A real time multi-objective optimization Guided-MPC strategy for power-split hybrid electric bus based on velocity prediction

Considering the frequent acceleration and deceleration of bus vehicles, the working conditions are complex, efficiency-oriented power-split hybrid electric bus (PSHEB) typically require frequent shifting to stay in high-efficiency areas, driving comfort and fuel economy may be affected. Therefore, to achieve a good balance be-tween overall efficiency and shifting stability, the study proposes a Real time Multi-objective optimization Guided-MPC strategy (RMGMPC) for PSHEB based on velocity prediction. Firstly, considering the different driving habits of drivers, combining with multi-source data fusion technology, a vehicle speed prediction controller is established; secondly, based on global optimization algorithm and multi-source data fusion tech-nology, a SOC reference generator is designed, which will determine the SOC guidance at predicted vehicle speed time domain online; then, to coordinate fuel efficiency, shifting stability and online optimization control real-time, the novel RMGMPC based on the direct multiple shooting method and sequential quadratic program-ming algorithm for PSHEB is proposed; finally, to avoid experience value of uncertain weight coefficient affecting the MPC, a weighted method of objective function with orientation is proposed. To verify the effectiveness of RMGMPC, the fuel economy reaches 98.41% of the global optimum; the shifting times are improved by 12.5%; Compared with MPC-DP, the calculation time is improved by 93.97%; And HIL test was carried out to further verify the real-time performance of the algorithm. The results manifest the excellent performance of the proposed RMGMPC.

Automated summary

This study proposes a Real time Multi-objective optimization Guided-MPC strategy (RMGMPC) for efficiency-oriented power-split hybrid electric buses (PSHEB). The strategy includes a vehicle speed prediction controller, a SOC reference generator, and the novel RMGMPC based on the direct multiple shooting method and sequential quadratic programming algorithm. The proposed RMGMPC achieves high fuel economy, improved shifting times, and reduced calculation time compared to MPC-DP. The results are verified through HIL testing.