GA-based fuzzy controller design for tunnel ventilation systems
Baeksuk Chu, Dongnam Kim, Daehie Hong, Jooyoung Park, Jin Taek Chung, Jae-Hun Chung and Tae-Hyung Kim
Graduate School, Department of Mechanical Engineering, Korea University, 5-1, Anam-dong, Sungbuk-gu, Seoul, 136-701, South Korea
Department of Mechanical Engineering, Korea University, 5-1, Anam-dong, Sungbuk-gu, Seoul, 136-701, South Korea
Department of Control and Instrumentation Engineering, Korea University Jochiwon, Chungnam, 339-700, South Korea
Department of Mechanical Engineering, Stevens Institute of Tech., Castle Point on Hudson, Hoboken, NJ 07030, USA
Fire and Engineering Services Research Department, Korea Institute of Construction Technology, Kyunggi-do, 411-712, South Korea
Available online 13 June 2007.
Abstract
The main purpose of a tunnel ventilation system is to maintain CO pollutant concentration and visibility index (VI) under an adequate level to provide drivers with a comfortable and safe driving environment. Moreover, it is necessary to minimize power consumption used to operate the ventilation system. To achieve the objectives, fuzzy control (FLC) methods have been usually utilized due to the complex and nonlinear behavior of the system. The membership functions of the FLC consist of the inputs such as the pollutant level inside the tunnel, the pollutant emitted from passing vehicles, and the output such as the number of running jet-fans. Conventional fuzzy control methods rely on simple experiences and trial and error methods. In this paper, the FLC was optimally redesigned using the genetic algorithm (GA), which is a stochastic global search method. In the process of constructing the objective function of GA, two objectives listed above were included: maintaining an adequate level of the pollutants and minimizing power consumption. The results of extensive simulations performed with real data collected from existing tunnel ventilation system are provided in this paper. It was demonstrated that with the developed controller, the pollutant level inside the tunnel was well maintained near the allowable limit and the energy efficiency was improved compared to conventional control schemes.