Scholarlink

<p>Electrohydrodynamic direct-writing (EDW), also known as near-field electrospinning, is an advanced additive manufacturing technique that uses electric forces to produce polymeric functional fibers with micro- to nanometer-scale diameters. However, the process involves complex multiphysics phenomena, such as rapid solvent evaporation, making it challenging to accurately predict fiber dimensions and deposition patterns, often leading to reliance on trial-and-error methods. In this study, we developed a multilayer perceptron (MLP) model to predict jet profiles and optimize process parameters for EDW. Using a real-time monitoring system that captures and analyzes jet profile images, the MLP model was trained to predict key parameters with high accuracy, achieving root mean square errors of 0.025 for jet diameter and 0.078 for jet center position. The model’s rapid inference time of 0.043 s demonstrates its feasibility for integration into real-time systems. This approach represents a significant advancement in optimizing EDW processes for applications such as tissue engineering, printed and flexible electronics, and sensors.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Optimizing electrohydrodynamic direct-writing with multilayer perceptron: accurate and efficient predictions of jet profiles

  • 期刊论文

  • 发表日期：

    2025年1月11日

查看全文
  • Dongwoon Shin,
  • Jiyoung Chang

摘要

Electrohydrodynamic direct-writing (EDW), also known as near-field electrospinning, is an advanced additive manufacturing technique that uses electric forces to produce polymeric functional fibers with micro- to nanometer-scale diameters. However, the process involves complex multiphysics phenomena, such as rapid solvent evaporation, making it challenging to accurately predict fiber dimensions and deposition patterns, often leading to reliance on trial-and-error methods. In this study, we developed a multilayer perceptron (MLP) model to predict jet profiles and optimize process parameters for EDW. Using a real-time monitoring system that captures and analyzes jet profile images, the MLP model was trained to predict key parameters with high accuracy, achieving root mean square errors of 0.025 for jet diameter and 0.078 for jet center position. The model’s rapid inference time of 0.043 s demonstrates its feasibility for integration into real-time systems. This approach represents a significant advancement in optimizing EDW processes for applications such as tissue engineering, printed and flexible electronics, and sensors.