ANN-Aided Virtual-Space-Vector PWM Strategy for Three-Level NPC Inverter Using Sextant-Coordinate Mapping for Heavy-Duty EV Applications

The three-level neutral-point-clamped (3L-NPC) inverter is a mature topology that tends to be utilized in the next-generation powertrain in heavy-duty electric vehicles (EVs). However, the wide operating range under off-road scenarios inevitably causes a high modulation index and low load angle, which poses neutral-point (NP) voltage of the 3L-NPC inverter at an unbalanced condition. To solve this disadvantage, the prior-art virtual-space-vector pulse-width-modulation (VSVPWM) has been explored owing to average-zero NP currents for all ranges of load conditions. But this method raises computational burden due to complicated subsector and dwell-time determination. In this paper, a new artificial neural network (ANN)-aided VSVPWM is thus proposed by using a sextant-coordinate system. The designed ANN attains excellent training performance with negligible errors. More importantly, all the trained nets are designed with simple structures for running efficiently on commercial digital signal processors (DSPs). This makes the proposed artificial intelligence (AI)-based PWM scheme possible to be executed in a commercial controller of the studied EV powertrain. Based on the training data collected by coordinate-based derivations and the trained nets, the validity of the proposed PWM was verified by simulation and experimentation.