Inicio  /  Applied Sciences  /  Vol: 11 Par: 12 (2021)  /  Artículo
ARTÍCULO
TITULO

Novel Data-Driven Models Applied to Short-Term Electric Load Forecasting

Manuel Lopez-Martin    
Antonio Sanchez-Esguevillas    
Luis Hernandez-Callejo    
Juan Ignacio Arribas and Belen Carro    

Resumen

This work brings together and applies a large representation of the most novel forecasting techniques, with origins and applications in other fields, to the short-term electric load forecasting problem. We present a comparison study between different classic machine learning and deep learning techniques and recent methods for data-driven analysis of dynamical models (dynamic mode decomposition) and deep learning ensemble models applied to short-term load forecasting. This work explores the influence of critical parameters when performing time-series forecasting, such as rolling window length, k-step ahead forecast length, and number/nature of features used to characterize the information used as predictors. The deep learning architectures considered include 1D/2D convolutional and recurrent neural networks and their combination, Seq2seq with and without attention mechanisms, and recent ensemble models based on gradient boosting principles. Three groups of models stand out from the rest according to the forecast scenario: (a) deep learning ensemble models for average results, (b) simple linear regression and Seq2seq models for very short-term forecasts, and (c) combinations of convolutional/recurrent models and deep learning ensemble models for longer-term forecasts.

 Artículos similares

       
 
Ilia Zaznov, Julian Martin Kunkel, Atta Badii and Alfonso Dufour    
This paper introduces a novel deep learning approach for intraday stock price direction prediction, motivated by the need for more accurate models to enable profitable algorithmic trading. The key problems addressed are effectively modelling complex limi... ver más
Revista: Applied Sciences

 
J. D. Tamayo-Quintero, J. B. Gómez-Mendoza and S. V. Guevara-Pérez    
Objective: This study aims to introduce and assess a novel AI-driven tool developed for the classification of orthodontic arch shapes into square, ovoid, and tapered categories. Methods: Between 2016 and 2019, we collected 450 digital dental models. Appl... ver más
Revista: Applied Sciences

 
Haoyu Lin, Pengkun Quan, Zhuo Liang, Dongbo Wei and Shichun Di    
In the context of automatic charging for electric vehicles, collision localization for the end-effector of robots not only serves as a crucial visual complement but also provides essential foundations for subsequent response design. In this scenario, dat... ver más
Revista: Applied Sciences

 
Xue Yang, Jingkai Zhi, Wenjun Zhang, Sheng Xu and Xiangkun Meng    
Arctic navigation faces numerous challenges, including uncertain ice conditions, rapid weather changes, limited communication capabilities, and lack of search and rescue infrastructure, all of which increase the risks involved. According to an Arctic Cou... ver más

 
Jan Sawicki, Maria Ganzha, Marcin Paprzycki and Yutaka Watanobe    
As the largest open social medium on the Internet, Reddit is widely studied in the scientific literature. Due to its structured form and division into topical subfora (subreddits), conducted research often concerns connections and interactions between us... ver más
Revista: Algorithms