Weld Event Tracking With Machine Learning for Missed Weld Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional part tracking systems cannot identify which specific welds are missed during the welding process, and may fail to detect issues if an extra weld is accidentally added, as they only compare the number of detected welds to the expected number.
Innovation Solution
The system uses machine learning techniques to identify missing welds by comparing feature characteristics of actual welds with those of a typical part model, and can notify operators of missing welds in real-time or after completion, while also disabling welding equipment if necessary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional part tracking systems only compare the number of detected welds to the expected number, then the system complexity is low, but the measurement precision of identifying which specific welds are missed is insufficient
Solution Approach 1:
The system segments the welding process into individual weld events, tracking each weld's unique characteristics (location, timing, parameters) separately. This allows precise identification of which specific welds are missing by comparing the sequence and properties of detected welds against the expected weld plan, rather than just counting total welds.
Solution Approach 2:
The patent replaces simple numerical counting with machine learning-based analysis of weld characteristics. The system uses ML models to analyze weld features (such as weld bead geometry, thermal patterns, or sensor signatures) to automatically identify missing or defective welds, substituting mechanical counting with intelligent pattern recognition.
2Reliability
If conventional systems only count welds, then the device complexity is low, but the reliability of detecting welding quality issues is insufficient
Solution Approach 1:
The system implements continuous feedback by monitoring each weld event in real-time, comparing detected weld characteristics against expected parameters from the weld plan. When deviations are detected (such as missing welds, out-of-sequence welding, or parameter violations), the system provides immediate feedback to operators or automatically adjusts the welding process to maintain quality.
Solution Approach 2:
The patent replaces simple weld counting with machine learning-based quality assessment. The ML system analyzes multiple weld characteristics (geometric features, thermal patterns, process parameters) to reliably detect welding issues, substituting basic numerical tracking with intelligent quality evaluation.
3Measurement precision
If the system uses machine learning techniques to identify missing welds, then the measurement precision of weld identification is improved, but the loss of time for data processing increases
Solution Approach 1:
The system performs preliminary actions by pre-processing weld data during the welding process itself, extracting and storing key features as they occur. The machine learning models are pre-trained on historical welding data, enabling rapid inference during production. This preliminary preparation minimizes real-time processing delays while maintaining high identification precision.
4Manufacturing precision
If the system tracks detailed weld characteristics, then the manufacturing precision of weld quality control is improved, but the loss of information processing resources increases
Solution Approach 1:
The system extracts only the most critical weld characteristics needed for quality control (such as weld location, timing, and key geometric parameters) while filtering out redundant information. This selective extraction approach maintains manufacturing precision by focusing on essential quality indicators while reducing the computational burden of processing all possible weld data.
Data Source
AI summary
Systems and methods for missing weld identification using machine learning techniques are described. In some examples, a part tracking system uses machine learning techniques to identify whether an operator has missed one or more welds when assembling a part. The part tracking system may additionally identify which specific welds were missed (e.g., the first weld, the third weld, the fifteenth weld, etc.). The part tracking system may be able to identify missing welds after a part has been completed, or in real-time, during assembly of the part. Identification of the particular weld(s) missed during the welding process can help an operator quickly assess and resolve any issues with the part being assembled, saving time and ensuring quality