Wire Arc Cladding With AI Vision for In-Process Defect Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing additive manufacturing methods using arc-wire deposition welding suffer from process errors such as cracks, gaps, and oxidation, which are difficult to detect and correct, leading to reduced automation and increased personnel costs due to visual inspection and subjective worker assessment.
Innovation Solution
Implementing multiple capturing devices to monitor different surfaces of the 3D object during production, using AI and computer vision to determine a building-quality parameter, allowing for automated detection and correction of defects, and adjusting process parameters to maintain quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If visual inspection by a worker is used to monitor process errors, then detection of defects such as cracks, gaps, and oxidation is possible, but the process must be interrupted and personnel costs increase
Solution Approach 1:
The patent replaces the mechanical/visual inspection system with an optical monitoring system using capturing devices (cameras) that continuously record the additive manufacturing process without requiring process interruption. The system captures images of the build platform and object from multiple angles, enabling continuous defect detection while maintaining process continuity.
Solution Approach 2:
The patent creates optical copies (images) of the manufactured object during the building process using capturing devices. These images are then analyzed to detect defects such as cracks, gaps, and oxidation, allowing quality monitoring without physical contact or process interruption.
2Reliability
If visual inspection by a worker is used to assess component quality, then subjective evaluation is possible, but automation is reduced and personnel costs increase
Solution Approach 1:
The patent replaces the human visual inspection system with an automated image processing system. Capturing devices record the manufacturing process, and evaluation algorithms automatically analyze the images to detect defects and assess quality, eliminating subjective human assessment and increasing automation.
Solution Approach 2:
The system enables self-inspection of the manufactured object through automated image capture and analysis. The evaluation algorithm independently assesses quality parameters such as crack detection, gap measurement, and oxidation identification without human intervention.
3Device complexity
If a single capturing device is used to monitor the additive manufacturing process, then device complexity is low, but multiple surfaces of the object cannot be captured simultaneously
Solution Approach 1:
The patent divides the monitoring task among multiple capturing devices positioned at different locations and angles. Each device is responsible for capturing specific surfaces or regions of the object, ensuring complete coverage of all build surfaces while maintaining manageable individual device complexity.
Solution Approach 2:
The patent adds spatial dimensions to the monitoring system by positioning capturing devices at multiple locations around the build platform. This multi-dimensional arrangement enables simultaneous capture of multiple surfaces that would be inaccessible to a single device, comprehensive surface coverage.
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
Enables continuous, automated monitoring and correction of defects, improving component quality and reducing personnel costs by eliminating the need for visual inspection and enhancing process automation.
Implementation Method 1
a metallic wire is melted on by way of an electric arc
Data Source
AI summary
A process for additively manufacturing three-dimensional objects using arc-wire deposition welding, in particular by layer-by-layer melting and solidifying of a building material that is in the form of a wire in its original state. During the manufacturing process, at least two different surfaces of at least one partially manufactured three-dimensional object are captured by at least two different capturing devices, and at least one building quality parameter describing a building quality of the object is determined.