A Probability-based Approach Resolving Thermal Preference Conflicts through Optimal Cubicle Assignment

Providing an acceptable thermal environment for occupants has been the long-term objective of HVAC systems. However, it is challenging to ensure every occupant feels thermally comfortable in a shared space owing to individual differences in thermal comfort. We present our approach of resolving thermal preference conflicts and providing personalized thermal conditioning for multi-occupants in a shared space without additional personal comfort systems. This approach relies on probability-based personal comfort profiles and uneven temperature distribution of a space to calculate each occupant's thermal comfort probabilities given temperatures of cubicles. A differential evolution algorithm is employed to find the optimal matches of occupants with cubicles and generates a solution that has the highest average thermal comfort probability. We compare the average thermal comfort probability of optimal assignment with that of the worst assignment. The results suggest that our approach through cubicle assignment has great performance in improving occupants' thermal comfort in a shared space.