Conference Paper

Proceedings of eSim 2018: 10th Conference of IBPSA-Canada

An open data science approach for building performance studies using refitXML and Jupyter Notebooks

Steven K Firth, Gareth Cole, Tom Kane, Farid Fouchal, Tarek M Hassan

Abstract: Open data science is increasingly viewed as a priority for collaboration and advancement of scientific disciplines. This paper presents two open data science methods for building performance studies i) a method for storing building performance information using an open data format, including the creation of a custom XML schema and the novel use of both xml and csv formats for data storage. ii) a method for carrying out building performance analysis in an open and reproducible manner using the Jupyter Notebook tool. A Notebook is developed to demonstrate the potential of this technique for building performance modelling and analysis. The work is based on the open-access REFIT Smart Home Dataset (DOI10.17028/rd.lboro.2070091.v1), a published dataset of building performance information in 20 UK homes. The dataset includes detailed building survey information and over 1.3 billion sensor readings.
Keywords: Building performance, Smart Homes, Open data science, XML, Jupyter Notebook
Pages: 473 - 481
Paper: