期刊
ENERGY
卷 115, 期 -, 页码 1042-1054出版社
PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/j.energy.2016.09.017
关键词
Primary energy consumption; Combined forecasting model; Input-output; Grey methods
资金
- National Natural Science Foundation of China [70701034, 71173210, 61273208]
- Energy Foundation (Research on key mechanism and institution for sustainable energy development in China) [G-1306-18458]
- Jose Castillejo Outgoing International Fellowship of the Spanish Government- Spanish Ministry of Education, Culture and Sport [CAS12/00159]
A combined forecast of Grey forecasting method and neural network back propagation model, which is called Grey Neural Network and Input-Output Combined Forecasting Model (GNF-10 model), is proposed. A real case of energy consumption forecast is used to validate the effectiveness of the proposed model. The GNF-IO model predicts coal, crude oil, natural gas, renewable and nuclear primary energy consumption volumes by Spain's 36 sub-sectors from 2010 to 2015 according to three different GDP growth scenarios (optimistic, baseline and pessimistic). Model test shows that the proposed model has higher simulation and forecasting accuracy on energy consumption than Grey models separately and other combination methods. The forecasts indicate that the primary energies as coal, crude oil and natural gas will represent on average the 83.6% percent of the total of primary energy consumption, raising concerns about security of supply and energy cost and adding risk for some industrial production processes. Thus, Spanish industry must speed up its transition to an energy-efficiency economy, achieving a cost reduction and increase in the level of self-supply. (C) 2016 Elsevier Ltd. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据