Study of Adaptive Learning Models for Human Comfort Management in Intelligent Building

• Main Authors: Ming-Ruei Lee, 李明叡
• Other Authors: Shuo-Yan Chou
• Format: Others
• Language: en_US
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/29852889119428625515

碩士 === 國立臺灣科技大學 === 工業管理系 === 100 === As time wheel move forward, people have been paid attention to their quality of life ever more and have wanted to improve comfort and health in their living spaces. In order to improve the environment become more comfortable, thermal comfort plays one of important roles in this issue. Besides, energy efficiency in buildings become a major world-wide challenge over the past few years due to the growth of energy costs, energy consumption and environmental impacts, especially those related to global warming. The purpose of this study is to develop a system that can provide comfortable environment for individual while save energy. In this study, we designed an adaptive thermal comfort system through fuzzy inference system (FIS) and adaptive-network-based fuzzy inference system (ANFIS). We not only simplify the complex PMV equation by neuro–fuzzy system but also provide more suitable comfortable environment for individual. It provides an optimal setting of temperature which approximates the user's current PMV without user preferences through air velocity calculation, fuzzy PMV calculation, and modification of temperature calculation. Furthermore, it provides three different scenarios for user choose which are single person scenario, multiple people scenario, and energy-saving scenario. Final, it can find the correlation between variables and PMV values by learning process which through adaptive-network-based fuzzy inference system so user no needs to adjust next time.