A Nonintrusive Load Monitoring Based on Multi-Target Regression Approach

• Published in: IEEE access Vol. 9; pp. 163033 - 163042
• Main Authors: Buddhahai, Bundit, Makonin, Stephen
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 2021; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Data aggregation; Data models; Design of experiments; Hidden Markov models; Home appliances; Home energy monitoring; Load modeling; Machine learning; machine learning process; Markov chains; multi-target regression; nonintrusive load monitoring; Performance prediction; Power factor; Predictive models; Reactive power; Regression models; Task analysis
• ISSN: 2169-3536; 2169-3536
• DOI: 10.1109/ACCESS.2021.3133292

This paper proposes an experimental design process for the application of energy disaggregation using multi-target regression, a new data learning approach in this application area. The approach shows to be a suitable model for dealing with energy disaggregation problems in which the task is to predict multiple appliances usage from the aggregate data. The experiments were conducted by analyzing AMPds2 and ECO public data sets for verifying the effectiveness of the approach. The data were analyzed through the machine learning process to select the optimal set of electrical features, learning algorithm, and model parameter so that the system resulting from the process could deliver the optimal performance for loads inference. Results of the data learning showed that the electrical features set of current (<inline-formula> <tex-math notation="LaTeX">I </tex-math></inline-formula>), real power (<inline-formula> <tex-math notation="LaTeX">P </tex-math></inline-formula>), reactive power (<inline-formula> <tex-math notation="LaTeX">Q </tex-math></inline-formula>), and power factor ( PF ) for the aggregate data and Random Forest as the base regressor for multi-target regression model could provide the best disaggregation performance. The overall predictive performance of disaggregation accuracy and F-score outperformed the benchmarking Super State Hidden Markov Model (SSHMM) and Denoising Autoencoder (DAE) network approaches.