Energy Disaggregation refers to the process of separating aggregating power reading from a meter of a house/building to its individual appliances. This is generically also known as Non-Intrusive Load Monitoring (NILM)

It is a way for homeowners and commercial building owners to monitor energy consumption on an appliance-by-appliance basis without having to install dedicated sensors across the entire house or building. As the advent of smart meters is inevitable, the demand for such application provides greater value add to customers and energy companies.



How it works:


NILM builds on the basic principles of smart meter by adding another layer of algorithmic disaggregation on the data collected from the meter. In a household, each appliance has a unique energy “signature”. By analyzing the smart meter data, NILM identifies a signature for each appliance in the household, and then uses a deep learning algorithm to separate those signatures from the overall energy data of the house.

Combinatorial Optimization minimizes the difference between the sum of predicted appliance power and the measure aggregate power from the smart meter.

Where,


=> Measured aggregate power
[x1, ....., xN] => Predicted appliance power for each device


To find the minimum we use backtracking and pruning. First we find a value of one combination that becomes the upper bound. Any value that might tend to a predicted aggregate greater than this upper bound can be pruned. If any value that happens to be lower than this upper bound, that becomes the new upper bound.

Probyto is collaborating with academia, businesses and organizations to develop a prototype for SmartBill. We welcome your feedbacks and happy to partner with you or your organization. Please fill the form below to get in touch.


Request for Demo



Related Topics