Application of occupancy-interlinked inhabitant behavior variables for improved energy and load profiles modeling

Abstract

Most simulation tools replicate the deterministic physical behavior of households particularly in energy load with repeated typical patterns of occupants' activities and occupancy without reproducing the active occupancy and occupants' interactions within such households. In so doing, this imparts peak demand/energy inaccurate information as encountered worldwide and the exaggerated energy savings estimation undertaken by government and utilities. This study entails the performance assessment of an ANN-based approach with the application of occupancy-interlinked inhabitant behavior variables in residential households. The application of such variables reinforces the ANN model to handle uncertainty and volatility of data to ascertain adroit forecasting of energy load profiles. The model produced a good coefficient of determination ($R^{2}$) and correlation coefficient ($r$). This model is projected to contribute mostly to energy load profile modeling, energy, utilities and measurement, and verification exercise.