ANFIS-Based Energy Management for Electric Heavy-Duty Vehicles With a Hybrid Energy Storage System Considering Payload Variations

Abstract One of the main contributors to global greenhouse gas emissions is the public transportation sector. The streamlined operation of heavy-duty vehicles in this area is crucial for improving energy efficiency. Focusing on heavy-duty battery electric vehicles (BEVs), this paper proposes to add a hybrid energy storage system (HESS) to support the batteries and extend their lifespan by adding supercapacitors and fuel cells. By implementing advanced energy management strategies (EMS) in the HESS, such as a fuzzy logic controller (FLC) and an adaptive neurofuzzy inference system (ANFIS), the power distribution between the energy sources can be optimized. This study highlights the adaptability of ANFIS in handling nonlinear relationships among power demand, battery state of charge (SOC), and vehicle mass while addressing its changes due to dynamic passenger loads. An analysis is conducted between the ANFIS-based EMS and FLC-based controller, as well as with the baseline BEV model without HESS. Results demonstrate that the HESS equipped with the ANFIS controller improves power distribution and minimizes battery stress during the reference driving cycle by reducing the root mean square (RMS) of battery C-rate. Specifically, ANFIS achieves reductions of approximately 3% and 7% when compared to the FLC, with vehicle masses of 14036 kg and 15072 kg, respectively, which correspond to different passenger loads. This reduction in the RMS of the battery C-rate, a critical factor influencing battery lifespan, will benefit heavy-duty vehicles in public transportation systems in the long term. These findings underline the optimization capability of the ANFIS controller to enhance the power delivery of electric heavy-duty vehicles with HESS under dynamic operating conditions.