Forecasting the Wind and Solar Power using Deep Learning based techniques
作者
Vidushi Bajpai,Shantanu Joshi,Hen‐Geul Yeh
标识
DOI:10.1109/igessc59090.2023.10321764
摘要
With the continuous development of the energy industry, it became necessary to meet the needs of consumers and reduce power redundancy. The task can be achieved by accurate prediction of future energy demand and the production. With the accurate prediction, the reliability of power networks can be enhanced and the power allocation can be optimized. This paper illustrates an in-depth analysis of forecasting the solar and wind power for the European electricity system. The dataset contains information about the load data each hour for 1494 buses. The main objective is to compare the predictive power by Feedforward Neural Network (FFNN) and Long Short-Term Memory (LSTM). In this article, we proposed the method to predict the total load of all the buses for later hour’s data using multistep forecasting. The results from experiments shows that LSTM outperforms the FFNN because of its long-term and short-term pattern memorizing capabilities. The proposed system can contribute to the intelligent energy system, opening the doors for sustainable and efficient energy landscape.