Typical daily occupancy profiles of express hotels and its stochasticity effect on building heating and cooling loads

Occupancy schedule is a key input parameter affecting building performance simulation. Current research mainly applies the fixed occupancy schedules proposed in building codes, which is unsuitable for representing actual occupancy status. Unreasonable occupancy schedules will affect the prediction results of building energy consumption. This study collects the real-time occupancy data of six express hotels with 744 guest rooms, and determines typical occupancy profiles in three scenarios (single day of the year, spring/summer/autumn/winter, weekdays/weekends & holidays) by using the clustering method. The typical occupancy profiles for single day of the year are divided into seven categories based on different samples (all samples from the six hotels, and samples from each hotel). A Monte Carlo-based model is developed to convert the probability distribution functions of typical occupancy profiles to a time series of random occupancy (ROTS). Next, building heating and cooling load differences caused by the fixed occupancy schedules and the ROTS, three ROTS in different scenarios, and seven categories of the ROTS based on different samples are compared and analyzed by simulation. Clustering results indicate that the typical occupancy profiles can be divided into four main categories: 'occupied for two nights', 'occupied from 0:00 to 14:00', 'occupied from 14:00 to 23:00' and 'vacant room', respectively. Simulation results indicate that annual building loads resulting from ROTS is 72 kW h/(m(2)center dot a), 30%-50% lower than that resulting from the fixed occupancy schedules. The differences in the heating and cooling loads resulting from three ROTS in different scenarios and seven categories of the ROTS based on different samples are relatively small. Absolute error and root mean square percentage error are within 10% and 8%, respectively. Therefore, it is recommended that the ROTS for single day of the year processed from a large number of actual occupancy status can be used as an occupancy schedule to predict and assess the building heating and cooling loads. These conclusions can contribute greatly to accurate building load simulation and the optimization of building design strategies for express hotels.