Virtual power plant optimization scheduling method and device considering demand and output uncertainty

By constructing a confidence interval and price elasticity coefficient matrix based on the t-distribution, and combining a game theory model and the XGBoost algorithm, the problems of new energy fluctuations and multi-entity collaborative optimization in the optimal scheduling of virtual power plants are solved, achieving more accurate scheduling and stable operation.

CN122137022APending Publication Date: 2026-06-02NANJING NORMAL UNIVERSITY
View PDF 1 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
NANJING NORMAL UNIVERSITY
Filing Date
2026-04-30
Publication Date
2026-06-02

AI Technical Summary

Technical Problem

There are problems in the optimized dispatch of virtual power plants, such as the impact of fluctuations in renewable energy output, insufficient consideration of the correlation between demand-side prices and electricity load, and poor multi-entity collaborative optimization.

Method used

The uncertainty of renewable energy output is quantified by constructing a confidence interval based on the t-distribution. A price elasticity coefficient matrix is ​​set, and a game model is constructed to optimize scheduling. The price elasticity coefficient is dynamically updated through the XGBoost algorithm. Combined with load change forecasts and electricity price changes, the Nash equilibrium condition is solved to optimize multi-entity collaborative operation.

Benefits of technology

Precise quantification of the uncertainty of new energy output improves the practicality of dispatching and the efficiency of collaborative optimization; dynamic updates of the price elasticity coefficient enhance the practicality of dispatching schemes; and incorporating the balance among multiple entities into dispatching considerations improves overall operational stability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122137022A_ABST
    Figure CN122137022A_ABST
Patent Text Reader

Abstract

This invention discloses a virtual power plant optimization scheduling method and apparatus that considers the uncertainty of demand and output. The method includes: constructing confidence intervals based on historical power generation samples of new energy sources and t-distribution; setting a price elasticity coefficient; constructing an upper-level function of a game model with the revenue of the dispatch control center as the objective; constructing a first lower-level function of the game model with the shared revenue of load aggregators and users as the objective; adjusting the electricity load of load aggregators based on the price elasticity coefficient; constructing a second lower-level function of the game model with the revenue of energy aggregators as the objective; and calculating the output of new energy sources and the fluctuation range of output based on the corresponding confidence interval. By employing the above technical solution, the uncertainty and fluctuation range of new energy output are quantified relatively accurately through the t-distribution, providing effective support for dispatch; and by using the price elasticity coefficient, the correlation between demand-side prices and electricity load is incorporated into the dispatch considerations.
Need to check novelty before this filing date? Find Prior Art