A Feature-Based Model for the Identification of Electrical Devices in Smart Environments.

Area of Science:

• Energy Systems Engineering
• Artificial Intelligence
• Data Science

Background:

• Smart Homes (SHs) are integral to the Smart Grid (SG), necessitating efficient energy management.
• Understanding household energy consumption is crucial for dynamic load management, cost savings, and reducing carbon emissions.
• Automated appliance identification is key for advanced monitoring and control in SHs.

Purpose of the Study:

• To propose a novel model for automatic identification of electrical appliances in Smart Homes.
• To leverage machine learning techniques for accurate appliance recognition based on energy usage patterns.
• To analyze the importance of extracted features for effective appliance identification.

Main Methods:

• Extraction of 19 distinct features from device profiles, including energy consumption, time usage, and location.
• Application of various machine learning classifiers to identify appliances based on the extracted features.
• Performance evaluation of different models and analysis of feature importance.

Main Results:

• The proposed model demonstrates effective automatic identification of appliances within Smart Homes.
• Machine learning classifiers showed varying degrees of success in appliance recognition based on the feature set.
• Feature importance analysis identified key indicators for accurate appliance identification.

Conclusions:

• The developed model provides a viable solution for automated appliance identification in SHs.
• This technology supports enhanced energy management within the SG, benefiting consumers and grid operators.
• The findings contribute to more efficient energy redistribution and a reduced carbon footprint.