Abstract
The residential sector is one of the major consumers of energy produced in the world. According to International Energy Balances (IEA, 2013), the residential sector demand represents about a quarter of the primary energy used in the world. Therefore, most energy efficiency programmes targeting large savings on a national or regional level pay particular attention to the opportunities in the residential sector. Lighting retrofitting on a large number of sites constitutes one of the most used strategies of energy conservation in the residential sector. However, given the large number of sites involved in this type of project, conventional measurement and verification (M&V) techniques based on the audit of each site, are not cost effective. Often, a statistical assessment approach based on the audit of a limited number of sites is the methodology used to mitigate the cost and the logistical challenges associated with the project. The major challenge in projects of this nature is to accurately estimate the energy consumption of a large number of sites using the measurement performed on a sample of sites selected from the overall population. In this research, baseline methodologies used in a selected number of light retrofitting projects have been analysed and, based on the observations made during this analysis, some improvements are suggested. The proposed methodology has been tested on a number of residences located on the premises of the University of Johannesburg. This paper describes the existing baseline methodologies and presents the improvements suggested to enhance the credibility of M&V results. The key results of the experimental phase of this project are also presented in this paper.