CN-122026507-A - Agricultural comprehensive energy system collaborative optimization method and device based on high-order water tower-wind-light complementation

CN122026507ACN 122026507 ACN122026507 ACN 122026507ACN-122026507-A

Abstract

The invention relates to the technical field of energy Internet and modern agricultural engineering, and particularly provides a high-level water tower-wind-solar complementary-based agricultural comprehensive energy system collaborative optimization method and device, wherein wind power and photovoltaic predicted output in a day-ahead optimization period are substituted into an upper-layer scheduling model and solved to obtain a day-ahead optimization result; the technical scheme provided by the invention improves the capacity of absorbing renewable energy sources, reduces the electricity purchasing cost and the running cost of a power grid, improves the safety of agricultural irrigation water, enhances the robustness of a system to prediction errors and disturbance, and reduces the carbon emission of the system.

Inventors

ZHANG SIRUI
ZHANG HAIFENG
LI DEXIN
WANG ZHANBO
LI WEN
GAO ZIHAN
CHENG LING
JIANG LIMIN
LIN JINGYI
QU BO
LI CHUNHONG
ZHANG HAILONG
BU FANPENG
NIU ZHEN
ZHAO LINXIANG
HE XIN
ZHANG QINGMIN
ZHANG JIAJUN
ZHANG JING

Assignees

中国电力科学研究院有限公司
国网甘肃省电力公司
国网吉林省电力有限公司电力科学研究院
国家电网有限公司

Dates

Publication Date: 20260512
Application Date: 20251223

Claims (20)

