An Ensemble Learning Classifier to Predict Net Electricity Generation from Nuclear Power Plants
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
A nuclear power plant (NPP) is an important source that can generate lowemission electricity. It is important to make appropriate predictions about electricity generation to operate an NPP very effectively, which is an essential statistic for assessing the efficiency of an NPP and its performance in terms of its actual impact on the generation of energy. The net electricity generation from NPPs can change on a monthly basis. In this research, a time-series dataset containing the net electricity generation in Thousand Megawatt Hours (TMWHs) from January 2001 to March 2023 is utilized for the prediction of net electricity generation from NPPs in the United States. The dataset, containing 267 data points, ranges from 54547 TMWH to 74649 TMWH. Data preprocessing is performed, and some ensemble learning classifiers such as Random Forest, Voting Ensemble, Bagging, and AdaBoost are applied for the experiment. The Random Forest classifier achieves the highest performance based on its 99.82% accuracy, 0.0018 root mean square error (RMSE), 0.0633 mean absolute percentage error (MAPE), and a minimum training time of 42.4 seconds. This research can aid in preventing system loss and ensuring electricity generation.
Fields of Study & Subjects
Energy Load and Power Forecasting
Forecasting Techniques and Applications
Stock Market Forecasting Methods
Publication details
- Journal / source
- CRC Press eBooks
- Pages
- 153–166
- Publication date
- 27 March 2026
- Language
- EN
- Document type
- Book chapter