Thermal Weld Bead Monitoring for Real-Time Quality Classification

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current welding quality monitoring methods are inadequate for ensuring consistent and non-destructive assessment of weld quality, particularly in industrial applications like automotive battery production, where uniform electrical resistance is critical.

Innovation Solution

A method utilizing a thermal camera to monitor the welding zone, dividing it into sub-areas, and analyzing temperature evolutions to classify weld quality through a trained classifier, which processes cooling curves and other features to estimate weld quality and detect defects.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional welding quality monitoring methods are used, then weld quality can be assessed, but the monitoring is destructive and cannot provide continuous real-time feedback

Engineering Contradiction:
Improveweld quality assuranceVSAvoidcontinuous non-destructive monitoring
Core Design Contradiction:
ReliabilityVSEase of operation

Solution Approach 1:

The patent replaces traditional mechanical/destructive testing methods with optical detection using a camera system. The camera captures images of the welding zone, and image processing algorithms analyze the visual data to assess weld quality non-destructively and continuously, eliminating the need for physical sampling or destruction of welds for quality verification

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent implements continuous real-time monitoring by capturing a sequence of images during the welding process and processing them continuously. The system provides ongoing quality assessment throughout the welding operation, enabling continuous feedback rather than intermittent or post-process inspection, thus maintaining continuous useful action for quality assurance

Inventive Principle:
Principle #20Continuity of useful action

2Productivity

If welding quality is monitored in real-time, then continuous feedback is provided, but the complexity of the monitoring system increases

Engineering Contradiction:
Improvecontinuous monitoring capabilityVSAvoidmonitoring system complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent divides the welding zone into multiple regions of interest (ROIs) and processes images from different perspectives (front view, side view, rear view). By segmenting the monitoring task into multiple focused analysis zones and viewing angles, the system achieves comprehensive continuous monitoring while managing complexity through modular region-based processing rather than attempting to analyze the entire welding zone as a single complex unit

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent uses image copying and processing techniques where multiple images of the welding zone are captured and processed to extract quality information. The system creates digital copies of the welding process through imaging and analyzes these copies to assess weld quality, avoiding the need for direct physical intervention in the welding process itself

Inventive Principle:
Principle #26Copying

3Measurement precision

If the welding zone is divided into multiple sub-areas for analysis, then measurement precision improves, but the processing complexity increases

Engineering Contradiction:
Improvetemperature evolution analysis precisionVSAvoidimage processing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent divides the welding zone into multiple regions of interest (ROIs) and further segments each ROI into sub-areas for detailed temperature evolution analysis. This segmentation enables precise measurement of temperature changes in specific zones by tracking pixel intensity variations over time in each sub-area, achieving high measurement precision through systematic division of the analysis domain into manageable segments

Inventive Principle:
Principle #1Segmentation

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, non-destructive monitoring of weld quality by accurately classifying welds as good or defective, identifying potential defects, and providing real-time feedback to improve the welding process.

Implementation Method 1

a thermal camera (3) is provided, which is configured for obtaining, in a given area corresponding to the welding zone (SA), a sequence of thermal images or frames

Methodology Applied
Scientific EffectThermal radiation: Thermal Radiation

Data Source

PatentEP4096861B1Method of monitoring the quality of a weld bead, related welding station and computer-program product
Publication Date: 2024.03.20 COMAU SPA
  • EP4096861B1 patent drawingFigure 1~2C
  • EP4096861B1 patent drawingFigure 3~4
  • EP4096861B1 patent drawingFigure 5~6

AI summary

Described herein is a method for analysing the quality of a weld bead in a welding zone. The weld bead is generated by means of a continuous welding operation, wherein an energy beam emitted by a source with corresponding welding head follows a welding path, thereby melting the material of at least two metal pieces. The method comprises monitoring the welding zone via a thermal camera, wherein the thermal camera supplies a thermal image (IMG) in which a given area corresponds to the welding zone, and dividing (300) the area into a plurality of sub-areas and determining for each sub-area a respective temperature (Ti). During a learning step, the temperature evolution (Ti(t)) of each sub-area is monitored for different welding conditions. During a training step, the temperature evolutions (Ti(t)) are processed for training a classifier (304). For this purpose, a respective cooling curve is extracted (302) from each temperature evolution (Ti(t)), and parameters (F) are determined that identify the shape of each cooling curve. In particular, these parameters (F) are used as input features for the classifier (304). During a normal welding operating step (1006), the temperature evolution (Ti(t)) of each sub-area (Ai) can thus be monitored again, and the classifier (304) can be used for estimating the respective weld quality (S).