Energy management of hybrid energy system sources based on machine learning classification algorithms

• Published in: Electric power systems research 2021-10, Vol.199, p.107436, Article 107436
• Main Authors: Musbah, Hmeda, Aly, Hamed H., Little, Timothy A.
• Format: Article
• Language: English
• Subjects: Algorithms; Confusion matrix; Decision trees; Energy management; Hybrid energy systems; Hybrid systems; K-nearest neighbors algorithm; Machine learning; Renewable energy sources; Scheduling and managing; Solar energy; Studies
• ISSN: 0378-7796; 1873-2046
• DOI: 10.1016/j.epsr.2021.107436

•Solar and wind along with gasoline and diesel sources are combined for HES.•Different algorithms are applied to forecast the optimal energy source supplier.•Optimal SAM is achieved based on supervised algorithms.•Algorithms are ranked based on their performance. Hybrid energy systems (HES) that contain renewable energy sources, such as wind and solar energy help to minimize CO2 emissions. Therefore, studying these systems to improve their performance has become one of the critical needs these days due to the environmental crisis. Within HES, energy management (EM) of HES is an essential topic that has been covered in detail by numerous studies, as errors in EM can lead to HES blackouts. Recent research has experimented with energy management strategy (EMS) to achieve optimal EM. This work aims to generate a robust forecasting model for one hour ahead of EM. The present research work has two main objectives. The first objective is to determine which energy source should supply the demand side, using different machine-learning algorithms such as Random Forest (RF), Decision Tree (DT), Gaussian Naive Bayes (Gaussian NB) and K-Nearest Neighbors (KNN). The second objective is to compare the results of these algorithms to choose the algorithm with the best performance and to rank them based on performance as well as accuracy. The work is validated using different algorithms. The results show that DT algorithm has achieved the best performance compared to the RF and Gaussian NB algorithms. KNN algorithm gives the lowest accuracy especially over class 3. The results proof that RF, DT, and Gaussian NB algorithms are reliable.