Segmented dynamic control method for highway tunnel lighting: A sparrow search algorithm-optimized fuzzy neural network

To address the issue of excessive energy consumption in traditional lighting control systems, this study proposes a segmented dynamic tunnel lighting (SDTL) control system based on a fuzzy neural network (FNN) optimized by the sparrow search algorithm (SSA). When sensors detect a vehicle entering the tunnel, the SSA-FNN model dynamically predicts the required luminance for six tunnel segments including threshold zone one (TH1), threshold zone two (TH2), transition zone (TR), interior zone (IN), exit zone one (EX1) and exit zone two (EX2), by integrating real-time data on tunnel exterior luminance, vehicle speed and traffic volume. If no vehicles are present in the tunnel or within a specific tunnel segment, the lighting control system adjusts the luminaires in that segment to their base illumination power. Experimental results demonstrate that the proposed SDTL system achieves an overall energy-saving rate of 71.5%, outperforming the traditional dynamic tunnel lighting system at 58.0% and the manual segmented tunnel lighting system at 47.9%. Key findings include: (1) lighting demand in the TH1 and TH2 segments is predominantly influenced by exterior luminance in a tunnel, whereas in the segment, it is more sensitive to traffic volume and vehicle speed; and (2) due to the higher lighting requirements caused by the ‘black hole effect’, the energy-saving performance in the TH1, TH2 and TR segments is relatively lower compared to the IN, EX1 and EX2 segments. These insights provide valuable guidance for optimizing tunnel lighting strategies through zone-specific adaptive control, contributing to the development of sustainable transportation infrastructure.