CN-122022506-A - Virtual power plant double-layer optimization regulation strategy formulation method, system and storage medium

CN122022506ACN 122022506 ACN122022506 ACN 122022506ACN-122022506-A

Abstract

The invention provides a method, a system, a storage medium and equipment for formulating a double-layer optimization regulation strategy of a virtual power plant. The method comprises the steps of obtaining operation data of a virtual power plant, processing wind power uncertainty based on the operation data, obtaining wind power predicted output, and formulating a double-layer optimization regulation strategy of the virtual power plant according to the operation data and the wind power predicted output. The method for formulating the double-layer optimization regulation strategy of the virtual power plant improves economic and environmental benefits.

Inventors

WANG JUN
HUANG YAN
CHEN JINDIAN
HAO YINGPENG
LIANG YAN
Chi Lixun
Tong Yongjing

Assignees

中国石油天然气股份有限公司

Dates

Publication Date: 20260512
Application Date: 20241104

Claims (14)

1. A method for formulating a double-layer optimization regulation strategy of a virtual power plant is characterized by comprising the following steps: Acquiring virtual power plant operation data; based on the operation data, processing wind power uncertainty and obtaining wind power prediction output; And according to the operation data and the wind power predicted output, formulating a double-layer optimization regulation strategy of the virtual power plant.
2. The method of claim 1, wherein the processing wind power uncertainty based on the operational data to obtain a wind power predicted force comprises: Wind power uncertainty is processed based on Latin hypercube sampling scene method, wind power predicted output is obtained, and reliability of a power system is evaluated.
3. The method of claim 2, wherein the latin hypercube based scenario approach processes wind power uncertainty, obtains wind power predicted output, and evaluates reliability of the power system, comprising: defining wind power output prediction error distribution compliance normal distribution based on historical data in the operation data; generating a plurality of wind power output samples from the prediction error distribution of the wind power plant output by using Latin hypercube sampling, wherein each wind power output sample represents a wind power output scene; adopting synchronous back substitution reduction method to perform scene reduction on the wind power output sample, and screening out a typical scene; And evaluating the reliability of the power system by using the typical scene.
4. The method of claim 1, wherein the two-layer optimization regulation strategy comprises an upper layer regulation strategy and a lower layer regulation strategy; the upper layer regulation strategy is that based on historical data in the operation data and the wind power predicted output, a daily declaration output model is established by taking the maximum daily net gain of a virtual power plant as an objective function, and daily optimization is carried out to obtain a daily declaration output plan; The lower layer regulation strategy is to establish a real-time optimization scheduling model based on the daily declaration output plan and real-time data in the operation data by taking the maximum real-time benefit as an objective function, and perform real-time optimization to obtain a daily real-time scheduling plan.
5. The method of claim 4, wherein the pre-day declaration force model further includes: calculating electricity selling and heat selling income according to the historical data, and calculating the daily net income of the virtual power plant by combining the running cost of each power generation unit; and (5) formulating a real-time daily scheduling plan by combining constraint conditions.
6. The method of claim 4, wherein the real-time optimized scheduling model further comprises: Acquiring ultra-short-term wind power predicted output according to the real-time data, and carrying out real-time adjustment on the output power of controllable equipment by combining the daily declaration output plan so as to maximize the real-time benefit of the virtual power plant; and combining the punishment cost of the real-time output and the daily declaration output deviation and constraint conditions to formulate a daily real-time scheduling plan.
7. The method according to any of claims 1-6, characterized in that the real-time optimization is performed by modifying the output power of the controllable device by means of a rolling-domain optimization.
8. A virtual power plant bi-level optimization regulation strategy formulation system, the system comprising: The data acquisition module is used for acquiring virtual power plant operation data; the wind power prediction module is used for processing wind power uncertainty based on the operation data and obtaining wind power prediction output; And the strategy making module is used for making a double-layer optimization regulation strategy of the virtual power plant according to the operation data and the wind power predicted output.
9. The system of claim 8, wherein the wind power prediction module is further configured to: Wind power uncertainty is processed based on Latin hypercube sampling scene method, wind power predicted output is obtained, and reliability of a power system is evaluated.
10. The system of claim 9, wherein the wind power prediction module is further configured to: defining wind power output prediction error distribution compliance normal distribution based on historical data in the operation data; generating a plurality of wind power output samples from the prediction error distribution of the wind power plant output by using Latin hypercube sampling, wherein each wind power output sample represents a wind power output scene; adopting synchronous back substitution reduction method to perform scene reduction on the wind power output sample, and screening out a typical scene; And evaluating the reliability of the power system by using the typical scene.
11. The system of claim 8, wherein the policy formulation module is further configured to: the double-layer optimized regulation strategy comprises an upper regulation strategy and a lower regulation strategy; the upper layer regulation strategy is that based on historical data in the operation data and the wind power predicted output, a daily declaration output model is established by taking the maximum daily net gain of a virtual power plant as an objective function, and daily optimization is carried out to obtain a daily declaration output plan; The lower layer regulation strategy is to establish a real-time optimization scheduling model based on the daily declaration output plan and real-time data in the operation data by taking the maximum real-time benefit as an objective function, and perform real-time optimization to obtain a daily real-time scheduling plan.
12. The system of any of claims 8-11, wherein the policy formulation module is further configured to: And correcting the output power of the controllable equipment through rolling time domain optimization to perform real-time optimization.
13. A computer-readable storage medium storing one or more programs, characterized in that, The virtual power plant double-layer optimization regulation strategy formulation method of any one of claims 1-7 can be implemented when the one or more programs are executed.
14. An electronic device comprising a processor, a communication interface, the computer-readable storage medium of claim 13, and a communication bus, wherein the processor, the communication interface, and the computer-readable storage medium are in electronic communication with each other via the communication bus, The processor is configured to execute a program stored in a computer-readable storage medium.

