Sensitivity analysis of household factors and energy consumption in residential houses: A multi-dimensional hybrid approach using energy monitoring and modeling

Abstract The energy forecast modeling has long been referred to as the function of estimating the energy consumption of a building and providing sustainable design options for energy-efficient policies. In addition to the existing conventional schemes for energy performance and energy simulation, there is the potential to connect steady-state energy modeling to energy monitoring data, along with household factors such as family size, housing design, and occupancy ratio to better understand hard-to-measure energy-related behaviors. This paper introduces a multi-dimensional hybrid approach that combines multiple interactions between observation-based and simulation-based data using energy modeling’s graphical interface software. Compared to the observed energy consumption, the model samples show goodness-of-fits with the Pearson coefficient of determination in the range of 67% to 91% probability of the energy model data. Accordingly, research has carried out specific air conditioning set points and schedules within each household. Finally, the variation of simulated household models illustrates the greatest effect of air conditioning setpoint on savings of 20% to 60%, compared to the baseline usage. Sensitivity analysis shows that larger household size, higher occupancy ratio, lower thermal resistance for wall insulation, and slighter airflow rate can reduce the end-use and the gross site or source energy. The study underlines that investigations combining household characteristics, occupant schedules, and energy monitoring data are crucial to stimulate the application of building energy modeling in the early design stage.