Deepika Rohit Chavan
Research Subjects
Deepika Rohit Chavan has a total of 3 co-authors. Here are the top co-authors by number of shared publications:
Amruta M. Khot
1 shared publicationDagadu S. More
1 shared publication- Dagadu Shankar More
Walchand College of Engineering
1 shared publication
Latest Publications
- 2 February 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1277-1280
Electric Load Disaggregation using Machine Learning Approach on IEDL Dataset: a Comparative Study
Managing energy costs is crucial because rapid urbanization increases energy demand, which affects existing energy resources. Load disaggregation is a promising non-intrusive tool for better demand management and power grid development. In this regard, feature selection and characterization are important. The paper represents a comparati ve analysis of the Indian Energy Dataset with Low-frequency (IEDL) using four different supervised machine learning models in a classification context. Five different appliances (ten years older) opted for the energy disaggregation purpose. The comparison between Random Forest (RF), Decision Tree (DT), Naïve Bayes (NB), and k-Nearest Neighbor (k-NN) classifiers was observed and assessed for performance on accuracy, precision, recall, and f-score. Furthermore, grid search is performed on RF and k-NN to prevent loss of accuracy.
- 2 February 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1277-1280
Electric Load Disaggregation using Machine Learning Approach on IEDL Dataset: a Comparative Study
Managing energy costs is crucial because rapid urbanization increases energy demand, which affects existing energy resources. Load disaggregation is a promising non-intrusive tool for better demand management and power grid development. In this regard, feature selection and characterization are important. The paper represents a comparati ve analysis of the Indian Energy Dataset with Low-frequency (IEDL) using four different supervised machine learning models in a classification context. Five different appliances (ten years older) opted for the energy disaggregation purpose. The comparison between Random Forest (RF), Decision Tree (DT), Naïve Bayes (NB), and k-Nearest Neighbor (k-NN) classifiers was observed and assessed for performance on accuracy, precision, recall, and f-score. Furthermore, grid search is performed on RF and k-NN to prevent loss of accuracy.
- Open Access
- 6 June 2022
- Published by Elsevier in Energy Reports
- Vol. 8, 701-709
IEDL: Indian Energy Dataset with Low frequency for NILM
- 6 June 2022
- Published by Elsevier in Energy Reports
- Vol. 8, 701-709
Open AccessIEDL: Indian Energy Dataset with Low frequency for NILM