Circular knitting machine and respective method to control textile quality by use of digital camera
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for detecting defects in circular knitting machines are limited in their ability to automatically identify and prevent various types of fabric defects during production, particularly continuous Lycra, dashed Lycra, contamination, and non-uniformities, and require human intervention, leading to inefficient quality control and potential for defective rolls to be produced.
Innovation Solution
A detector system comprising a digital camera, data processor, and lighting setup integrated into the knitting machine, capable of capturing and analyzing 2D images of the fabric in real-time, using machine-learning methods to detect and differentiate between defects, and automatically stopping the machine or sending alerts to prevent further defective production.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If human workers manually check machines periodically to detect defects, then quality control can be performed, but defective rolls are already produced by the time defects are detected, leading to loss of production time and material
Solution Approach 1:
The patent implements real-time defect detection during the knitting process itself, rather than after production. The detection system continuously monitors the fabric as it is being knitted, enabling immediate identification of defects before they propagate through the entire roll, thus preventing waste of production time and material
Solution Approach 2:
The system provides immediate feedback to the control system when a defect is detected, triggering automatic machine stoppage. This closed-loop feedback mechanism ensures that defects are detected and addressed in real-time, eliminating the delay inherent in manual periodic inspection
2Reliability
If hundreds of sensors are installed on the knitting machine to detect broken yarns and needles, then machine failures can be prevented, but the system cannot detect fabric defects that occur during production
Solution Approach 1:
The patent employs a multi-functional detection system that combines the roles of traditional sensors (for detecting broken yarns and needles) with imaging technology (for detecting fabric defects). This universal system performs both machine monitoring and quality inspection functions, eliminating the need for separate specialized sensors for each type of detection
Solution Approach 2:
The system merges the functions of mechanical sensors and optical imaging into a unified detection platform. The imaging system captures fabric defects while traditional sensors continue to monitor machine components, creating an integrated monitoring solution that addresses both machine reliability and fabric quality
3Measurement precision
If commercial sensors are used to detect defects, then some defect types (holes, oil spots) can be detected, but most fabric defects (Lycra defects, contamination, non-uniformities) remain undetected
Solution Approach 1:
The system uses imaging technology that captures visual parameters of the fabric (color, texture, pattern) rather than relying on physical sensors that detect only specific properties like holes or oil spots. This parameter change enables detection of a broader range of defect types including Lycra defects, contamination, and non-uniformities that are invisible to traditional sensors
Solution Approach 2:
The patent replaces mechanical sensing systems with optical imaging and machine learning analysis. This substitution enables the system to detect defects based on visual patterns and characteristics rather than physical contact or specific sensor responses, dramatically expanding the types of defects that can be identified
4Loss of information
If workers manually inspect fabrics after production, then defects can be identified, but the inspection process is time-consuming and defective rolls have already been produced
Solution Approach 1:
The system replaces manual visual inspection with automated optical imaging and machine learning algorithms. This substitution provides immediate quality information during production rather than after, eliminating the time delay and enabling real-time quality control that maintains both information availability and production efficiency
Solution Approach 2:
The system performs self-inspection by automatically capturing images, analyzing defects through machine learning, and triggering machine stoppage without human intervention. This self-service capability eliminates the need for manual inspection while maintaining continuous production monitoring
Data Source
AI summary
A circular knitting machine, a method for controlling textile fabric defects, and a method for retrofitting a circular knitting machine are provided. The circular knitting machine includes a fixed support structure, a rotational support structure, and a system for controlling textile fabric defects. The system includes a digital camera for capturing digital images of knitted textile fabric, a data processor for processing the captured digital images, a camera support structure for holding the camera, and a lighting system to illuminate the knitted textile fabric from the camera side for capture by the digital camera. The camera support structure is fixed to the rotational structure.