Data mining: Geo-dependent energy supply in relation to urban form

Supervisor: Nahid Mohajeri  / PhD student: Dan Assouline

Funding: Swiss Competence Centre for Energy Research “Future Energy Efficient Buildings and Districts” (SCCER – FEEB&D) – Urban Decentralised Energy Systems (WP3)
Project duration: July 2014 – January 2016

Decentralized systems will require novel types of geo-spatial databases, new methods of urban pattern analysis, and new technology and modelling approaches. The aim is to develop geo-dependent energy-related tools, based on Geographic Information Systems (GIS), together with remote sensing and geo-statistics, as well as machine learning so as to identify the potential renewable energy resources (wind, solar, biomass, geothermal heat, and waste heat) and to model their spatio-temporal distributions from large to neighbourhood scale across Switzerland.

Spatio-temporal solar irradiance model using machine learning (Support Vector Regression) by Dan Assouline