Arc Welding Signal Spectral Analysis for Weld Failure Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing arc welding methods face challenges in accurately determining weld failures due to noise susceptibility in time series waveforms of weld voltage and current, leading to inadequate detection of subtle changes and types of weld failures.

Innovation Solution

An arc welding system that performs frequency analysis on time series waveforms of weld voltage and current to generate frequency spectral data, which is then used by a determiner to assess the welding state, including the presence and type of weld failures, through comparison with normal spectral data and machine learning algorithms.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If frequency analysis is performed on time series waveforms to generate frequency spectral data, then measurement precision of welding state is improved, but device complexity increases due to addition of analyzer and determiner components

Engineering Contradiction:

Improveweld failure detection accuracyVSAvoidsystem structure

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent introduces an analyzer as an intermediary component that processes the raw time series waveforms of weld voltage and current through frequency analysis. This intermediary transforms the complex time-domain signals into frequency spectral data, making subtle weld failures detectable without requiring direct complex processing in the main control system.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces traditional statistical analysis methods with frequency analysis methodology. Instead of using conventional time-domain statistical parameters, the system uses frequency-domain spectral analysis to detect weld failures, substituting one analytical approach with another that is more sensitive to subtle changes.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Device complexity

If statistical analysis of time series waveforms is used to determine weld failures, then device complexity is kept simple, but measurement precision deteriorates due to noise susceptibility

Engineering Contradiction:

Improvesystem structureVSAvoidweld failure detection accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent transitions from analyzing data in the time domain to analyzing data in the frequency domain. By performing frequency analysis on the time series waveforms, the system extracts frequency spectral data that reveals patterns and subtle changes not visible in the original time-domain signals, effectively adding a new dimension to the analysis.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

Data Source

PatentUS10906116B2Arc welding system and arc weld determination device

Publication Date: 2021.02.02 YASKAWA DENKI KK

AI summary

An arc welding system includes an arc welding portion that executes an arc welding, an arc welding power supply that supplies a weld voltage and a weld current to the arc welding portion, an analyzer including analyzer circuitry that performs a frequency analysis on a time series waveform of at least one of the weld voltage and the weld current and generates frequency spectral data, and a determiner including determiner circuitry that obtains a welding state of a welding work by the arc welding portion based on the frequency spectral data generated by the analyzer.