Adaptive Critic Design Based Control of Tunnel Ventilation System with Variable Jet Speed

In this paper, an adaptive critic design (ACD) based control algorithm is proposed for road tunnel ventilation systems with jet fans. Firstly, a comprehensive dynamic model is established for control purpose, which considers both the static and dynamic nonlinearities between the pollutant concentration and the jet speed or the air flow inside the tunnel. Vehicle speed and density are also integrated into the model as system external disturbances. To cope with the time-varying model parameters and unknown system uncertainties, one artificial neural network (NN) unit is employed in controlling the jet fans to regulate the pollutant concentration to a desired level, even in the presence of varying working conditions. An additional NN is utilized to approximate the cost-to-go function required for the performance optimization. The proposed scheme is also verified and compared with traditional control design in simulation environment. The simulation results substantiate the conclusion that the adaptive controller can achieve a better regulation performance along with less energy consumption.