A Probabilistic Model to Predict Household Occupancy Profiles for Home Energy Management Applications

Due to the impact of human lifestyle on building energy consumption, development occupants' behavior models is crucial for energy-saving purposes. In this regard, occupancy modeling an effective approach intend such a purpose. However, literature reveals that existing have limitations related representation state duration and integration variability among individuals. Accordingly, paper proposes explicit differentiated probabilistic model generate realistic daily profiles in residential buildings. The discrete-time Markov chain theory semi-parametric Cox proportional hazards (Cox regression) are used predict household profiles. proposed able capture states integrate according individuals' characteristics. Moreover, parametric analysis employed investigate these characteristics' performance consequently, select most significant input variables. A validation process conducted by comparing with previous methods, presented literature. For purpose, k crossvalidation technique utilized. Validation results demonstrate highly efficient generating