CAREER: Irregular Modulation: Harnessing the Hidden Potential of PWM

Modulation, since the birth of modern power electronics, bridges control algorithms with switch operations. Despite its significance, research related to modulation has seen relatively little progress in the past decade, generally due to the widespread use of sequential computing devices (e.g. digital signal processors). However, recent breakthroughs in artificial intelligence (AI) technologies have propelled parallel computing devices into the forefront. To leverage the potential benefits of parallel computing devices and AI, this project proposes an irregular modulation theory, which can be easily applied to parallel computing devices and memristive devices. Additionally, this project will fully unlock the control capacity of modulation, enabling the control bandwidth to truly benefit from the high switching frequency brought in by Wide Bandgap (WBG) devices. This novel modulation approach will inevitably lead to the emergence of new topologies and control strategies, naturally giving rise to novel inverter modeling methodologies. Furthermore, the PI proposes a Mini-Task Inspired Learning Experience (M-TILE) approach to support students, particularly those from underrepresented groups, in overcoming imposter syndrome. To extend the impact of M-TILE, this concept will be shared with local K-12 teachers through a one-day workshop. By deeply integrating irregular modulation and the M-TILE concept, education activities will offer world-class education, inspiring and training future leaders in science, technology, engineering, and mathematics (STEM). The proposed irregular modulation theory will break free from the conventional idea of sawtooth and triangular carriers and lead to a new genre of converter modulation technologies. It will fundamentally change the operating principle of future power converters. Due to the fast response nature of modulation-control integration, it is possible to enable a new ultrafast control loop in addition to the classic double-loop control architecture. This new control loop will accommodate the recent advancement of AI technologies and lead to an entirely new genre of grid services. As a result of irregular modulation, a wide variety of new converter topologies, controls, and models will be revealed, holding the potential to eliminate vulnerable components and reduce voltage/current stress. Therefore, the irregular modulation theory proposed in this project will serve as the fundamental enabler of new topologies. This, in turn, will ultimately reduce device stress and component count, significantly lowering converter cost and enhancing reliability. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.