Re-thinking of energy consumption classification by the patterns of occupant behaviour in dwellings: A conceptual framework

Abstract

The aim of this paper is to present the conceptual framework of a TUBİTAK funded project titled "Developing a New Methodology to Improve Housing Quality in Turkey Based on Effects of Occupant Behavior on Energy and Comfort of the Dwellings". Although the objective of the project is the development of new methodologies and tools to be used for the definition of effects of behavior profiles of housing occupants on the energy consumption and usage of this knowledge for building new houses and renovation of existing buildings, here the first step of research has been executed. By defining sensitivity of occupant behavior on energy consumption, it is planned to classify different models of occupant behavior. With data provided, the aim is to develop an "occupant behavior labeling" which rates occupants instead of buildings. To provide the mentioned data, sensitivity analysis of existing occupant behavior will be analysed based on the Monte Carlo Methodology. This methodology is one of the most used methodologies to analyse accurate distribution of possible outputs relied on inputs based on probability. Inputs for this research are (1) number of occupants for each space (for weekdays and weekend) (2) behaviour for ventilation (Window open/closed and mechanical ventilation on/off) (3) control of heating systems (radiator on/off and/or thermostat degree). These data will be provided with survey and data logging of chosen a house occupant group. As a consequence of the research, the aim is not only to rate the behavior of housing occupants but also determine occupancy groups/labels. With this approach, based on occupant's behavior labeling, the aim is to realize fast and affective applications for renovation of existing buildings. Moreover, by evaluation/assessment of houses which will be designed in the future based on determined occupant profiles allow to produce high performance dwellings.