首页|A robust stacking model for predicting oil and natural gas consumption in China
A robust stacking model for predicting oil and natural gas consumption in China
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
Taylor & Francis
Accurate prediction of oil and natural gas consumption (ONGC) is crucial forenergy security and greenhouse gas emission control. This study usesmachine learning to improve forecast accuracy by transforming time seriespredictions into supervised learning models. A novel stacking learningmethod, with added cross-validation, enhances model diversity and robustness.The key findings are: (1) The stacking model outperforms base modelsin predicting China’s ONGC. It achieves R2 scores of 94.44% for oil and98.33% for natural gas, with corresponding RMSE scores of 0.5325 and0.2919. (2) When comparing the scores of the models in the validation setusing cross-validation, it can be observed that the stacking model exhibitsthe most consistent performance. (3) Through the diversification of models,the stacking approach enhances robustness and achieves better generalizationon new datasets. The study provides fresh insights into model stackingfor energy consumption prediction.
Machine learningstackingoil consumptionnatural gas consumptionprediction
Yali Hou、Qunwei Wang、Tao Tan
展开 >
College of Information Engineering, Nanjing Xiaozhuang University, Nanjing, China
College of Economics andManagement, Nanjing University of Aeronautics and Astronautics, Nanjing, China
College of Public Administration,Nanjing Agricultural University, Nanjing, China
NETL
NSTL
Taylor & Francis
