Real-time recommendations for energy-efficient appliance usage in households

Area of Science:

• Energy Systems
• Artificial Intelligence
• Human-Computer Interaction

Background:

• Household energy consumption is significantly influenced by user behavior.
• Optimizing energy usage offers economic savings and environmental benefits.
• Existing systems often require appliance-level monitoring, which is not always feasible.

Purpose of the Study:

• To propose a framework for assisting households in improving energy efficiency.
• To provide real-time recommendations for efficient appliance use based on user behavior.
• To develop an end-to-end solution for energy disaggregation, profiling, and recommendation.

Main Methods:

• Developing a framework for household-specific and appliance-specific energy consumption profiling.
• Utilizing non-intrusive load monitoring (NILM) and data mining for energy disaggregation.
• Implementing pattern mining and a recommendation engine with an intuitive user interface.

Main Results:

• The proposed framework accurately captures household energy consumption patterns.
• Generated recommendations align with actual household energy habits.
• Potential to reduce household energy consumption by 2-17%.

Conclusions:

• User behavior is a key factor in household energy consumption.
• The developed framework effectively assists households in reducing energy usage.
• The system offers a practical approach to energy efficiency through personalized recommendations.