In Vehicular Ad Hoc Networks (VANETs), ensuring reliable, energy-efficient communication under varying traffic densities remains a major challenge, particularly when relying completely on conventional Radio Frequency (RF) communication. This study presents an integrated communication framework that combines Visible Light Communication (VLC) with RF to improve vehicular connectivity. Motivated by the limitations of fixed-power techniques in dynamic environments, we introduce two power allocation algorithms based on Density-Based Spatial Clustering of Applications with Noise (DBSCAN): one using Fixed Power Allocation (DFPA) and another employing Dynamic Power Allocation (DDPA). Both approaches are evaluated using realistic traffic simulations based on Simulation of Urban Mobility (SUMO) and OpenStreetMap (OSM) data, with a focus on energy efficiency and communication coverage. Simulation results show that DDPA outperforms DFPA in cases with cluster radii below 1159 m, reducing power consumption by up to 20% and serving 90% of vehicles. Conversely, DFPA proves more efficient at larger radii, saving up to 12% power. These results establish a scalable, integrated VLC/RF framework for next-generation VANETs, adaptable to diverse density and coverage requirements.