Data Driven Modelling and Analysis
There is currently a widespread and significant disparity of predicted vs. actual building energy performance, whilst the mechanisms of ‘human factor’ impact and technical risk are poorly understood. Furthermore, poor understanding of energy sub-system complexities and how to evaluate them is compounded by shortcomings in modelling/simulation, diagnostics and control approaches with regards building energy systems. This results in significant uncertainty with regards energy performance contracting, future post-occupancy regulatory compliance, building energy affordability, and ultimately the attainment of energy and carbon policy goals.
In this research, with the support of RCUK, industry and the European Commission, we utilise a growing database of both modelled/simulated and measured data from sub-system level upwards to support better understanding of such complex issues. Via comparison of modelled and monitored sub-system energy use, and the use of techniques such as relational statistical correlation, we are able to link cause and effect for both prognostic and diagnostic evaluation.
Academics: Paul Rowley, Ralph Gottschalg
Researchers: Nick Doylend, Philip Leicester, Adam Thirkill, Becky Gough, Eleni Koumpli