Akademik Veri Yönetim Sistemi
ENERGY, cilt.338, sa.2025, ss.138854-138869, 2025 (SCI-Expanded, Scopus)
Amid global efforts toward decarbonization and sustainable urban mobility, maximizing energy recovery in electric public transportation has become a strategic priority. This study presents a data-driven approach to optimizing regenerative braking (RB) performance in hybrid electric trolleybuses operating in Türkiye, utilizing over 79 million real-world operational data points collected across five years. Departing from traditional simulation-based methodologies, this work develops a high-fidelity predictive model using a novel meta-heuristic algorithm inspired by the physiological mechanism of water uptake and transport in plants termed the Water Uptake and Transport in Plants-Based Electric Bus Regenerative Braking Model (WUTP-EBREM). By incorporating eight critical variables including vehicle speed, road gradient, acceleration, ambient temperature, passenger load, and auxiliary system consumption the model achieves exceptional predictive accuracy (RMSE: 0.12%).WUTP-EBREM was subsequently applied to 50 high-density urban bus routes with varying topographical and operational conditions. The analysis revealed significant variability in RB potential, indicating that energy recovery is influenced not only by vehicle design but also by route-specific environmental and operational factors. The model supports optimization of battery sizing, fleet energy planning, and route selection, offering actionable insights for transit authorities and electric vehicle manufacturers.Moreover, the model’s scalable and generalizable architecture enables adaptation to diverse climatic and geographic contexts, bridging the gap between theoretical energy models and field-level implementation. This research contributes a novel decision-support tool for intelligent transport systems, emphasizing the role of artificial intelligence in advancing real-time, energy-efficient public transit strategies across urbanized environments.