Additive Manufacturing Defect Detection Using Thermal ML Signals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Additive manufacturing systems face challenges in detecting and characterizing defects, such as lack of fusion, porosity, and inclusions, which can lead to premature failure of finished parts due to the difficulty in tracking and detecting these defects during the manufacturing process.
Innovation Solution
The implementation of a machine learning algorithm that uses sensors to collect data on thermal emission density and spectral peaks to identify defects in real-time, allowing for the detection and classification of defects like lack of fusion, porosity, and inclusions during the additive manufacturing process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional defect detection methods are used in additive manufacturing, then the manufacturing process is simple, but defects cannot be detected or characterized in real-time
Solution Approach 1:
The machine learning model is trained in advance with labeled defect data from training builds before production manufacturing. This preliminary training enables the model to automatically detect and classify defects during real-time production without adding complex detection hardware during the manufacturing process.
Solution Approach 2:
The patent replaces traditional mechanical or manual defect detection methods with a machine learning-based automated detection system. The model processes sensor data from the additive manufacturing process to identify defects, substituting physical inspection methods with computational analysis.
2Productivity
If real-time defect detection is implemented, then manufacturing efficiency improves, but the complexity of the system increases
Solution Approach 1:
The machine learning model autonomously detects and classifies defects without requiring external intervention or complex additional hardware. The system uses existing sensor data from the additive manufacturing process, enabling self-service defect detection that improves productivity without proportionally increasing system complexity.
3Loss of substance
If defects are not detected during manufacturing, then the manufacturing process is fast, but defective parts lead to waste and rework
Solution Approach 1:
The machine learning model provides real-time feedback during the additive manufacturing process by analyzing sensor data and identifying defects as they occur. This immediate feedback enables operators to stop the build process when defects are detected, preventing material waste from continuing production of defective parts while minimizing detection time through automated analysis.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enables real-time defect detection and classification, allowing operators to make informed decisions to either repair or stop the build process, significantly improving manufacturing efficiency and reducing waste.
Implementation Method 1
A sensor is arranged to detect electromagnetic energy emitted during the fusing of the metallic powder
Implementation Method 2
A power source is arranged to emit a beam of energy at the build plane and fuse the metallic powder
Implementation Method 3
The various processes that are used for making metallic parts have in common the sintering and/or melting of powdered or granular raw material
Implementation Method 4
determining a thermal emission density (TED) that includes measuring an amount of energy radiated from the build plane during one or more scans
Data Source
AI summary
An additive manufacturing system comprises an apparatus arranged to distribute layer of metallic powder across a build plane and a power source arranged to emit a beam of energy at the build plane and fuse the metallic powder into a portion of a part. The system includes a processor configured to steer the beam of energy across the build plane and receive data generated by one or more sensors that detect electromagnetic energy emitted from the build plane when the beam of energy fuses the metallic powder. The received data is converted into one or more parameters that indicate one or more conditions at the build plane while the beam of energy fuses the metallic powder. The one or more parameters are used as input into a machine learning algorithm to detect one or more defects in the fused metallic powder.