Model predictive and fuzzy control of a road tunnel ventilation system

In this paper we describe a control method for longitudinal ventilatio of two main elements: (a) prediction of a number of jet-fans and (b) fuzzy control of pollutant levels. Based on measurements of traffic intensity and weather conditions and by knowing tunnel parameters, production of carbon monoxide, nitrite oxides and small particles (soot) is predicted. Estimated values of pollutants are then used for calculation of fresh air volume demand, i.e. required air flow is determined. One-dimensional force equation is used for estimation of a number of jet-fans that would produce a thrust force sufficient to provide calculated air flow. In the same time a fuzzy controller compares measured and admissible levels of pollutants and adjusts a predicted number of jet-fans in order to keep the pollutant levels within predefined boundaries. The proposed method is tested by simulation and obtained results are compared with a method currently used in the ventilation system of the tunnel Ucka. Finally, the field results, obtained by implementation of the proposed control method, are also presented at the end of the paper. (c) 2007 Elsevier Ltd. All rights reserved.