State Recognition of Multi-Nozzle Electrospinning Based on Image Processing.

Area of Science:

• Materials Science and Engineering
• Chemical Engineering
• Process Control

Background:

• Stable electrospinning is crucial for industrial applications, requiring precise control over multi-jet systems.
• Online monitoring is essential for detecting and correcting anomalies in real-time during electrospinning.
• Existing methods may lack the precision needed for large-scale, multi-nozzle electrospinning processes.

Purpose of the Study:

• To develop an automated image processing system for recognizing the ejection state of multi-jet electrospinning.
• To enable real-time monitoring and intelligent control of large-scale multi-nozzle electrospinning equipment.
• To improve the stability and reliability of the electrospinning process for industrial manufacturing.

Main Methods:

• Utilized CMOS industrial cameras for real-time recording of multi-nozzle electrospinning ejection behaviors.
• Applied image processing techniques including Roberts operator edge detection, Hough transform line detection, and mask histogram analysis.
• Developed algorithms to identify characteristic information of the multi-jet cone tip and detect anomalies like hanging droplets.

Main Results:

• Successfully constructed an ejection state recognition system for multi-jet electrospinning.
• Achieved real-time identification of jet anomalies, specifically hanging droplets at the nozzle outlet.
• Demonstrated an identification accuracy of over 85% for target hanging droplets.
• Laid the groundwork for intelligent control of large-scale multi-nozzle electrospinning equipment.

Conclusions:

• The developed vision system enables effective online monitoring of multi-jet electrospinning processes.
• Accurate identification of jet anomalies contributes to achieving stable mass electrospinning.
• This technology supports the intelligent control and industrial application of large-scale electrospinning systems.