Automated waveform optimization for electroaerodynamic actuators using Genetic Programming Algorithms

This study utilizes Genetic Programming to optimize voltage waveforms for electroaerodynamic actuators, maximizing electromechanical power conversion efficiency. We evaluate mechanical power via net kinetic energy fluxes and dissipation within a control volume, and address accurate electrical power estimation for non-sinusoidal waveforms. We developed an automated experimental chain to generate the more than 1000 waveforms needed for convergence. Asymmetric waveforms, with different positive and negative cycle durations and high form factors, outperform sinusoidal waveforms by 80%. The optimized waveforms reduce power consumption, minimize power transfer to fluctuating fields, and reduce the amount of internal dissipation relative to net kinetic power.