Improving sustainability of ever-changing building spaces affected by users' fickle taste: A focus on human comfort and energy use

The building spaces have been planned to satisfy the various conditions required for each use. However, actual users often utilize the building spaces by the changes in their preferences for other purposes outside the initial planning. In such situations, the existing building space and indoor control strategies may not operate properly, and the changed conditions may cause unnecessary energy use. This study presents a network based deterministic model responding to occupant's preferences that can deteriorate indoor comfort and energy efficiency. By use of two distinct situations assuming gradual changes in using spaces, a deterministic model combining two different algorithms analyses unexpected abnormal situations derived from the characteristics of spaces and occupants. In a building space where the changes in users' metabolic rate and energy use are not large, the model improves human comfort level by 4.3% for the library and 2.5% for the workroom, respectively. In the other building space where the changes in both of the factors occur greatly, the model improves energy efficiency by 26.4% as compared to the thermostat on/off model and 44.1% to the conventional neural network model, respectively. It is theoretically confirmed that the deterministic process improves thermal attributes and energy savings, and that it is appropriately responsive to gradual changes in the use of indoor spaces derived from users' preferences. (C) 2019 Elsevier B.V. All rights reserved.