Quasi-Steady Roll Plane Analysis and Metaheuristic Optimization for Enhancing Lateral Stability of Electric Vehicles using Active Height Control

With the increasing adoption of Electric Vehicles (EVs), the need for ensuring safe driving arises obviously. Though the EVs are designed to have a lower Center of Gravity (CG) height compared to Internal Combustion Engine Vehicles (ICEVs), due to their dynamic operating and road conditions, probabilities of rollover risk prevail. In order to ensure lateral stability and mitigate rollover risk, this paper presents the application of a metaheuristic optimization method to get the optimal key design parameters, one of which is the CG height, which is used as a reference signal to the Model Predictive Controller (MPC) of an Active Height Control (AHC) system comprising a hydraulic active suspension system at each of the wheels. The key parameters that influence the quasi-steady lateral acceleration are first optimized using the Shuffled Frog Leaping Algorithm (SFLA), and the results are benchmarked against the Genetic Algorithm (GA) to prove the salient feature of SFLA. A constrained Model Predictive Control (MPC) system that could estimate the disturbance and track the CG height offset-free is developed to modulate the vehicle’s CG height to the SFLA reference. The SFLA, GA and MPC are simulated using MATLAB coding. Simulation results of SFLA and GA are tabulated. The AHC system simulation results demonstrate that the proposed approach effectively maintains the CG height near its optimal value while minimizing lateral acceleration. Furthermore, the rollover index is also significantly reduced, remaining well below critical safety thresholds. An experimental demonstration comprising a LVDT with a metric micrometer to emulate measurement of CG height in real-time, associated signal conditioning circuit to derive a usable voltage and a servo positioning system to emulate electro-hydraulic valve functioning in the Hydraulic active suspension system is also presented to prove the feasibility of the proposed methods real-time implementation and control.