Power demand-side flexible load active power prediction method

Abstract

The invention discloses a power demand-side flexible load active power prediction method The prediction method includes the steps of acquiring the historical data of the power demand-side flexible load active power, and counting and calculating the flexible load active power values under the conditions of different weather and time; determining the influence factors of the flexible load active power based on the change trend and features of the flexible load active power; constructing a demand-side flexible load model based on the conventional normal load component, the special event load component, and the random load component of the flexible load; and performing the combined prediction of the demand-side flexible load active power based on the optimal criterion of weighted average of the weight and the fitting degree. According to the method, the prediction accuracy of the demand-side flexible load active power is improved, the technological support for the safe, economic, and stable running of a power grid is provided, and the wide application and efficiency running of the demand-side electric automobile industry and the distributed power and energy storage system are greatly enhanced.

Description

technical field [0001] The invention relates to a method for predicting active power of flexible loads on the power demand side. Background technique [0002] With the country's continuous cultivation and promotion of the development of the new energy vehicle industry, pure electric vehicles, which are the main strategic orientation for the development of new energy vehicles and the transformation of the automotive industry, have achieved rapid development and a large number of demonstration applications. The typical power demand is the charging load of electric vehicles. The lateral flexible load has become the research direction and research focus at this stage. The active power prediction of flexible loads is extremely important to the safe, economical and stable operation of the power system. Accurate prediction is conducive to the formulation of power system operation schemes and power equipment maintenance plans. [0003] Many factors such as weather, holidays, and spe...

AI Technical Summary

Problems solved by technology

[0003] Many factors such as weather, holidays, and special events affect the active power curve of the demand-side flexible load, which has great randomness and complex uncertainties. There is a certain correlation between various factors, which is likely to lead to information overlap. leading to a decrease in the prediction accuracy of the forecasting model

Traditional forecasting methods can be roughly divided into parameter estimation methods and artificial intelligence methods. After continuous evolution, traditional forecasting methods mostly use indirect influencing factors or power load data sequences for modeling and analysis, and effectively use load-related economic data or sequence self-response. However, this information is not comprehensive and complete enough. Traditional forecasting models and forecasting methods can no longer meet the requirements of the ever-growing demand-side flexible load forecasting index. How to analyze the influencing factors of flexible load more comprehensively and accurately predict the flexible load Load active power has become an important research topic at present

Images