Nonlinear Waveform Optimization for Enhanced Ink Droplet Formation in Material Jetting.

Abstract

Material jetting, as a critical additive manufacturing technology, relies on precise control of the driving waveform to achieve high-quality droplet formation. During the droplet ejection process, pressure fluctuation at the nozzle outlet plays a significant role in droplet formation. Traditional experimental methods for optimizing the driving waveform often struggle to address the complex nonlinearities inherent in the jetting process. In this study, a numerical simulation model of the droplet ejection process is established to elucidate the influence mechanism of nozzle outlet pressure oscillations on droplet formation. A novel optimization method combining Convolutional Neural Networks (CNNs) and Particle Swarm Optimization (PSO) is proposed, targeting the suppression of residual pressure oscillations and achieving the desired pressure fluctuation. The method leverages nonlinear regression and optimization to obtain the optimal waveform design. Simulation and experimental results demonstrate that the optimized waveform effectively suppresses residual pressure oscillations, significantly improves droplet formation quality, and reduces pressure fluctuation convergence time by approximately 32.19%. The findings demonstrate that the optimized waveform effectively improves droplet ejection quality and stability.