| Title |
Stochastic Model Based Prediction of Occupants' Presence in Residential Apartment Buildings |
| Authors |
Kim Young-Jin ; Park Cheol-Soo |
| Keywords |
Ventilation ; Uncertainty ; Apartment Buildings ; Markov Chain ; Sensitivity Analysis ; Monte-Carlo |
| Abstract |
With the recent increased expectation of better indoor environment, the need of reliable ventilation simulation has become significant. Such simulation results vary with uncertain parameters including (1) stochastic nature of weather, (2) occupants' behavior, (3) building components, (4) uncertainties in the simulation parameters. It has been recognized that the occupant's presence and behavior significantly influenced the prediction of the ventilation rates as well as indoor air quality. The paper shows that the occupant's schedule is more influential than other inputs to ventilation simulation. In light of this, the paper present a prediction method of occupant's presence employing Markov Chain. A typical of a residential building consisting of a master room, two bedrooms, a living room and kitchen was chosen for the study. The occupant's presence for 30 families were obtained. Based on the collected data, a transition probability matrix was developed to predict the future state from the current state. The paper shows that the proposed method performs better than a conventional deterministic approach (using occupant's schedule). |