Description

Virtual power plant double-layer optimization regulation strategy formulation method, system and storage medium Technical Field The invention belongs to the technical field of electric power, and particularly relates to a method, a system, a storage medium and equipment for formulating a double-layer optimization regulation strategy of a virtual power plant. Background In recent years, with the continuous adjustment of the structure of Chinese electric power energy, the proportion of renewable energy accessed into a power grid is higher and higher. However, renewable energy power generation output represented by wind power and photovoltaic power generation has volatility and randomness, and has weak controllability and difficult accurate prediction. This volatility and randomness presents a significant challenge for safe operation of the power system. The wind-combustion-storage virtual power plant aggregates the distributed power (wind power), the controllable power (gas turbine), the controllable load, the energy storage and other resources scattered in different areas through advanced control, measurement and communication technologies, but due to the randomness of wind power output and the dynamic fluctuation of user load demands, the wind-combustion-storage virtual power plant often has inconsistency and unmatched performance in time dimension and space dimension. Disclosure of Invention Aiming at the defects of the prior art, the invention provides a method, a system, a storage medium and equipment for formulating a double-layer optimization regulation strategy of a virtual power plant, which are used for processing wind power uncertainty based on virtual power plant operation data, accurately predicting wind power predicted output, formulating a double-layer optimization regulation strategy of the virtual power plant according to the wind power predicted output and the operation data, and can effectively solve the problem that wind power output is random and optimize regulation of the virtual power plant from multiple dimensions, thereby improving economic and environmental benefits. The invention is realized by the following technical scheme: Acquiring virtual power plant operation data; based on the operation data, processing wind power uncertainty and obtaining wind power prediction output; And according to the operation data and the wind power predicted output, formulating a double-layer optimization regulation strategy of the virtual power plant. Alternatively to this, the method may comprise, Based on the operation data, processing wind power uncertainty to obtain wind power predicted output force, including: Wind power uncertainty is processed based on Latin hypercube sampling scene method, wind power predicted output is obtained, and reliability of a power system is evaluated. Alternatively to this, the method may comprise, The scene method based on Latin hypercube sampling processes wind power uncertainty, obtains wind power predicted output, and evaluates reliability of a power system, and comprises the following steps: defining wind power output prediction error distribution compliance normal distribution based on historical data in the operation data; generating a plurality of wind power output samples from the prediction error distribution of the wind power plant output by using Latin hypercube sampling, wherein each wind power output sample represents a wind power output scene; adopting synchronous back substitution reduction method to perform scene reduction on the wind power output sample, and screening out a typical scene; And evaluating the reliability of the power system by using the typical scene. Alternatively to this, the method may comprise, The double-layer optimized regulation strategy comprises an upper regulation strategy and a lower regulation strategy; the upper layer regulation strategy is that based on historical data in the operation data and the wind power predicted output, a daily declaration output model is established by taking the maximum daily net gain of a virtual power plant as an objective function, and daily optimization is carried out to obtain a daily declaration output plan; The lower layer regulation strategy is to establish a real-time optimization scheduling model based on the daily declaration output plan and real-time data in the operation data by taking the maximum real-time benefit as an objective function, and perform real-time optimization to obtain a daily real-time scheduling plan. Alternatively to this, the method may comprise, The day-before declaration output model further includes: calculating electricity selling and heat selling income according to the historical data, and calculating the daily net income of the virtual power plant by combining the running cost of each power generation unit; and (5) formulating a real-time daily scheduling plan by combining constraint conditions. Alternatively to this, the method may comprise, The real-time