Tunnel fire smoke control based on the PID method: A numerical study

Abstract

A proportional-integral-differential (PID) ventilation controller is numerically investigated in the context of smoke control to protect the region upstream of the tunnel fire, while also considering the effect on the downstream region. This study proposes a method by coupling a full-field CFD model and the PID algorithm to control the fire-induced smoke automatically. In this method, the ventilation system automatically adjusts the ventilation velocity such that the smoke back-layering is controlled at the control point. The stability of the PID ventilation system is analyzed first, after which a set of values for the controller coefficients are determined for practical reference. The results manifest excellent stability, which guarantees the rationality and practicability of the system. Furthermore, this study investigates the influence of the control distance on the robustness of the PID ventilation system. For a tunnel with a 30 m upstream region and under the determined coefficients, there are three control patterns: the first is that the smoke crosses the control point only once before it is controlled; the second is that the smoke crosses the control point twice before it is controlled; the third is that the smoke crosses the control point just once before it is controlled, but the smoke can reach the entrance of the tunnel. The results show that the PID ventilation system performs in an excellent manner for various control distances. Finally, the comparison of the time-averaged quasi-steady state ventilation velocity as acquired by the PID ventilation system to experimental data as obtained for a constant ventilation based system shows good agreement, illustrating that the simulation results are reliable.