Modelling and Control of a Hybrid Electric Vehicle Powertrain for Nigerian Urban Driving Cycles Using Model Predictive Control

The adoption of hybrid electric vehicles in Nigeria is constrained by the absence of HEV powertrain designs optimized for local driving conditions, which differ substantially from the standard European and American driving cycles used in most existing HEV energy management research. This study develops a dynamic model of a parallel hybrid electric vehicle powertrain and applies model predictive control for energy management strategy optimization under a Nigerian urban driving cycle developed from GPS speed data collected on the Lagos-Ibadan expressway and Kano urban road network. The powertrain model, encompassing an internal combustion engine, electric motor-generator, battery pack, and power split mechanism, is developed and validated in MATLAB-Simulink against component characterization data. The MPC controller optimizes torque split between the ICE and electric motor at each control timestep to minimize fuel consumption subject to battery state-of-charge constraints and drivability requirements. The MPC strategy is benchmarked against a rule-based control baseline and a dynamic programming reference solution over the developed Nigerian driving cycle. Results show the MPC strategy achieves fuel economy within 8.3 percent of the theoretically optimal dynamic programming solution, compared to the rule-based controller which performs 22.1 percent below optimal. The study contributes the first empirically derived Nigerian urban driving cycle dataset and demonstrates that MPC-based energy management significantly outperforms conventional rule-based strategies under the characteristically stop-and-go traffic patterns prevalent in Nigerian cities. Keywords: hybrid electric vehicle, model predictive control, energy management, driving cycle, powertrain modelling