Direct voltage MTPA control of interior permanent magnet synchronous motor driven electric vehicles

This manuscript proposes an efficient, straightforward, direct voltage maximum torque per ampere (MTPA) control scheme for an interior permanent magnet synchronous motor (IPMSM) propelling an electric vehicle (EV). The main feature of the traction control scheme is that the MTPA is attained by directly varying the amplitude and angle of the voltage vector, eliminating the need for current control loops and associated regulators. Instead, a single-speed controller is adopted. Furthermore, an analytical formulation based on the motor voltage model is developed to extract the desired voltage’s magnitude and angle to run the motor within the MTPA operating points, disregarding numerical solutions, control law approximation, long-winded iterative calculations, or approximate representation of the IPMSM. Such a methodology significantly reduces control scheme complexity, enhances computational efficiency, and mitigates the delays associated with cascaded-based control systems. Additionally, it facilitates straightforward real-time implementation. The performance of the designed algorithm is experimentally validated using commonly adopted driving cycles, namely the Federal Test Procedure (US06) drive cycle and the New European Driving Cycle (NEDC). The validity test is performed using a 5 HP IPMSM. Based on the driving cycles employed, an intensive comparative evaluation against MTPA field-oriented control (FOC) is established. A quantitative assessment is conducted using the MTPA FOC as a benchmark to investigate energy consumption. This assessment reveals that the designed strategy achieved energy savings of 1.318% and 2.26% under US06 and NEDC, respectively, compared to the MTPA FOC. The proposed method’s speed-tracking accuracy and computational efficiency are also investigated and compared to the FOC and existing direct voltage approaches, demonstrating an average improvement of 14% in speed-tracking accuracy and 6.8% in computational efficiency.