.png){kind=icon}
.png){kind=icon}
Return
Effective energy consumption forecasting using enhanced bagged echo state network
.png){kind=icon}
PRE
AI
DOI：10.1016/j.energy.2019.116778.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Precise analysis and forecasting of energy consumption not only affects energy security and environment of a nation but also provides a useful decision basis for policy makers. This study proposes a new enhanced optimization model based on the bagged echo state network improved by differential evolution algorithm to estimate energy consumption. Bagging is applied to reduce forecasting error and improve generalization of network. Further, three parameters of echo state network are optimized using differential evolution algorithm. Thus, the proposed model combines the merits of three techniques which are echo state network, bagging, and differential evolution algorithm. The proposed model is applied to two comparative examples and an extended application to verify its accuracy and reliability. Results of the comparative examples show the proposed model achieves better forecasting performance compared with basic echo state network and other existing popular models. Mean absolute percentage error of the proposed model is 0.215% for total energy consumption forecasting of China. Therefore, the proposed model can be a satisfactory tool for forecasting energy consumption because of its high accuracy and stability. (C) 2019 Elsevier Ltd. All rights reserved.
Keywords:
Energy consumption forecasting
Echo state network
Bagging
Differential evolution
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.