An Indoor Occupant Distribution Estimation Method Based on Improved Stay-Time Algorithm

The distribution of indoor occupants plays a vital role in intelligent building management since occupants are the operators of facilities in building. Information of occupant number in each zone is helpful to both intelligent control of building systems under normal conditions and evacuation in emergency situations. In general, it is difficult to obtain an optimal estimation of occupant distribution because of the sensor error, the randomness of human movement and the large state and action spaces. In this paper, we take this important problem into consideration and make the following contributions. First, we formulate the moving process of occupant as a Markov chain and use the event-based optimization (EBO) theory to model the overall indoor movements of occupants. Second, we develop a new method based on improved stay-time algorithm and utilize it to estimate the number of occupants. Third, the performance of this method is evaluated through numerical results. It is our hope that this work brings insight into the problem of indoor occupant distribution estimation.