Welding Defect Prediction Using Surface Valley Geometry
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing non-contact inspection methods for defect detection in additively manufactured objects, such as ultrasonic flaw detection and X-ray CT, face limitations in applicability and cost, making it difficult to accurately predict and prevent defects in welded structures.
Innovation Solution
A learning device that generates an estimation model to predict defect sizes by learning the relation between welding conditions, dimensions of narrow portions, and positional relations in the surface shape of additively manufactured objects, using sensors to measure bead formation and generate a corrected welding plan to minimize defects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If ultrasonic flaw detection is used to inspect the quality of manufactured objects, then defect detection capability is improved, but it becomes difficult to apply the probe to complex manufactured objects and surfaces
Solution Approach 1:
The patent replaces the mechanical contact-based ultrasonic probe inspection system with a non-contact optical measurement system. The laser displacement sensor measures surface shape data without physical contact, enabling inspection of complex manufactured objects that cannot accommodate probe contact, while still achieving defect detection capability through surface shape analysis.
2Measurement precision
If X-ray CT apparatus is used for non-contact inspection, then defect detection capability is improved, but the apparatus itself is expensive and object size is limited
Solution Approach 1:
The patent employs a cost-effective laser displacement sensor instead of expensive X-ray CT apparatus. The laser sensor provides sufficient measurement precision for surface shape analysis and defect detection through computational methods, eliminating the need for costly imaging equipment while removing size limitations on inspectable objects.
Solution Approach 2:
The patent extracts only the essential measurement function (surface shape data acquisition) from complex imaging systems like X-ray CT. By using a simple laser displacement sensor to capture surface topology and combining it with computational analysis of welding conditions and surface shape relationships, the system achieves defect detection capability without requiring expensive volumetric imaging apparatus.
3Reliability
If visual sensor is used to capture welding state images, then welding quality monitoring is improved, but it cannot accurately predict defect sizes inside the welded structure
Solution Approach 1:
The patent introduces surface shape data as an intermediary measurement that bridges visual monitoring and internal defect detection. The laser displacement sensor captures surface topology, which serves as a measurable indicator correlated with internal defect formation. By analyzing the relationship between surface shape characteristics and welding conditions, the system can predict internal defect sizes without directly imaging the interior structure.
Solution Approach 2:
The patent performs preliminary measurement of surface shape data during or immediately after the welding process, before defects fully manifest or require destructive testing. The system uses the measured surface shape information in combination with welding condition data to predict defect sizes in advance, enabling proactive quality control rather than post-processing inspection.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A learning device includes: a data acquisition unit configured to acquire information on a welding condition when beads are deposited, a dimension related to a narrow portion forming a valley portion in a surface shape of an additively manufactured object before the beads are deposited, a positional relation between the narrow portion and a target position of the bead, and a defect size of an unwelded defect; and a learning unit configured to generate the estimation model by learning a relation between the welding condition, the dimension related to the narrow portion and the positional relation, and the defect size, and the dimension related to the narrow portion includes at least one of a bottom width of the valley portion, an opening width representing an interval between top portions on both sides of the valley portion, both sides constituting the valley portion, and a valley depth from the top portion to a bottom of the valley portion.