1. The agricultural comprehensive energy system collaborative optimization method based on high-order water tower-wind-light complementation is characterized by comprising the following steps of: Substituting the wind power and photovoltaic predicted output in the day-ahead optimization period into an upper layer scheduling model and solving to obtain a day-ahead optimization result; Taking the day-ahead optimization result as a day-ahead boundary constraint, and solving a lower scheduling model corresponding to a day-ahead optimization period to obtain a day-ahead optimization result; Taking the daily optimization result as a scheduling plan instruction of a daily optimization period to perform collaborative optimization scheduling on the agricultural comprehensive energy system; Each time period in the day-ahead optimization time period comprises a plurality of day-ahead optimization time periods, and the optimization result comprises at least one of wind power generation grid-connected power, photovoltaic power generation grid-connected power, wind curtailment and light curtailment power, pumping station pumping flow and power, irrigation water flow, high-level water tower water quantity, rigid electric load, adjustable electric load total power, grid purchase power and grid selling power.
2. The method of claim 1, wherein the upper layer scheduling model includes a first objective function and constraints.
3. The method of claim 2, wherein the first objective function is as follows: In the above-mentioned method, the step of, For the first value of the objective function, For the electricity price of electricity purchase at the moment t, The power is purchased for the power grid at the moment t, For the total number of time instants in the day-ahead optimization period, In order to provide for the time interval of time, The electricity price is the electricity selling price at the moment t, The electric power is sold for the power grid at the moment t, In order to discard the electrical penalty factor, The wind and light power is abandoned at the moment t, The punishment coefficient for the start and stop of the pump station, The starting action quantity of the pump station at the moment t, For the under-irrigation penalty factor, Is the amount of under-filled water.
4. The method of claim 3, wherein the lower layer scheduling model includes a second objective function and constraints.
5. The method of claim 4, wherein the second objective function is as follows: In the above-mentioned method, the step of, For the second objective function value, For the total number of time instants in the day optimization period, The track of the water quantity of the water tower is tracked and weighted, The water quantity of the high-order water tower at the moment t, For optimizing the water quantity of the high-level water tower at the time t in the result before the day, And k is the starting time, which is the amount of under-filled water in the current prediction time domain.
6. The method of claim 5, wherein the constraint conditions comprise wind and light output and power rejection constraint, high-level water tower water balance and potential energy constraint, pump station power and running state constraint, agricultural irrigation water constraint, agricultural energy load constraint, grid-connected point power balance and power purchase constraint, and under-filling water constraint.
7. The method of claim 6, wherein the wind-solar power output and power rejection constraints are as follows: The water quantity balance and potential energy constraint of the high-level water tower are as follows: the pump station power and the running state are constrained as follows: The agricultural irrigation water is constrained as follows: the agricultural energy load constraints are as follows: The grid-connected point power balance and electricity purchase and sale constraint are as follows: In the above-mentioned method, the step of, The grid-connected power for the wind power generation at the moment t, The grid-connected power of the photovoltaic power generation at the moment t, The wind power output is predicted for the time t, The force is predicted for the photovoltaic at time t, The water quantity of the high-order water tower at the time t+1, The pumping flow of the pump station at the moment t, The water flow is irrigated at the moment t, Is the lower limit of the water quantity of the high-level water tower, Is the upper limit of the water quantity of the high-level water tower, For the water quantity of the high-level water tower at the starting moment of the next period, For the amount of water in the high-level water tower at the beginning moment of the current period, For potential energy storage level of the high-order water tower at the moment t, Is the water density of the water, the water is in a water-tight state, The acceleration of the gravity is that, Is the effective lift of the water tower, The pumping power of the pump station at the moment t is calculated, Is the equivalent efficiency of the pump station, The state of the pump station is the start-stop state of the pump station at the moment t, The upper limit of the pumping flow of the pump station is provided, The lower limit of the pumping flow of the pump station, For the total amount of water needed by agricultural irrigation, In order to irrigate the upper limit of the water flow, To define a set of irrigation allowable periods, In order for the window of time to be operable, The power of the electrical load k is adjustable for time t, For the total power demand of the adjustable electrical load k, For the operating state of the adjustable electrical load k at time t, For the upper power limit of the adjustable electrical load k, For the total power of the adjustable electric load at the time t, In order to adjust the set of loads, Is a rigid electrical load at time t.
8. The method of claim 7, wherein the under-fill volume is constrained as follows: 。
9. The method of claim 8, wherein the step of iteratively solving the pump station pumping flow and pump station pumping power by introducing lagrangian multipliers and augmented lagrangian functions in the step of substituting wind power and photovoltaic predicted output in a day-ahead optimization period into an upper-layer scheduling model and solving.
10. The method of claim 9, wherein the introducing lagrangian multipliers and augmenting lagrangian functions iteratively solve for pump station pumping flow and pump station pumping power, comprising: initializing a pumping flow update value of a pump station; substituting the pumping flow updated value of the pumping station into an electric energy sub-problem function corresponding to the extended Lagrangian function and solving the electric energy sub-problem function to obtain the pumping power updated value of the pumping station; Substituting the pump station pumping power updated value into a water energy problem function corresponding to the extended Lagrangian function and solving the water energy problem function to obtain a pump station pumping flow updated value; And d, judging whether convergence conditions are met, if yes, outputting a pumping station pumping power updating value and a pumping station pumping flow updating value, otherwise, returning to the step b after updating the Lagrangian multiplier.
11. The method of claim 10, wherein the augmented lagrangian function is as follows: In the above-mentioned method, the step of, To be a cost term that contains only power related variables, To include only the cost terms of water and water tower related variables, In order to be a lagrange multiplier, To augment the coefficient, the cost term containing only the power related variables is as follows: The cost term for the water and water tower related variables only is as follows: 。
12. The method of claim 11, wherein the electrical energy sub-problem function corresponding to the augmented lagrangian function is an augmented lagrangian function that removes the cost term that includes only water and water tower related variables, and wherein the water energy sub-problem function corresponding to the augmented lagrangian function is an augmented lagrangian function that removes the cost term that includes only power related variables.
13. The method of claim 11, wherein the lagrangian multiplier is updated as follows: In the above-mentioned method, the step of, Is the updated lagrangian multiplier.
14. The method of claim 11, wherein the convergence condition is as follows: In the above-mentioned method, the step of, Is a preset precision threshold.
15. Agricultural comprehensive energy system collaborative optimization device based on high-order water tower-wind-light complementation, which is characterized by comprising: the first analysis module is used for substituting the wind power and the photovoltaic predicted output in the day-ahead optimization period into an upper-layer scheduling model and solving to obtain a day-ahead optimization result; The second analysis module is used for taking the day-ahead optimization result as a day-ahead boundary constraint, and solving a lower scheduling model corresponding to a day-ahead optimization period to obtain a day-ahead optimization result; The scheduling module is used for carrying out collaborative optimization scheduling on the agricultural comprehensive energy system by taking the daily optimization result as a scheduling plan instruction of a daily optimization period; Each time period in the day-ahead optimization time period comprises a plurality of day-ahead optimization time periods, and the optimization result comprises at least one of wind power generation grid-connected power, photovoltaic power generation grid-connected power, wind curtailment and light curtailment power, pumping station pumping flow and power, irrigation water flow, high-level water tower water quantity, rigid electric load, adjustable electric load total power, grid purchase power and grid selling power.
16. The apparatus of claim 15, wherein the upper layer scheduling model comprises a first objective function and a constraint.
17. The apparatus of claim 16, wherein the first objective function is as follows: In the above-mentioned method, the step of, For the first value of the objective function, For the electricity price of electricity purchase at the moment t, The power is purchased for the power grid at the moment t, For the total number of time instants in the day-ahead optimization period, In order to provide for the time interval of time, The electricity price is the electricity selling price at the moment t, The electric power is sold for the power grid at the moment t, In order to discard the electrical penalty factor, The wind and light power is abandoned at the moment t, The punishment coefficient for the start and stop of the pump station, The starting action quantity of the pump station at the moment t, For the under-irrigation penalty factor, Is the amount of under-filled water.
18. The apparatus of claim 17, wherein the lower layer scheduling model comprises a second objective function and a constraint.
19. The apparatus of claim 18, wherein the second objective function is as follows: In the above-mentioned method, the step of, For the second objective function value, For the total number of time instants in the day optimization period, The track of the water quantity of the water tower is tracked and weighted, The water quantity of the high-order water tower at the moment t, For optimizing the water quantity of the high-level water tower at the time t in the result before the day, And k is the starting time, which is the amount of under-filled water in the current prediction time domain.
20. The apparatus of claim 19, wherein the constraint conditions comprise a wind-solar power output and power rejection constraint, a high-level water tower water balance and potential energy constraint, a pump station power and operating state constraint, an agricultural irrigation water constraint, an agricultural energy load constraint, a grid-tie point power balance and power purchase constraint, and an under-fill water constraint.

