Short Term Load Forecasting Method for Power System Based on Neural Network
Authors:
- 27 December 2024
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 481-484
Abstract
A prediction method has been proposed for short-term electricity load forecasting. Collect historical load time series data of multiple power systems and historical exogenous variable time series data at the same time scale according to the set time step, and construct the original dataset; Preprocess the original dataset to obtain a complete and usable sample set; Constructing a sequence to sequence (seq2seq) LSTM-seq2seq load forecasting model using long short-term memory neural networks; Train the LSTM seq2seq load forecasting model constructed based on the sample set; Preprocess the historical load time series data and historical exogenous variable time series data within the set time step before the prediction period, and input them into the trained LSTM seq2seq load prediction model to obtain the power load prediction results during the prediction period. The proposed method can efficiently and accurately achieve short-term power load forecasting.Keywords
This publication has 4 references indexed in Scilit:
- Short-term load forecasting of regional integrated energy system based on spatio-temporal convolutional graph neural networkElectric Power Systems Research, 2024
- Short-term load forecasting for microgrid energy management system using hybrid SPM-LSTMSustainable Cities and Society, 2023
- A Novel Evolutionary-Based Deep Convolutional Neural Network Model for Intelligent Load ForecastingIEEE Transactions on Industrial Informatics, 2021
- A self-partitioning local neuro fuzzy model for short-term load forecasting in smart gridsEnergy, 2020