Additive Manufacturing Defect Detection With Weighted Sensor Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Defects during additive manufacturing processes, particularly in automated wire arc additive manufacturing of metals and metal alloys, can lead to component failure and increased costs due to the complexity of thermal history and the number of heating and cooling cycles involved.
Innovation Solution
A system comprising a robotic arm, sensors, and a processing unit that monitors the metal material during manufacturing, receives information about the microstructure, and determines the likelihood of defects by weighting sensor data based on microstructure, sensor characteristics, and geometry.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple sensors are used to monitor the additive manufacturing process, then defect detection accuracy is improved, but device complexity increases
Solution Approach 1:
The system divides defect detection into multiple specialized sensor types, each monitoring specific parameters (optical sensors for surface defects, thermal sensors for temperature anomalies, acoustic sensors for cracking). This segmentation allows each sensor to be optimized for its specific function while collectively achieving comprehensive defect detection.
Solution Approach 2:
The control system serves multiple functions: it processes data from various sensor types, performs defect analysis, controls the additive manufacturing process parameters, and coordinates repair operations. This multi-functionality consolidates what could be separate systems into a unified platform.
2Reliability
If real-time defect identification and repair is implemented, then component reliability is improved, but manufacturing time increases
Solution Approach 1:
The system performs preliminary defect identification during the additive manufacturing process itself, before the component is fully completed. By detecting and marking defects in real-time, the system prepares for repair without interrupting the overall manufacturing flow, allowing repair operations to be scheduled efficiently.
Solution Approach 2:
The system continuously monitors the additive manufacturing process and provides real-time feedback on defect formation. This feedback loop enables immediate adjustment of process parameters or initiation of repair operations, preventing defect propagation and reducing the need for extensive post-processing.
3Measurement precision
If weighted sensor data analysis is used to determine defect likelihood, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The system dynamically adjusts the weightings assigned to different sensor data based on the specific additive manufacturing process conditions, material being deposited, and defect types being monitored. This parameter adjustment allows the same hardware system to adapt to different manufacturing scenarios without requiring complex reconfiguration.
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
The system efficiently identifies and repairs defects in real-time, reducing the risk of component failure and costs associated with defect detection and correction, while maintaining near-net-shape production.
Implementation Method 1
a robotic arm configured for manufacturing a three-dimensional object by melting and solidifying a metal material via a heat source
Data Source
AI summary
Methods and systems for identifying a defect in an additive manufacturing process. One or more robots are configured for manufacturing, inspection, and repair of material. One or more sensors are configured for analyzing the material. A memory includes instructions and at least one processor configured to execute the instructions. The instructions are configured to cause the robot(s) and/or a corresponding computing platform to perform steps including receiving information from the one or more sensors during deposition of the material and obtaining microstructure information specific to the material being deposited from a predictive index. The steps also include assigning a weight to the output of, or to a defect probability determined from the output of, at least one of the one or more sensors. The steps also include determining, using the weighted information, the likelihood of a defect in the material.