AUTHOR=Zhang Daohua , Jin Xinxin , Shi Piao , Chew XinYing 

TITLE=Real-time load forecasting model for the smart grid using bayesian optimized CNN-BiLSTM

JOURNAL=Frontiers in Energy Research

VOLUME=Volume 11 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/energy-research/articles/10.3389/fenrg.2023.1193662

DOI=10.3389/fenrg.2023.1193662

ISSN=2296-598X

ABSTRACT=Smart grid is a new type of power system with features of real-time response, efficient energy utilization, and sustainability. However, due to its complexity and uncertainty, how to effectively perform real-time prediction is an important challenge. This paper proposes an attention mechanism-based convolutional neural network (CNN) combined with Bi-directional Long Short-Term Memory BiLSTM for a smart grid real-time prediction model.CNN is a feed-forward neural network with a two-dimensional matrix as input. The main feature of CNN is its ability to process multi-channel input data. BiLSTM can capture the semantic dependencies in both directions, effectively preventing gradient explosion and gradient disappearance while better capturing the longer-distance dependencies. CNN-BiLSTM extracts the data for features, then optimized by Bayes. The model can further improve the reliability and efficiency of the smart grid by collecting real-time data from the power system, including power, load, weather, and other factors, and using deep learning algorithms to forecast real-time load.