energy storage, which are the maximum value of the energy storage charge state, the upper value of the energy storage charge state, the lower value of the energy storage charge state, and the minimum value of the energy storage charge state; its discharge power is: When frequency △f< f3 When the frequency is in the low extreme frequency modulation area, the energy storage is discharged according to mode 3. At this time, the energy storage SOC range is SOC∈(SOC min ,SOC max ), SOC min Indicates the minimum value of the energy storage charge state; Since the SOC has a lower limit constraint in this SOC range, the discharge power constraint factor λ1 is added for variable power discharge, 0<λ1<1 and decreases as the SOC decreases. The discharge power is: Where: P d1 , P d2 and P d3 are the real-time discharge powers of modes 1, 2 and 3 respectively; P d,max The maximum output discharge power of energy storage frequency modulation; and are the AGC power instructions in the normal frequency modulation area, the warning frequency modulation area and the emergency frequency modulation area respectively; P g is the real-time output power of the thermal power unit; P e is the rated power of energy storage, S i is the real-time SOC value of energy storage; △f is the frequency deviation; f1 It is the boundary between the dead zone and the normal frequency modulation zone (negative value); f2 It is the boundary between the normal frequency modulation area (negative value) and the warning frequency modulation area (positive value); f3 is the lower limit of frequency deviation; S min , S low , S high , S max They are the minimum, lower, upper and maximum values ​​of energy storage respectively; When the load is in the middle frequency regulation area, and the frequency deviation exceeds the dead zone range and is high, energy storage charging regulation is required when the grid frequency is high; When the frequency deviation In the normal frequency modulation area, the energy storage is charged according to mode 4. At this time, the energy storage SOC charging range is SOC∈(SOC min ,SOC low ); its charging power is: When the frequency deviation When it is in the warning frequency modulation area, the energy storage is charged according to mode 5. At this time, the energy storage SOC charging range is SOC∈(SOC min ,SOC high ); its charging power is: When the frequency deviation When it is in the emergency frequency modulation area, it is charged according to mode 6. At this time, the energy storage SOC range is SOC∈(SOC min ,SOC max ), within this range, the energy storage has an upper limit constraint, and variable power charging is performed using the charging power constraint factor λ2, 0<λ2<1 and decreases as the SOC increases; its charging power is: Where: P c4 , P c5 and P c6 are the real-time charging power of Mode 4, Mode 5 and Mode 6 respectively; P c,max The maximum charging power for energy storage frequency regulation; and They are the AGC power instructions in the normal frequency modulation area, the warning frequency modulation area and the emergency frequency modulation area (positive), respectively; It is the boundary between the dead zone and the normal frequency modulation zone; It is the boundary between the normal frequency modulation area and the warning frequency modulation area; is the upper limit of frequency deviation; When the load is in the upper peak shaving area, energy storage discharge is used to supplement the lack of electric energy in the power grid; the energy storage is discharged in sequence according to mode 7, mode 8 and mode 9; In mode 7, SOC∈(SOC high ,SOC max ), at this time the energy storage discharge power is: P d7 =[P t -P1-P g ,(WITH i -WITH high )P e ] Then, it is determined whether the energy storage has completed peak shaving. If mode 8 is not executed, SOC∈(SOC low ,SOC max ), the energy storage has a larger capacity to participate in peak shaving, and its discharge power is: P d8 =min{P t -P1-P g ,(S i -S low )P e } Determine whether the energy storage has completed peak shaving. If mode 9 is not executed, then SOC∈(SOC min ,SOC max ), at which point the energy storage is at the lower boundary of charge SoC min There is a charge limit, and it is necessary to add a discharge power constraint factor λ3 when the energy storage is discharged, 0<λ3<1, and λ3 decreases as the energy storage SOC decreases, so that the energy storage can take into account the state of the energy storage SOC during the valley filling period; Its discharge power is: P d9 =λ3(P t -P1-P g )+(1-λ3)min(S i -S min )P e Where: P d7 , P d8 and P d9 are the real-time discharge powers of mode 7, mode 8 and mode 9 respectively; P t is the load power; P1 is the upper limit power that needs to be peak cut; P g is the real-time output power of the thermal power unit; P e is the rated power of energy storage, S i is the real-time SOC value of energy storage; S min , S low , S high , S max are the minimum, lower, upper and maximum values ​​of energy storage respectively; λ3 is the constraint factor; When the load is in the lower valley filling area, the energy storage is charged to absorb the excess power in the grid; the energy storage is charged in mode 10, mode 11 and mode 12 in sequence; when in mode 10, SOC∈(SOC min ,SOC low ), at this time the energy storage charging power is: P c10 =[P2-P t -P g ,(S low -S i )] Then determine whether the energy storage has completed valley filling. If not, continue to execute mode 11 charging. At this time, SOC∈(SOC min ,SOC high ), the energy storage has a larger capacity to participate in valley filling; its charging power is: P c11 =min{P2-P t -P g ,(S high -S i )} Determine whether the energy storage has completed valley filling at this time. If it has not completed execution mode 12, then SOC∈(SOC min ,SOC max ), since the upper limit of energy storage charge is SOC max In order to ensure that the energy storage capacity does not exceed the set upper limit during charging, a charging power constraint factor λ4 is added during energy storage charging, 0<λ4<1, and λ4 decreases as the energy storage SOC increases, so that the energy storage can be charged in the valley period while taking into account the state of the energy storage SOC; its charging power is: P c12 =λ4(P2-P t -P g )+(1-λ4)min(S max -S i )P e Where: P c10 , P c11 and P c12 They are the real-time discharge power of mode 10, mode 11 and mode 12 respectively; Pt is the load power; P2 is the lower limit power required to fill the valley; P g is the real-time output power of the thermal power unit; P e is the rated power of energy storage, S i is the real-time SOC value of energy storage; S min , S low , S high , S max They are the minimum, lower, upper and maximum values ​​of energy storage respectively; λ4 is the constraint factor.

