Assessment of Adaptive Self-Learning-Based BLDC Motor Energy Management Controller in Electric Vehicles Under Real-World Driving Conditions for Performance Characteristics

The superior performance of an electric vehicles (EVs) is dependent on the related energy management controller (EMC). The current study devoted to develop various EMCs such PID, intelligent, hybrid and supervisory strategy to enhance the performance of EVs under real-time driving conditions. Also, the work integrates the various novel methodologies to develop EV model, efficiency maps and real-time driving cycle (DC). In this instance, a mathematical model of an EV with BLDC motor is developed using MATLAB/Simulink. Further, the efficiency maps for the motor and controller with different EMC's are generated using the innovative experimental approach. Then, the developed efficiency maps are incorporated into model-in-loop (MIL)-based EV test platform to analyze the performance of various EMCs. Additionally, to validate the EV model, a real time DC has been developed for different types of road conditions, including urban, rural, and highway. Subsequently, the developed DC is integrated with MIL-based EV test platform for analysis of energy consumption and battery discharge behavior under real-time conditions. From the results, the proposed supervisory controller (68.4%) exhibits minimal SOC drop than the PID (21.5%), intelligent (44.9%) and hybrid (59.1%) controllers. As well, the energy consumption (EC) of the various EMCs is 85.63, 60.14, 44.67 and 33.4 Wh/km. In the case of regenerative efficiency of the developed EMCs under real-time driving conditions are −27.73, −41.64, −58.28 and −77.6 Wh respectively. The overall outcome of this work demonstrates that the proposed supervisory controller achieves a considerable improvement in battery consumption as well as a reduction in EC as compared to PID, intelligent, and hybrid controllers.