Fusion Welding Control Using Physics-Based Defect Prediction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Fusion welding processes are complicated by various physical and mechanistic factors, leading to defects such as porosity and lack of penetration, which result in undesirable joint properties and potential crack initiation.

Innovation Solution

A physics-based modeling approach that analyzes weld data using sensors and algorithms to predict defect formation, providing feedback for process optimization during design and active control of welding machines to eliminate defects, incorporating thermal, mechanical, and fluid flow models for accurate defect prediction and control.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Strength

If fusion welding is performed with high energy input to ensure complete penetration and strong joints, then weld strength is improved, but defect formation (porosity, cracks) increases

Engineering Contradiction:
Improveweld strengthVSAvoiddefect formation
Core Design Contradiction:
StrengthVSObject-generated harmful factors

Solution Approach 1:

The physics-based model predicts defect formation before welding occurs, allowing process parameters to be optimized in advance to prevent defects while ensuring adequate penetration and strength. The model calculates thermal history, melt pool dynamics, and solidification patterns to identify parameter combinations that achieve strong joints without defects.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system uses sensors to capture real-time weld data and feeds this information back to the physics-based model, which adjusts process parameters dynamically to maintain optimal welding conditions. This closed-loop control prevents defect formation while ensuring complete penetration and weld strength.

Inventive Principle:
Principle #23Feedback

2Manufacturing precision

If complex physics-based modeling and real-time control systems are implemented to eliminate defects, then weld quality is improved, but device complexity increases

Engineering Contradiction:
Improveweld qualityVSAvoidcontrol system complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The system replaces complex mechanical trial-and-error welding processes with physics-based computational models that simulate thermal, mechanical, and fluid flow phenomena. This substitution enables accurate defect prediction and control through software algorithms rather than extensive physical experimentation and complex mechanical adjustments.

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

Solution Approach 2:

The physics-based model creates a virtual copy of the welding process, simulating thermal history, melt pool behavior, and solidification patterns. This virtual model allows defect prediction and parameter optimization without requiring complex physical measurement and adjustment systems, simplifying the actual welding equipment while maintaining high weld quality.

Inventive Principle:
Principle #26Copying

3Productivity

If real-time sensor data collection and analysis are used to predict and control defects, then productivity is improved through reduced rework, but loss of time increases during data processing

Engineering Contradiction:
Improveweld production efficiencyVSAvoiddata processing time
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The physics-based model performs rapid calculations of thermal history, melt pool dynamics, and defect formation tendencies based on process parameters before welding begins. This preliminary prediction allows immediate adjustment of parameters without time-consuming data collection and analysis during the welding process, maintaining high productivity while preventing defects.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system replaces time-consuming physical measurement and analysis during welding with physics-based computational models that rapidly predict defect formation based on process parameters. This substitution eliminates the need for extensive real-time data processing while maintaining accurate defect prediction and control, preserving welding productivity.

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

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 the development and optimization of defect-free welds through virtual design and closed-loop control, increasing efficiency and ensuring high-quality welds by predicting and correcting defects in real-time, particularly in robotic welding processes.

Implementation Method 1

incorporating thermal, mechanical, and fluid flow models for accurate defect prediction and control

Methodology Applied
Scientific EffectThermal conduction: Conduction (thermal)

Implementation Method 2

incorporating thermal, mechanical, and fluid flow models for accurate defect prediction and control

Methodology Applied
Scientific EffectFluid flow: Convection

Implementation Method 3

The method further includes receiving weld data about the weld. In one embodiment, receiving the weld data about the weld comprises capturing the weld data about the weld using a sensor, wherein the sensor is one or more of a camera, a pyrometer, or a spectrometer

Methodology Applied
Scientific EffectPyrometry: Phosphor Thermometry

Implementation Method 4

The method further includes receiving weld data about the weld. In one embodiment, receiving the weld data about the weld comprises capturing the weld data about the weld using a sensor, wherein the sensor is one or more of a camera, a pyrometer, or a spectrometer

Methodology Applied
Scientific EffectSpectroscopy: Absorption Spectroscopy

Data Source

PatentEP4252958A1Predictive optimization and control for fusion welding of metals
Publication Date: 2023.10.04 RTX CORP
  • EP4252958A1 patent drawingFigure 1
  • EP4252958A1 patent drawingFigure 2
  • EP4252958A1 patent drawingFigure 3

AI summary

A method includes receiving weld data about a weld (902). The method further includes analyzing, using a physics-based model, the weld data to predict a formation of a defect (913, 914, 915) in the weld (902). The method further includes providing feedback to enable process optimization during a design stage or active control during welding to control a welding machine (110) to correct for and eliminate the formation of the defect (913, 914, 915) in the weld (902).