Use of generalised additive models to assess energy efficiency savings in buildings using smart metering data

Abstract:

Buildings and construction together are estimated to account for 36% of global final energy use and 39% of energy-related carbon dioxide emissions globally. Increasing energy efficiency in the building sector has become a priority worldwide and especially in the European Union, although it is clear that the energy efficiency potential that lies in buildings is far from being harnessed. Given the relatively low turn-over rate of the building stock, energy efficiency retrofit appears to be a fundamental step in reducing the energy consumption and CO2 emissions in existent buildings. In this study, a framework for the evaluation of the impact of energy retrofitting measures, with a statistical learning approach, is proposed. The research was developed to enhance the data analytics system at the core of European projects SHERPA and EDI-Net, with the main goal of facilitating energy consumption monitoring in buildings and allowing analysis and evaluation of applied energy efficiency measures (EEM).

An innovative approach considering user behaviour in the evaluation of EEM impact is proposed, based on a combination of Gaussian Mixture Models and Generalised Additive Models (GAM). The method was tested in three pilot buildings in the framework of projects SHERPA and EDI-Net through the analysis of hourly smart meter consumption data and weather data. The results show the viability of this quick and cost-effective approach to evaluate the impact of applied EEM and open to further research to verify the method’s scalability to a district, city or national level when applied in a big data environment.