Loading...

Indoor Air Quality and Brightness Control Using Fuzzy Logic in Intelligent Environments

Boroomandfar, Parviz | 2018

361 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 50568 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Manzuri, Mohammad Taghi; Nazari Shireh Jeini, Ali Asghar
  7. Abstract:
  8. The energy consumed by lighting, heating, ventilating, and air conditioning (LHVAC) systems have been increasing over the last decades. Thus, improving the efficiency of LHVAC systems has gained the attention of industry and academia. This concern has posed challenges for controlling and optimizing LHVAC systems. The traditional methods, such as on/off and PID approaches, usually involve conditions and parameters that may not hold in practice since LHVAC systems are complex, nonlinear, and dynamic. Also, controllers are one of the important components of a method for its efficiency. This research focuses on controlling LHVAC systems. Fuzzy logic controllers one of the best controller for this systems. These controllers are utilized to make an intelligent environment with the low cost and maximum efficiency. This research use three type of fuzzy inference for controlling; Mamdani inference systems, Takagi-Sugeno inference systems, and adaptive neuro-fuzzy inference systems. In the end of this research, the controller with adaptive neuro-fuzzy inference system has better results than other controllers. It has 44 percent performance in energy saving and Mean absolute error for predicated mean vote is equal to 0.061. This numbers for the controller with Sugeno fuzzy inference system is 41 percent and 0.075 and for Mamdani inference system is 32 percent and 0.078
  9. Keywords:
  10. Fuzzy Control ; Indoor Quality ; Adaptive Neuro-Fuzzy Inference System (ANFIS) ; Energy Consumption Optimization ; Ambient Intelligence ; Lighting, Heating, Ventalation, Aircondition (LHVAC) ; Predicated Mean Vote (PMV)

 Digital Object List

 Bookmark

No TOC