Adaptive variable-exponent sliding mode control for energy management of battery-supported hybrid electric vehicle

This study presents a robust converter-level control framework for a fuel cell electric vehicle (FCEV) equipped with a multi-source hybrid power system (HPS) comprising a fuel cell (FC), an ultracapacitor (UC), and a lithium-ion battery. The considered powertrain employs a boost converter for the FC and bidirectional converters for the energy storage units and is represented by a fourth-order nonlinear model. To achieve accurate source-current tracking and stable DC-link voltage regulation under varying load demands and source-voltage uncertainties, an adaptive variable-exponent integral sliding mode control scheme is developed. The controller ensures fixed-time convergence of the closed-loop system, independent of the initial conditions, and compensates for lumped disturbances without requiring prior knowledge of their bounds. Lyapunov analysis establishes the practical fixed-time stability of the closed-loop states, which converge to a neighborhood of the origin within a fixed time. The proposed controller is evaluated through simulations and hardware-in-the-loop (HIL) experiments. Compared with the benchmark controller, it regulates the DC-link voltage at 400 V, reduces the settling time by approximately 64%, decreases the maximum overshoot by more than 90%, lowers the tracking error indices by up to 85%, and reduces the control effort by up to 16%. The HIL results further demonstrate consistent real-time performance under the tested current-demand and source-voltage variations. • Adaptive variable-exponent ISMC developed for FCEV energy management. • Practical fixed-time stability achieved without disturbance bounds. • Robust DC-link voltage regulation under source-voltage uncertainties. • Improved current sharing among fuel cell, battery, and ultracapacitor. • Reduced tracking errors and control effort under standard driving cycles.