A novel EMS for FCV combined with condition-adaptive search ranges of decision variables based on the spatial and temporal traffic characteristics

Condition-adaptive energy management strategies (EMSs) have been widely used in fuel cell vehicles (FCVs). There are two main issues that need to be further addressed: classification or recognition methods for driving conditions and adaptive strategies under these conditions. To address these issues, a novel MPC–based EMS for FCV combined with condition-adaptive search ranges based on traffic characteristics is proposed. The real urban road segments are further classified into several conditions based on the proposed spatial-temporal traffic characteristics, then the fuel cell system power intervals under these conditions are obtained using an efficiency-probability principle. A novel optimization algorithm, artificial protozoa optimizer is used to solve the short-term optimization problem under the novel MPC framework, which searches the optimal decision variables within the condition-adaptive fuel cell system power intervals. The results show that the proposed EMS combined with condition-adaptive fuel cell system power intervals and artificial protozoa optimizer–based solver reduces the total cost by 4.49 % and 4.30 % under real urban cycles compared to the conventional MPC–based EMS. In addition, the novel MPC framework with condition–adaptive fuel cell system power intervals reduce the computational burden of the short–term optimization problem, and the artificial protozoa optimizer–solver reduces the FCS power fluctuations compared to the solvers based on other algorithms.