A particle swarm optimization algorithm combined with penalty function is proposed to reduce energy consumption of tunnel lighting so as to ensure safe driving. The proposed technique considers the total power of tunnel lighting as a fitness function and the spacing, height, and elevation of lights as independent variables. The optimization model of highway tunnel lighting parameters is based on the constraints of average road brightness, total uniformity of road brightness, and longitudinal uniformity of road median brightness. The optimization model is simulated and solved using the field data of the Yanglushan tunnel. After optimization, the average brightness of each lighting section fulfills the required brightness, and the energy saving of the whole tunnel achieves 10.53% compared with that before optimization. The Dialux is used to establish simulation scenarios and validate the optimization results. The results show that the error between the lighting parameters and the calculated values of the optimization model is within 10%, which indicates good applicability of the optimization model.