Chapter 09 · Sustainability & AISustainability & AI
Energy disaggregation
Definition
Inferring the consumption of individual appliances from a building's aggregate meter signal, using signal processing or machine learning — also called non-intrusive load monitoring.
References
This source provides part of the technical or institutional basis for the definition.
This source supports the explanation of how the term is applied, measured or governed in practice.
Overview
What it means
The technique dates to Hart's foundational 1992 paper on non-intrusive appliance load monitoring, which showed that device 'signatures' could be separated from a single measurement point. Modern implementations apply machine learning to smart-meter data.
How it is used
Utilities and efficiency programmes use disaggregation to give households and facility managers appliance-level feedback without installing sub-meters, targeting the largest loads and verifying savings from retrofits and behaviour programmes.
Why it matters
Feedback is among the cheapest efficiency interventions, and disaggregation makes it scalable. It also raises the privacy side of the sustainability bargain: fine-grained consumption data reveals household behaviour, requiring careful governance.