Fast Quasi-Optimal Control of On-Board Energy Storage Systems for Regenerative Braking Energy Recovery Based on Area-Equivalence Principle

Existing threshold-based control strategies for on-board energy storage systems (OESSs) often fail to achieve optimal energy-saving performance, while AI-based strategies typically suffer from low computational efficiency. To address these limitations, this article proposes a fast quasi-optimal control strategy for OESSs by leveraging the electrical characteristics of supercapacitors and the area-equivalence principle. Specifically, during train braking, OESSs should absorb as much regenerative braking energy as possible. By applying the area-equivalence principle, this goal is reformulated as a geometric problem-maximizing the area bounded by the train power curve and the OESS power curve. Thus, the optimal charging current and initial charging voltage of OESSs can be determined. Correspondingly, a discharging strategy is developed to ensure that the OESS voltage at the end of train traction reaches the previously obtained optimal initial charging voltage. Case studies reveal that the proposed area-equivalent strategy can achieve 98% of the optimal energy-saving performance while maintaining low root-mean-square (rms) current and average di/dt. Meanwhile, its computational time is only 1/92 of that of dynamic programming (DP) method. With a well-balanced trade-off among energy recovery performance, computational efficiency, and OESS lifetime, the superiority of the area-equivalence-based control strategy is verified.