Laser Machining Parameter Learning for Precise Process Prediction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for determining laser machining parameters are limited by the need for empirical tests and extrapolation, leading to imprecise results and restricted applicability to local data ranges, failing to maximize the capabilities of laser machining devices efficiently and effectively.

Innovation Solution

An automated learning method that uses previous experimental data to determine optimal machining parameters through a learning function, considering material-specific interaction parameters like delta, threshold fluence, incubation coefficient, and complex refractive index, allowing for faithful reproduction of target machining results across extended ranges.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If linear response models are used to model laser-material interaction, then the model is simple to implement, but the range over which the model exhibits a reliable response is limited and does not cover sufficient range to allow the laser machining device to be used to its maximum capabilities

Engineering Contradiction:
Improvemodel complexityVSAvoidparameter range coverage
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent transitions from linear response models to non-linear models that account for cumulative damage effects through the incubation coefficient. This parameter change allows the model to accurately represent laser-material interaction across a wider range of fluence values and pulse numbers, enabling the system to operate at its maximum capabilities while maintaining model reliability.

Inventive Principle:
Principle #35Parameter changes

2Loss of time

If extrapolation of experimental machining results is used to determine laser machining parameters, then the method requires fewer preliminary tests, but the results are imprecise and the assessment of parameter adequacy is left to the user

Engineering Contradiction:
Improvepreliminary testing timeVSAvoidparameter determination precision
Core Design Contradiction:
Loss of timeVSMeasurement precision

Solution Approach 1:

The patent implements a feedback mechanism where the model continuously compares predicted machining results with actual experimental results. The incubation coefficient and other parameters are adjusted based on this feedback, allowing the system to self-correct and improve precision over time rather than relying on user assessment of extrapolated results.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent performs preliminary modeling and simulation before actual machining operations. By using the non-linear model to predict optimal parameters beforehand and then validating these predictions with minimal experimental tests, the system reduces the need for extensive preliminary testing while maintaining high precision in parameter determination.

Inventive Principle:
Principle #10Preliminary action

3Measurement precision

If extensive empirical tests are conducted to determine optimal laser machining parameters, then the precision of parameter determination is improved, but the time and cost required for parameter optimization increases significantly

Engineering Contradiction:
Improveparameter determination precisionVSAvoidparameter optimization time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent replaces the mechanical trial-and-error approach with a computational model-based system. The non-linear model incorporating incubation effects allows for virtual experimentation and parameter optimization through simulation, significantly reducing the need for physical empirical tests while maintaining or improving parameter determination precision.

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

Solution Approach 2:

The system performs preliminary parameter optimization through computational modeling before actual machining. By pre-calculating optimal parameters using the non-linear model and then verifying with minimal tests, the patent achieves high precision parameter determination without requiring extensive empirical testing campaigns.

Inventive Principle:
Principle #10Preliminary action

4Device complexity

If local extrapolation around experimental results is used, then the method is simple to implement, but it does not allow determination of laser machining parameters over an extended data range

Engineering Contradiction:
Improvemethod complexityVSAvoiddata range coverage
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent introduces the incubation coefficient as a key parameter that enables the model to handle extended data ranges. This parameter accounts for cumulative damage effects that occur over multiple pulses, allowing the model to accurately predict machining results across a wide range of fluence values and pulse numbers rather than being limited to local extrapolation around single experimental points.

Inventive Principle:
Principle #35Parameter changes

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 enables precise and efficient determination of laser machining parameters without extensive empirical testing, ensuring high-quality machining results for various materials and machining systems by leveraging machine learning algorithms to predict optimal parameters for different machining conditions.

Implementation Method 1

Laser beams are sometimes used for machining parts. It is in fact possible to melt, evaporate or sublimate part of a material exposed to a laser beam.

Methodology Applied
Scientific EffectAbsorption (EM radiation): Absorption (EM radiation)

Implementation Method 2

Laser beams are sometimes used for machining parts. It is in fact possible to melt, evaporate or sublimate part of a material exposed to a laser beam.

Methodology Applied
Scientific EffectLaser ablation: Laser Ablation

Implementation Method 3

The method of the invention comprises an automated learning step which allows, on the basis of previous experimental results, the determination of machining parameters for the machining of a result to be achieved.

Methodology Applied
Scientific EffectMachine learning:

Data Source

PatentEP3743236B1Method for determining laser machining parameters and laser machining device using this method
Publication Date: 2023.11.29 LASER ENG APPL
  • EP3743236B1 patent drawingFigure 1
  • EP3743236B1 patent drawingFigure 2
  • EP3743236B1 patent drawingFigure 3a~3c

AI summary

A method for determining laser machining parameters for machining a material using a laser machining system comprises the following steps, inter alia: (a) providing the central unit with a machining learning function capable of learning on the basis of a plurality of machining data samples, the machining learning function comprising an algorithm capable of defining the following laser machining parameters for the machining result sought and for the machining system used: • a polarisation, • a pulse energy Ep, • a diameter at the focal point w, • a Gaussian order p, • a pulse repetition rate PRR of n pulses, • a wavelength; (b) learning on the basis of the machining learning function so as to enable the laser machining system to machine the material to be machined according to the machining result sought.