Laser Welding Monitoring Using Correlated Multi-Sensor Defect Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current laser welding monitoring systems often result in false rejections and escapes due to independent evaluation of physical quantities, lacking a reliable and automated method to determine weld joint defects effectively.
Innovation Solution
A method that correlates sensor data from multiple sensors and processing parameters to evaluate the quality of weld joints, reducing false rejection and escape rates by combining sensor data for defect determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple individual monitoring systems are used to evaluate different physical quantities separately, then the monitoring coverage is improved, but the reliability of defect detection deteriorates due to false rejections and escapes
Solution Approach 1:
The patent combines sensor data from multiple individual monitoring systems that evaluate different physical quantities (geometric features, heat radiation, process light) into a unified evaluation approach. By correlating and jointly analyzing these previously separate data streams, the system achieves both comprehensive monitoring coverage and improved defect detection reliability, eliminating false rejections and escapes caused by independent evaluation.
2Device complexity
If sensor data from multiple sensors are evaluated independently, then the complexity of individual sensor evaluation is reduced, but the manufacturing precision of weld quality assessment deteriorates
Solution Approach 1:
The patent merges the evaluation of sensor data from multiple sensors by correlating their outputs in a unified analysis framework. This approach maintains the simplicity of individual sensor evaluation while significantly improving weld quality assessment precision through the combined information from geometric features, heat radiation, and process light measurements.
3Ease of manufacture
If separate defect detection and classification processes are used, then the modularity of the monitoring system is improved, but the productivity of quality determination deteriorates due to independent processing steps
Solution Approach 1:
The patent combines the defect detection and classification processes into a single integrated evaluation step that simultaneously performs both functions. By correlating sensor data from multiple sources and processing it through a unified analysis method, the system maintains modular sensor components while achieving high productivity through simultaneous defect detection and classification without separate processing steps.
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
This approach enhances the reliability of weld joint evaluation, reducing false classifications and improving the quality assessment of laser welding processes by using correlated sensor data and processing parameters.
Implementation Method 1
the laser beam emitted from a laser beam source or one end of a laser guiding fibre is irradiated and focused onto the work pieces with the aid of beam guiding and focusing optics in order to weld work pieces
Implementation Method 2
the material of the work piece heats up to such an extent by the irradiated laser power that it vaporises. The vapour capillary is surrounded by a melt pool in which the material is in a molten state
Implementation Method 3
which use different physical quantities, e.g. geometric features in a camera image, heat radiation and process light, for evaluation and analysis
Data Source
AI summary
A method for monitoring a laser welding process includes the steps of: performing the laser welding process by irradiating a laser beam onto at least one work piece to form a weld joint, wherein the laser beam is guided along a processing path by a deflection device, detecting sensor data by at least two sensors during the laser welding process, and determining whether the weld joint has a defect based on the detected sensor data and at least one processing parameter of the laser welding process. An associated laser welding system is also provided.