Corrective Welding Seam Analysis Using AI Fault Classification

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Manual evaluation of welding seams for corrective welding is subjective, error-prone, and time-consuming, making it unsuitable for high-throughput analysis of large quantities of workpieces, especially when determining corrective welding feasibility and parameters.

Innovation Solution

Implementing artificial intelligence-based image analysis using deep convolutional networks and few-shot/one-shot training to identify seam positions and fault classes, enabling automated determination of corrective welding feasibility and parameters.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If manual evaluation of welding seams is used, then human cognitive abilities can identify fault classes and determine corrective welding parameters, but the process becomes subjective, error-prone, and time-consuming

Engineering Contradiction:

Improveevaluation accuracyVSAvoidevaluation time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent replaces the manual mechanical evaluation process with an automated optical measurement system using cameras and laser scanners. This substitution eliminates human subjectivity and errors while significantly reducing evaluation time, allowing for rapid automated classification of welding seam faults and determination of corrective welding parameters

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service by allowing the automated measurement system to independently identify fault classes and determine corrective welding parameters without human intervention. The computer-aided classification algorithm automatically processes measurement data, classifies defects, and generates welding parameters, making the entire evaluation process autonomous and eliminating dependency on manual human assessment

Inventive Principle:

Principle #25Self-service

2Reliability

If manual evaluation is used for determining corrective welding feasibility, then human expertise can be applied, but high throughput analysis of large quantities of workpieces becomes unsuitable

Engineering Contradiction:

Improvefault classification accuracyVSAvoidthroughput

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The patent replaces manual expert evaluation with an automated computer-aided classification system that uses optical measurement data and algorithms to identify fault classes and determine corrective welding feasibility. This substitution enables high-throughput processing of large quantities of workpieces while maintaining reliable and consistent classification accuracy, eliminating the bottleneck of manual evaluation

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system performs preliminary automated classification and feasibility determination before corrective welding is executed. By pre-processing and classifying all welding seams automatically, the system prepares welding parameters and feasibility assessments in advance, enabling rapid processing of multiple workpieces without requiring manual intervention for each piece, thus achieving high throughput

Inventive Principle:

Principle #10Preliminary action

3Productivity

If automated image recognition is implemented, then fast and reliable analysis can be achieved, but algorithmic positioning for all fault classes cannot currently be solved

Engineering Contradiction:

Improveanalysis speedVSAvoidalgorithmic positioning capability

Core Design Contradiction:

ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent performs preliminary automated positioning and measurement of welding seams using camera-based vision systems and laser scanners before classification. By capturing comprehensive measurement data and pre-positioning the seam geometry algorithmically, the system creates a foundation that enables subsequent automated fault class classification and corrective welding parameter determination, gradually expanding algorithmic capability across different fault types

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system uses parameter changes by adjusting camera angles, lighting conditions, and measurement parameters to capture welding seams with various fault classes. By varying acquisition parameters and using multiple viewing angles, the system enables algorithmic recognition and positioning of different fault types, expanding the versatility of automated image recognition across diverse welding defects

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS20250289077A1Corrective welding method and seam processing device for corrective welding

Publication Date: 2025.09.18 TRUMPF LASER GMBH CO KG

AI summary

A method for corrective welding of workpieces is provided. At least a first workpiece of the workpieces has at least one faulty welding seam. The method includes providing the first workpiece that has the at least one faulty welding seam, capturing image data of the first workpiece using a camera, performing a first artificial intelligence-based image analysis of the image data in order to identify a seam position of the at least one faulty welding seam, performing a second artificial intelligence-based image analysis of the image data in order to identify at least one fault class of the at least one faulty welding seam, and determining whether a corrective welding of the at least one faulty welding seam is possible based on the seam position and the fault class of the at least one faulty welding seam.