Weld Bead Concavity Analysis for In-Process Fault Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In arc welding, faults such as cavities can occur within multilayer manufacturing objects due to the shape or arrangement of weld beads, leading to unwelded portions, which are difficult to detect and evaluate, especially in complex structures, and require time-consuming inspections or disassembly for identification and repair.
Innovation Solution
A fault-monitoring device and method that predicts welding faults by acquiring shape profiles of existing weld beads, extracting feature amounts, identifying potential fault locations, and updating these profiles during the welding process, combined with a welding assistance system that generates instruction information to prevent faults through post-processing or adjusting welding conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If non-destructive inspection is performed after welding to identify faults, then fault detection is achieved, but time and effort are consumed and productivity is reduced
Solution Approach 1:
The patent applies preliminary action by performing real-time monitoring and prediction of welding faults during the welding process itself, rather than conducting inspection after welding is complete. The system continuously acquires welding data, predicts potential faults, and alerts operators before the welding is finished, thereby eliminating the need for time-consuming post-welding non-destructive inspection while maintaining high detection accuracy.
2Reliability
If the welded structure is cut to confirm faults, then fault identification is achieved, but time and effort are consumed and structure recovery is difficult
Solution Approach 1:
The system performs preliminary fault prediction during the welding process, identifying potential faults before they become actual defects. By predicting faults in advance based on real-time welding parameter analysis, the system eliminates the need to cut into the welded structure for confirmation, thereby preserving the integrity of the workpiece and avoiding difficult recovery operations.
Solution Approach 2:
The patent replaces the mechanical action of cutting the welded structure with a non-contact prediction system that uses sensors and data analysis to identify faults. This substitution of mechanical inspection with electronic prediction maintains high accuracy while completely avoiding damage to the workpiece.
3Productivity
If conventional welding continues without fault prediction, then productivity is maintained, but quality issues arise due to undetected unwelded portions
Solution Approach 1:
The patent implements feedback by continuously monitoring welding parameters in real-time, comparing them against predetermined criteria, and providing immediate alerts when potential faults are detected. This feedback mechanism allows the welding process to continue at high productivity while maintaining quality, as operators can take corrective action immediately upon receiving the alert, preventing defective welds from being produced.
Data Source
AI summary
A fault-monitoring device includes: a shape profile acquisition unit that acquires a shape profile of the existing weld bead; a feature amount extraction unit that extracts a feature amount of a concave shape formed by the plurality of existing weld beads included in the shape profile; a fault position identification unit that identifies a fault candidate location where the welding fault is expected to occur according to the extracted feature amount; and a control unit that causes the shape profile acquisition unit to update the shape profile when the welding device newly forms the weld bead and repeatedly executes the extraction of the feature amount by the feature amount extraction unit and the identification of the fault candidate location by the fault position identification unit.