An enhanced hypotrochoid spiral optimization algorithm based intertwined optimal sizing and control strategy of a hybrid electric air-ground vehicle

Hybrid electric air-ground vehicles (HEAGVs) are deemed as promising transportations due to their great versatility, mobility, and environmental values. Capable of ground-driving, vertical-take-off-and-land, and near-ground flight, HEAGVs are competent to pass poor road conditions such as broken bridges or cliffs. As a crucial part of the HEAGV development, size of the power sources is hard to design owing to different characteristics of air-driving and ground-driving. Large power sources would result in increased fuel consumption and accelerating battery degradation, while small ones might lead to insufficient power supply. Thus, optimal sizing of the power sources is needed at the functional level of a HEAGV design. Moreover, at the performance level, the control strategy should be taken into account simultaneously, which significantly affects the sizing process. However, the intertwined optimal sizing and control strategy of HEAGV becomes difficult to address due to the expanded design space. Motivated by this, an efficient co-optimization strategy is presented for the studied HEAGV, intended to simultaneously find optimal sizing of the battery pack and turbine-generator pack and optimal logic threshold control parameters. The initial mass, fuel consumption, and battery degradation are chosen as optimization objectives to formulate an objective function. Then, a novel enhanced hypotrochoid spiral optimization algorithm (EHSOA) is proposed to address the intricated co-optimization problem. In this algorithm, an enhanced bi-considering mechanism is firstly proposed to avoid the optimization process being trapped in local optima. The co-optimization is implemented under an air-ground driving cycle. Results show that, compared to the initial design, the proposed strategy reduces initial mass, fuel consumption, and battery degradation by 5.08%, 26.10%, and 2.08%, respectively. Finally, the proposed EHSOA is demonstrated to be more qualified to solve the intricated co-optimization problem in comparison to other optimization algorithms. The proposed co-optimization strategy might provide theoretical insights for future HEAGV powertrain designs.