Machine learning techniques for human comfort evaluation of HVAC systems

• Main Author: Teo, Sharon Hui Ling
• Other Authors: Soh Yeng Chai
• Format: Final Year Project
• Language: English
• Published: 2014
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Control and instrumentation::Control engineering
• Online Access: http://hdl.handle.net/10356/60821
• Institution: Nanyang Technological University

This project is to explore the use of machine learning technique such as ANN, ELM etc to derive at a better human comfort analysis and evaluation of air-conditioned spaces. Very often, the human comfort are derived based on certain empirical formulae derived on certain condiitons, and may not be appropriate for tropical setttings like in Singapore. In modern HVAC systems, much more information are available about the operation conditions of the systems. These information can best be exploited by using machine learning techniques to extract the imporatnt influencing factors on human comfort. Discoveries made using the machine learning techniques can be captured and analyzed to identify the important parameters that determine the human comfort of air-condiitoned spaces in a tropical setting. With these information, the impacts from changes in the layout of the building, the operating conditions, the air flow, the temperature, the humidity etc can be readily and quickly examined with respect to human comfort.