Additive Manufacturing Fault Detection Using Multi-Model Sensor Comparison

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional direct metal laser melting (DMLM) process quality control is inadequate for real-time monitoring, as it struggles to detect process deviations and faults due to measurement variability from part geometry, scan path history, and optic-induced distortions, leading to potential defects and poor part quality.

Innovation Solution

The implementation of time-dependent and spatially-dependent models for in-process monitoring, using sensors like photodiodes and pyrometers, along with a multiple model hypothesis testing algorithm to differentiate between normal and faulty conditions, and a probabilistic comparison of model outputs to measured sensor data for real-time fault detection and quality metric generation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional between-build diagnostics are used for quality control, then device complexity is reduced, but measurement precision and reliability of fault detection deteriorate due to inability to catch process deviations in real-time

Engineering Contradiction:
Improvefault detection precisionVSAvoidmonitoring system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the monitoring task by dividing the build process into discrete layers and further into laser-on transients (individual laser pulses). Each segment is independently analyzed by comparing sensor measurements against pre-computed model predictions for that specific segment, enabling precise local fault detection without requiring complex global analysis

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system performs preliminary action by pre-computing model predictions for normal process behavior under various conditions before the actual build. These pre-computed models (including spatially-dependent and time-dependent components) are stored and readily available for immediate comparison with real-time sensor data, eliminating the need for complex real-time computation during monitoring

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If steady state averaged measurement is used for process monitoring, then device complexity is reduced, but measurement precision deteriorates due to inability to resolve changes in individual parameters

Engineering Contradiction:
Improveparameter change resolutionVSAvoiddiagnostics complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies periodic action by analyzing laser-on transients (individual laser pulses) rather than continuous averaged signals. Each laser pulse creates a periodic measurement opportunity that captures transient melt pool behavior, allowing resolution of individual parameter changes through temporal sampling at the pulse frequency without requiring complex continuous analysis

Inventive Principle:
Principle #19Periodic action

Solution Approach 2:

The system transitions from spatial averaging (steady state) to temporal dimension analysis by examining time-dependent laser-on transient responses. This dimensional shift allows resolution of parameter changes through time-resolved measurements of melt pool characteristics during each laser pulse cycle

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Reliability

If multiple diagnostics are added to discriminate parameter shifts, then measurement precision improves, but device complexity and computation time increase

Engineering Contradiction:
Improvefault detection reliabilityVSAvoidcomputation time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent uses copying by creating multiple pre-computed model predictions representing different normal process conditions (different laser powers, scan speeds, positions). These model copies are stored and directly compared against sensor measurements without requiring complex real-time computation or multiple diagnostic algorithms, achieving reliable fault detection through simple pattern matching

Inventive Principle:
Principle #26Copying

4Manufacturing precision

If conventional quality control is performed weeks later after heat treatments, then manufacturing precision of final part is improved, but productivity deteriorates due to delayed defect detection

Engineering Contradiction:
Improvepart qualityVSAvoiddefect detection speed
Core Design Contradiction:
Manufacturing precisionVSProductivity

Solution Approach 1:

The system implements real-time feedback by continuously comparing sensor measurements during the build against model predictions and providing immediate fault indications. This allows operators to detect and correct process deviations during manufacturing rather than weeks later, maintaining high manufacturing precision while dramatically improving productivity through early defect detection

Inventive Principle:
Principle #23Feedback

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 ability to detect process deviations and faults in real-time, improving part quality by accounting for measurement variability and allowing for flexible tuning of performance specifications, thereby reducing the risk of defects and improving mechanical properties.

Implementation Method 1

using sensors like photodiodes and pyrometers

Methodology Applied
Scientific EffectPhotodetection: Photoelectric Effect

Implementation Method 2

uses a laser fired into a bed of powdered metal, with the laser being aimed automatically at points in space defined by a 3D model, thereby melting the material together

Methodology Applied
Scientific EffectLaser heating: Laser

Implementation Method 3

The metal powder is fused into a solid part by melting it locally using the focused laser beam

Methodology Applied
Scientific EffectMelting: Melting

Data Source

PatentEP4040250B1System and methods for generating fault indications for an additive manufacturing process based on a probabilistic comparison of the outputs of multiple process models to measured sensor data
Publication Date: 2023.09.13 GENERAL ELECTRIC CO
  • EP4040250B1 patent drawingFigure 1~2
  • EP4040250B1 patent drawingFigure 3
  • EP4040250B1 patent drawingFigure 4A

AI summary

Generating fault indications for an additive manufacturing machine (110) based on a comparison of the outputs of multiple process models (150) to measured sensor data.