Adaptive Neural Network Predictive Control Design for Hybrid Electric Vehicle with Hardware in the Loop (HIL) Verification

In this paper, the speed control of Hybrid Electric Vehicles (HEVs) in the environment of Hardware in the Loop (HIL) is implemented and designed based on different control methodologies. The concept of using HIL in this research is to check the proposed control design in real time environment. The Adaptive Artificial Neural Network based Model Predictive Control (MPC) is selected to be the core of control decision for the electric vehicles. In addition, The ANN is suggested to improve the behavior and results of MPCs based on the distinct history of ANN. To verify the efficient of the Adaptive ANN-MPC as a proposed control strategy for speed control of EVs, the results of ANN- MPC are compared with Proportional Integral (PI) controllers and MPC. Electric vehicles (EVs) is considered as one of the most important equipment which introducing an efficient and clean environment. Besides, the priority for utilizing EVs is that the transportation process is treated as one of the largest CO2 emitters. In this research paper, the technique of HIL is implemented to evaluate the efficient of introducing the Adaptive ANN- MPC to control the speed of EVs in the environment of hardware. The simulation and HIL results demonstrate the preference of applying the Adaptive ANN- MPC in the challenge of speed control of EVs. The recommendation of the proposed controller is introduced based on inserting system disturbance, parameters perturbation, and profile variations and observe the behavior of the Adaptive ANN- MPC. In this check, the proposed controller still gives distinct results. Finally, the Adaptive ANN- MPC is recommended to be applied on speed control of HEVs.