Modeling Regenerative Braking Efficiency in Urban Electric Vehicles as a Function of Initial Braking Speed and Average Deceleration

This research investigates how two fundamental motion parameters—initial braking speed and average deceleration—influence the regenerative braking (RB) efficiency of electric vehicles in urban conditions. Regenerative braking systems recover kinetic energy normally lost during braking and convert it into electrical energy, thereby extending driving range and enhancing sustainability. However, efficiency varies widely depending on how braking occurs. Rather than examining external factors like temperature or terrain, the study isolates these two kinematic variables to understand their direct, measurable impact on energy recovery.

Using a secondary analysis of previously published data (Szumska & Jurecki, 2022), the study reconstructs thirty braking events from 3D graphical data through a combination of axis-parallel projection and pixel interpolation. Each event is characterized by its speed, deceleration, and recovered energy, from which recovery efficiency is calculated as the ratio of recovered to total kinetic energy lost. After filtering outliers using the interquartile range rule, regression modeling is applied to the remaining 24 observations to statistically test the relationship between the two predictors and efficiency outcomes.

The findings reveal a clear asymmetry in influence: average deceleration emerges as a strong positive determinant of recovery efficiency—each 1 m/s² increase improves efficiency by roughly 16.8 percentage points—while initial speed shows a minor, statistically insignificant effect. This suggests that efficiency depends less on how fast a vehicle is moving and more on how it slows down. Applying the model to the Modified Indian Driving Cycle (MIDC) demonstrates that even at constant speeds, efficiency rises with stronger yet controlled deceleration. These results carry implications for EV design, traffic engineering, and urban transport policy, emphasizing how braking patterns and infrastructure design can be optimized to enhance fleet-level energy recovery and sustainability. The paper’s methodological innovation—reconstructing usable datasets from published visuals—also offers a replicable framework for future secondary modeling in automotive efficiency research.