Incremental Distribution Network Forecasting for Different Industries Based on Long and Short Term Memory Network

Abstract

In order to improve the quantitative evaluation and prediction ability of power grid investment benefit and solve the optimization problem of investment rhythm caused by the increase of user load in the new park, an incremental distribution network load forecasting method based on long-term and short-term memory network is proposed. Considering multiple influencing factors of investment decision-making, grey correlation analysis is carried out on the factors affecting saturated load, so as to quantitatively determine the impact Degree and size. Then, the influencing factors are used as independent variables, and the demand of electric power or electricity is used as the dependent variables to establish the prediction model to realize the medium and long-term load high-precision forecasting under the condition of small sample and high uncertainty.

Keywords

• medium-and long-term load forecasting
• deep learnining
• industrial classification
• incremental distribution network