4. According to claim 1, a double-layer optimization scheduling method for a wind, solar, thermal and energy storage multi-source joint system to participate in peak load and frequency regulation is characterized by: In step 3, the objective function of the lower-level optimization scheduling model is to minimize the system operation cost. The objective function of the lower-level optimization scheduling model is expressed as follows: min F3=C1+C2+C3 Where: F3 is the operating cost of thermal power units, wind power, photovoltaic power and energy storage participating in the optimal dispatch; C1 is the operating cost of thermal power units, including the coal consumption cost and start-up and shutdown cost of thermal power units; C2 is the operating cost of wind power, photovoltaic power and energy storage; C3 is the cost of wind and solar power abandonment; Operating costs of thermal power units: Where: f mh 、f qt are the coal consumption cost and start-up and shutdown cost of thermal power units respectively; a i , b i and c i are the consumption coefficients of thermal power unit i; P i,t is the output of thermal power unit i at time t; S i,t is the start-up and shutdown cost of thermal power unit i at time t; u i,t is a 0-1 variable representing the operating status of thermal power unit i at time t; T is a scheduling cycle; I is the set of all thermal power units; u i,(t-1) is a 0-1 variable representing the operating status of thermal power unit i at time t-1; Wind and solar energy storage operating costs: Where: w ,δ p ,δ s are the operating cost coefficients of wind power, photovoltaic power and energy storage respectively; are the wind power and photovoltaic power consumption at time t respectively; P ch,t is the charging power of the energy storage at time t; P dis ,t is the discharge power of the energy storage at time t; Cost of curtailing wind and solar power: The absorption capacity of renewable energy is represented by the sum of abandoned wind power and abandoned photovoltaic power during the dispatch period. The smaller the value, the stronger the absorption capacity, which is: Where: w ,δ pv are the cost of wind and solar power abandonment; P w,cut,t is the wind power abandoned at time t; P pv,cut,t is the abandoned optical power at time t; △t is a time period.

5. According to claim 1, a double-layer optimization scheduling method for a wind, solar, thermal and energy storage multi-source joint system participating in peak load and frequency regulation is characterized by: In step 4, the output of the thermal power units at each moment is optimized based on the energy storage charging and discharging power, the optimal wind and solar grid-connected power obtained by the upper-level optimization scheduling model and the net load curve, and the most economical thermal power unit scheduling strategy is determined through optimization solution.

Citation Information

Cited By

  • Multi-energy system double-layer optimization scheduling method for prediction error dynamic compensation

    CN120414731A

  • Fire storage system, control method and device thereof and medium

    CN120433281A

  • Fire storage system and control method, device and medium thereof

    CN120433281B

  • Fire, light and storage complementary active power control method and system

    CN120638522A

  • Power determination method and device of energy storage system and nonvolatile storage medium

    CN122178473A