Intelligent Model of Urban Road Tunnel Ventilation System Based on Multi-Level Neural Network

The urban road tunnel longitudinal ventilation is a complex non-linear process with the feature of strong time-varying, always affected by a good many uncertain factors, it is difficult to establish a precise mathematical model by using the traditional linear control theory. The paper describes an intelligent model consists of two levels of neural network, the first level is a traffic flow neural network model to predict the traffic flow changes in road tunnel, the result of predicting will be used as an input of the second level neural network which is used to describe an intelligent model of urban road ventilation system, through the different states of predicted traffic flow, to establish an intelligent model of urban road tunnel ventilation system based on multi-level neural network. This model was used for ventilation intelligent control under the environment of Wu-chang Yue-ma-chang tunnel. Results of field test indicate that the method can highly improve ventilation conditions.