A method and device for electric load forecasting

Abstract

The present invention is applicable to the field of energy technology, and provides a power load forecasting method and device. The method includes: obtaining time series data corresponding to the time to be predicted and external factor data corresponding to the time series data; using a residual attention network to Process time series data to obtain semantic data features of time series data; use autoencoder to process external factor data to obtain external factor features of external factor data; obtain combined features according to semantic data features and external factor features; use neural network The combined features are processed to obtain the electric load forecasting result at the moment to be forecasted. The combined use of the two forecasting methods makes up for the poor accuracy of single load forecasting and solves the problem of large forecasting deviations. Accurate forecasting results are conducive to the scheduling optimization of follow-up work; the overall operation of this method is simple, ensuring the accuracy and convenience of forecasting and rapidity, saving processing resources.

Description

technical field [0001] The invention belongs to the technical field of energy, and in particular relates to a method and device for predicting electric power load. Background technique [0002] Photovoltaic is the abbreviation of solar photovoltaic power generation system. It is a new type of power generation system that uses the photovoltaic effect of solar cell semiconductor materials and can directly convert solar radiation into electrical energy. At the same time, based on the classification of solar photovoltaic power generation systems, a One is centralized, such as the large-scale northwest ground photovoltaic power generation system; the other is distributed, such as the roof photovoltaic power generation system of industrial and commercial enterprises. At present, electricity supply users are divided into industrial, commercial, residential, office, etc. The electricity load, load magnitude, and load characteristics of different users are different, and the accuracy...

AI Technical Summary

Problems solved by technology

[0003] In view of this, an embodiment of the present invention provides a method, device, terminal equipment, and computer-readable storage medium for electric load forecasting, so as to solve the technical problem that the electric load cannot be accurately predicted in the prior art, especially the photovoltaic electric load

Images