Attentive ink MLP droplet detection algorithm based on the satellite droplets threshold domain.

Area of Science:

• Fluid Dynamics
• Materials Science
• Computational Modeling

Background:

• Satellite droplets in inkjet printing degrade print quality.
• Traditional methods for controlling satellite droplets are empirical and inefficient.
• Accurate prediction and control of droplet formation are crucial for advanced printing technologies.

Purpose of the Study:

• To establish a simulation model for predicting satellite droplet formation in piezoelectric inkjet nozzles.
• To identify the key parameters influencing satellite droplet generation.
• To develop an accurate and efficient algorithm for real-time prediction of satellite droplets.

Main Methods:

• Numerical analysis of a ring-shaped piezoelectric ceramic nozzle model.
• Simulation of 13,650 different voltage waveforms and collection of droplet behavior parameters.
• Development of an attention-based Multi-Layer Perceptron (MLP) algorithm (Attentive Ink MLP).
• Experimental validation using a droplet observation system.

Main Results:

• The relative differential of the negative pressure peak is identified as the primary cause of satellite droplet formation.
• A "Satellite droplet threshold domain" was established based on critical parameter values.
• The Attentive Ink MLP algorithm achieved a prediction accuracy of 0.988.
• Prediction time was significantly reduced from 40 minutes to 5 minutes.

Conclusions:

• The study provides a robust method for understanding and predicting satellite droplet formation.
• The developed Attentive Ink MLP algorithm offers a highly accurate and time-efficient solution for inkjet printing quality control.
• The findings are validated through both numerical simulations and experimental observations, confirming the practical applicability.