Machine-Learning Fluid Dispensing Control for Defect Correction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing fluid dispensing systems face challenges such as material accumulation at the nozzle, satellite defects on the substrate, and inconsistent performance across different environments due to varying operating conditions, leading to defects and inefficiencies.

Innovation Solution

Implementing machine learning tools, such as neural networks, to analyze images of dispensed fluid volumes and adjust operating parameters in a closed-loop control system for improved fluid dispensing precision and defect detection.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If conventional fluid dispensing systems are used, then basic dispensing function is achieved, but material accumulation occurs at the nozzle and satellite defects appear on the substrate

Engineering Contradiction:
Improvedispensing precisionVSAvoidnozzle accumulation and satellite defects
Core Design Contradiction:
Manufacturing precisionVSObject-generated harmful factors

Solution Approach 1:

The patent implements a closed-loop control system where images of dispensed fluid volumes are captured and analyzed by machine learning tools to determine classifications (e.g., defect detection). This feedback loop allows the system to adjust operating parameters based on actual dispensing results, thereby reducing nozzle accumulation and satellite defects while maintaining high dispensing precision.

Inventive Principle:
Principle #23Feedback

2Adaptability or versatility

If fixed dispensing parameters are used, then simple operation is maintained, but performance becomes inconsistent across different environments

Engineering Contradiction:
Improveenvironmental adaptabilityVSAvoidcontrol system complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent transitions from fixed dispensing parameters to dynamic, adaptive parameters. Machine learning tools continuously analyze dispensing results and automatically adjust operating parameters in real-time, allowing the system to adapt to varying environmental conditions without manual intervention. This dynamic adjustment mechanism enhances environmental adaptability while the automation reduces the operational complexity burden.

Inventive Principle:
Principle #15Dynamics

3Adaptability or versatility

If manual adjustment of dispensing parameters is performed, then adaptability to different environments is achieved, but productivity decreases due to time-consuming adjustments

Engineering Contradiction:
Improveenvironmental adaptabilityVSAvoiddispensing throughput
Core Design Contradiction:
Adaptability or versatilityVSProductivity

Solution Approach 1:

The patent implements self-service automation where machine learning tools automatically analyze dispensing images and adjust operating parameters without human intervention. The system performs its own calibration and adaptation to different environments, eliminating the need for manual parameter adjustment. This self-service capability maintains high environmental adaptability while significantly improving productivity by removing time-consuming manual operations from the dispensing workflow.

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP4107479B1Improved fluid dispensing process control using machine learning and system implementing the same
Publication Date: 2026.04.01 NORDSON CORP
  • EP4107479B1 patent drawingFigure 1A
  • EP4107479B1 patent drawingFigure 1B
  • EP4107479B1 patent drawingFigure 1C

AI summary

Systems and methods for improved fluid dispensing process control using a machine learning tool are disclosed. In an example method, successive portions of viscous fluid are dispensed by a dispensing device according to operating parameters to train a machine learning tool to associate defect classifications with images of dispensed portions and/or operating parameters associated with dispensing the dispensed portions. The trained machine learning tool is then used in a closed loop fashion in production to detect and correct for defects associated with the dispensed portions to improve quality and production efficiency.