The energy waste problem of illumination system in tunnel is serious. In order to reduce the energy consumption of tunnel illumination control system, a kind of tunnel illumination control strategy based on fuzzy neural network algorithm is put forward. The realtime collection of brightness outside the tunnel portal, the traffic flow and the vehicle speed are taken as input values of the illumination control system; the illumination loop dimming values collected from every illumination tunnel section are taken as the output values of the illumination control system; a illumination control model with tripleinput and singleoutput is established; and the mechanism and training of the establishment and caululation of the model are analyzed. The energysaving effect of the model for tunnel illumination control system is verified by experimental simulation, which can provide reference for present study of tunnel illumination control.