A short-term power load prediction method based on a GRU neural network and transfer learning

Abstract

The invention discloses a short-term power load prediction method based on a GRU neural network and transfer learning, and belongs to the field of power load prediction. A short-term power load prediction problem is solved. The invention provides a short-term power load prediction method based on a GRU neural network and transfer learning. A GRU-based recurrent neural network is utilized to solvethe problem that a traditional neural network cannot extract time characteristics, auxiliary information input influencing load change is expanded, including influence factors such as date, temperature and weather, a Dropout layer and a normalization layer are added to avoid the overfitting problem, and the accuracy of load prediction is improved; Historical knowledge is migrated through migrationlearning, a network which is normally put into use is adjusted, retraining and fine tuning are carried out through target prediction data, the value of the historical data is brought into play, and the precision and efficiency of load prediction are further improved.

Description

technical field [0001] The technical field of the present invention is the field of load forecasting, specifically a short-term power load forecasting method based on a Gated Recurrent Unit (GRU) neural network and transfer learning. Background technique [0002] The electric power industry is an important basic industry for the development of the whole country. With the rapid development of science and economy, the scale of power system continues to expand and the operating conditions become increasingly complex. Power load forecasting is the premise and basis of power grid decision-making and control. It has a key influence on the safe, reliable and economical operation of the power system. It is an important part of the power grid energy management system and power distribution management system. High-accuracy power load forecasting can not only provide a decision-making basis for the power grid to formulate a reasonable construction plan, determine the demand and locati...

AI Technical Summary

Problems solved by technology

[0007] The purpose of the present invention is to propose a short-term power load forecasting method based on GRU neural network and transfer learning for the load forecasting problem that requires increasing accuracy and efficiency; the use of GRU-based cyclic neural network solves the problem that traditional neural networks cannot extract The problem of time characteristics; increase the input of auxiliary information that affects the load change, improve the prediction accuracy; and add the dropout layer and the normalization layer to avoid the over-fitting problem; transfer the historical knowledge through transfer learning, give full play to the value of historical data, and make the load prediction Accuracy and efficiency further improved

Images