Inkjet Printer Base Material Expansion Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional inkjet printing apparatuses face challenges in accurately predicting and adjusting for the expansion/contraction of elongated strip-shaped base materials due to varying conditions such as tension, type of material, and ink, leading to distorted images and misregistration of ink positions.
Innovation Solution
A method and apparatus that utilize machine learning to estimate the expansion/contraction state of the base material by acquiring image data, generating an estimation model using learning data with marked regions, and adjusting ink ejection positions based on the estimated expansion/contraction state before printing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If conventional printing parameters are used, then printing operation is simple, but print quality deteriorates due to expansion/contraction of base material
Solution Approach 1:
The system performs preliminary estimation of the base material's expansion/contraction state using machine learning before printing occurs. By predicting the dimensional changes in advance based on submitted image data and material properties, the system can pre-calculate compensation values for ink ejection positions, thereby maintaining print quality without complicating the actual printing operation.
Solution Approach 2:
The system incorporates feedback mechanisms where the estimated expansion/contraction state is used to adjust ink ejection parameters. The machine learning model continuously refines its predictions based on actual printing conditions and measured dimensional changes, creating a closed-loop system that maintains precision while keeping operations simple.
2Manufacturing precision
If printing parameters are adjusted to compensate for expansion/contraction, then print quality improves, but device complexity increases
Solution Approach 1:
The patent replaces complex mechanical adjustment mechanisms with a computational approach using machine learning. Instead of physically adjusting printing parameters through multiple sensors and actuators in real-time, the system uses an estimation model that processes submitted image data and material information to predict expansion/contraction, then calculates compensation values algorithmically. This substitution reduces mechanical complexity while maintaining or improving precision.
Solution Approach 2:
The system changes the approach from real-time parameter adjustment to pre-calculated parameter compensation. By using machine learning to estimate expansion/contraction states and pre-computing correction values based on submitted data and material properties, the system simplifies the parameter adjustment process while maintaining high print quality.
3Measurement precision
If real-time measurement of expansion/contraction is performed, then ink position accuracy improves, but measurement time and processing complexity increase
Solution Approach 1:
The system performs expansion/contraction estimation in advance before printing occurs, using machine learning models that process submitted image data and material information. This preliminary estimation eliminates the need for real-time measurements during the printing process, reducing measurement time while maintaining ink position accuracy through pre-calculated compensation values.
Solution Approach 2:
The system creates a computational model (copy) of the base material's expansion/contraction behavior using machine learning. Instead of physically measuring the actual material dimensions in real-time, the system uses the learned model to predict dimensional changes based on submitted data, significantly reducing measurement time while maintaining accuracy.
4Measurement precision
If machine learning estimation model is used, then prediction accuracy of expansion/contraction improves, but data processing complexity increases
Solution Approach 1:
The system changes the data processing approach by using pre-trained machine learning models that take submitted image data and material properties as input and output expansion/contraction predictions. This parameter-based approach simplifies the processing complexity compared to real-time physical measurements, while maintaining high prediction accuracy through the learned models.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A printing apparatus includes a transport mechanism, heads, and an estimation part. The transport mechanism transports an elongated strip-shaped base material along a predetermined transport path in a longitudinal direction thereof. The heads eject ink onto a surface of the base material being transported by the transport mechanism, based on submitted data. The estimation part outputs an estimation result indicating the expansion/contraction state of the base material resulting from the ink, based on the submitted data, prior to printing of the submitted data. This provides the estimation result of the expansion/contraction state in accordance with the submitted data. Thus, the submitted data is printed in consideration of the estimation result. The printing apparatus according to the present invention is capable of estimating the expansion/contraction state of the base material in accordance with the submitted data prior to the printing.