AI Print Head Control for Real-Time Additive Manufacturing Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current additive manufacturing systems lack precise feedback and corrective action during the printing process, leading to inconsistencies in printed objects and difficulty in achieving desired mechanical, optical, and electrical properties.
Innovation Solution
Implementing artificial intelligence process control (AIPC) using image sensors, classifiers, and machine learning algorithms to analyze printed layers, identify anomalies, and adjust print parameters in real-time to correct errors and improve print quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional feedback methods (shadows) are used, then the system structure is simple, but measurement precision is poor due to obstructions and inability to provide precise feedback
Solution Approach 1:
The patent replaces traditional mechanical shadow-based feedback with an optical imaging system using image sensors to capture printed layers. This substitution enables precise measurement of printed object characteristics without mechanical obstructions, directly resolving the contradiction between measurement precision and system complexity.
Solution Approach 2:
The patent implements a feedback mechanism where captured images are processed to provide real-time information about printed layer quality. This feedback loop enables continuous monitoring and adjustment, improving measurement precision while managing system complexity through intelligent processing rather than additional hardware.
2Loss of time
If feedback is provided only after entire object printing, then the system is simple to implement, but loss of time increases due to delayed detection of printing errors
Solution Approach 1:
The patent applies preliminary action by capturing and analyzing printed layers during the printing process itself, rather than waiting for completion. Image sensors continuously monitor printed layers, enabling early detection of errors while they are still being formed, thus reducing time loss without requiring overly complex post-processing systems.
Solution Approach 2:
The patent maintains continuity of useful action through ongoing image capture and analysis throughout the printing process. This continuous monitoring enables real-time error detection while keeping the system manageable by integrating feedback into the existing printing workflow rather than adding separate post-print inspection stages.
3Manufacturing precision
If comprehensive real-time analysis of all printed layers is performed, then manufacturing precision is improved, but device complexity increases due to multiple classifiers and processing steps
Solution Approach 1:
The patent applies segmentation by dividing the analysis process into multiple specialized classifiers: a failure classifier to identify non-recoverable errors, a binary error classifier to detect correctable issues, and an extrusion quality classifier to assess material deposition. This segmentation enables comprehensive quality control while managing complexity by assigning specific functions to dedicated processing modules.
Solution Approach 2:
The patent applies partial action by selectively analyzing different aspects of printed layers through specialized classifiers rather than attempting uniform analysis of all layers. The failure classifier first filters for critical errors, then only problematic areas undergo more detailed analysis by subsequent classifiers, reducing overall processing complexity while maintaining high manufacturing precision.
Data Source
AI summary
Systems, methods, and media for additive manufacturing are provided. In some embodiments, an additive manufacturing system comprises: a hardware processor that is configured to: receive a captured image; apply a trained failure classifier to a low-resolution version of the captured image; determine that a non-recoverable failure is not present in the printed layer of the object; generate a cropped version of the low-resolution version of the captured image; apply a trained binary error classifier to the cropped version of the low-resolution version of the captured image; determine that an error is present in the printed layer of the object; apply a trained extrusion classifier to the captured image, wherein the trained extrusion classifier generates an extrusion quality score; and adjust a value of a parameter of the print head based on the extrusion quality score to print a subsequent layer of the printed object.