A Saturated Power Load Forecasting Method Based on Long Short-term Memory Neural Network

Abstract

The present invention relates to a method for forecasting saturated power load based on long-term short-term memory neural network, including: step 1, setting the influencing factors and forecasting scenarios of saturated load; The long-short-term memory neural network prediction model predicts the saturated power load of the area to be predicted, and obtains the power load saturation time and saturation scale of the area to be predicted under different prediction scenarios. Compared with the prior art, the present invention has the advantages of meeting the load time sequence continuity and the time delay requirements of influence factors on the load.

Description

A Saturated Electricity Load Forecasting Method Based on Long Short-Term Memory Neural Network technical field The present invention relates to a kind of power system load forecasting technology, especially relate to a kind of neural network based on long short term memory A method for predicting the saturated power load of the network. Background technique [0002] The saturated power load forecast refers to the prediction of the time when the regional power load enters saturation and the scale of saturated power consumption. The power load saturation scale and saturation time are affected by many factors, including regional population and economic characteristics. Saturation load gauge The mode and saturation time are the basis for the long-term planning goals of the power grid, and are of great significance for coordinating the short-term power grid construction. Traditional saturated power load forecasting method comprises: adopt improved K-means clustering algori...

AI Technical Summary

Problems solved by technology

These studies did not take into account the delay characteristics of load timing continuity and the influence of correlation factors

Images