Description

Agricultural comprehensive energy system collaborative optimization method and device based on high-order water tower-wind-light complementation Technical Field The invention relates to the technical field of energy Internet and modern agricultural engineering, in particular to a method and a device for collaborative optimization of an agricultural comprehensive energy system based on high-level water tower-wind-light complementation. Background With the continuous advancement of agricultural modernization technology, the energy demand of agricultural areas has the characteristics of diversification and centralization, and the loads of irrigation, greenhouse environment regulation, agricultural product cold chain storage, agricultural power equipment and the like have obvious seasonal peak characteristics. Such loads tend to occur centrally during certain periods of time and require high energy continuity, such that conventional agricultural distribution networks face high energy pressures. Meanwhile, the agricultural area has rich wind energy and solar energy resources, and is suitable for building wind power and photovoltaic power stations with certain scales. However, the wind-solar resource output has strong volatility and uncertainty, and has low time matching degree with agricultural load, so that the capability of the renewable energy sources for absorbing in agricultural production scenes is insufficient for a long time. The existing agricultural irrigation system generally adopts a simple pump station start-stop strategy, and the high-level water tower is only used as a static water storage facility, so that the potential gravitational potential energy storage characteristic of the high-level water tower is not fully developed. The pump station is usually passively started and stopped according to the upper and lower limits of the water level of the water tower, can not form effective coupling with the change of external wind and light resources, and can not realize load peak shifting or butting time-of-use electricity price strategies. Therefore, in the period of abundant wind and light resources, the pump station is possibly not started to discard wind and light, and in the period of concentrated load or the period of higher electricity price, the pump station has to be operated with high intensity, thereby aggravating the system operation cost. Furthermore, the prior art lacks system modeling of the complete link "wind and light power-pump station-high water tower-irrigation load". Most researches focus on single objects, such as high-level water tower structural design, agricultural irrigation load prediction, wind-light power generation grid-connected management and the like, and few methods can perform unified modeling on electric energy, water energy and potential energy and perform collaborative optimization scheduling. The lack of a unified model makes the coupling relationship between different subsystems difficult to use, and the overall efficiency of the system cannot be improved. Meanwhile, prediction errors, sudden demands of agricultural water and wind-light output fluctuation often cause system running deviation, the traditional one-time optimization method cannot realize dynamic correction of the real-time state of the system, and robustness is poor. In summary, the conventional agricultural comprehensive energy system still has a great number of defects in the cooperative operation level, such as low wind and light resource absorption capability, unreleased energy storage value of a high-level water tower, rough operation mode of a pump station, difficulty in cooperative optimization of agricultural load and renewable energy sources, and lack of a layered regulation method for adapting to prediction errors and system disturbance. Therefore, it is necessary to provide an optimization method capable of comprehensively considering factors such as wind and light output, high-level water tower water quantity, irrigation demand, electricity price change and the like, so that efficient, safe and low-carbon operation of the agricultural comprehensive energy system is realized. Disclosure of Invention In order to overcome the defects, the invention provides a cooperative optimization method and device for an agricultural comprehensive energy system based on high-order water tower-wind-solar complementation. In a first aspect, a cooperative optimization method of an agricultural integrated energy system based on high-level water tower-wind-solar complementation is provided, and the cooperative optimization method of the agricultural integrated energy system based on high-level water tower-wind-solar complementation comprises the following steps: Substituting the wind power and photovoltaic predicted output in the day-ahead optimization period into an upper layer scheduling model and solving to obtain a day-ahead optimization result; Taking the day-ahead optimization result as a day-ahead bo