Towards Trustworthy Energy Disaggregation: A Review of Challenges, Methods, and Perspectives for Non-Intrusive Load...

Area of Science:

• Energy Systems Engineering
• Artificial Intelligence
• Signal Processing

Background:

• Non-intrusive load monitoring (NILM) disaggregates total power consumption into individual appliance usage.
• Early research focused on mathematical formulation using machine learning, while current efforts emphasize practical, real-world applications.
• The field is maturing, with a growing focus on algorithm complexity, transferability, reliability, and trustworthiness.

Purpose of the Study:

• To provide a comprehensive literature review of NILM methods specifically for residential appliances.
• To analyze and summarize outcomes from recent scholarly articles on NILM.
• To bridge the gap between theoretical NILM and practical, trustworthy applications.

Main Methods:

• Comprehensive literature review of recently published scholarly articles on NILM for residential appliances.
• Analysis and summarization of the performance and characteristics of various NILM methods.
• Identification of key research dilemmas and considerations for practical NILM implementation.

Main Results:

• Identified key trends and advancements in NILM techniques for residential energy disaggregation.
• Highlighted the shift towards practical NILM applications and the associated challenges.
• Summarized the outcomes of numerous recent studies, providing a consolidated view of the state-of-the-art.

Conclusions:

• There is a critical need to transition traditional NILM models into practical and trustworthy frameworks for real-world deployment.
• Addressing issues of complexity, transferability, reliability, and trustworthiness is paramount for the mature NILM era.
• Future research should focus on developing robust and dependable NILM solutions for widespread adoption in residential settings.