Artificial Intelligence (AI)-based identification of appliances in households from their electrical smart meter data

Principle Goal of the Project: The goal of this project is to propose an algorithm that identifies main appliances in households (like heat pumps, Electric Vehicles) from their electrical demand, and then to assess automatically the potential electrical flexibility of households.

Description of the Project: The ReFLEX Orkney (Responsive Flexibility) project aims to create a ‘smart energy island’ - developing a 'virtual energy system' in Orkney which will monitor generation, grid constraint and energy demand and then use smart control of energy technologies to manage and improve the supply-demand balance.
The smart control requires to assess which households’ electric consumption are flexible in order to target these households. This is the aim of the project. Based on households’ electric consumption data, the student(s) will design and implement an AI-based algorithm that assess if households possess appliances that are flexible (as Heat Pumps, Electric Vehicles). The algorithm will then estimate and forecast what is the flexibility potential in short term horizon.

Completion Criteria
1. A report on appliances identification from smart meter data and short term flexibility assessment
2. The Python or Matlab codes with explanation on the implementation and replication

Other Comments: 

Essential skills and knowledge: Good mathematical skills. Basic knowledge in programming and electrical engineering
Desirable skills and knowledge: Experience with thermodynamics (thermal modelling in buildings)

Supervisor name: 
Professor David Flynn
Supervisor and Deputy email addresses: 
d.flynn@hw.ac.uk
Deputy name: 
Dr Merlinda Andoni