Computer implemented method of and optimisation tool for refinement of laser cutting processing parameters by means of an optimization tool

1. A computer-implemented method for calculating process parameters, being laser cutting parameters, which are machine and material specifically optimized for laser cutting a workpiece, being a sheet metal, with specific material properties by means of a laser machine with specific machine properties, comprising the method steps of:

Determining at least one of material of the workpiece and/or machine properties for which the process parameters are to be optimized;

Determining preconfigured initial process parameters;

Executing a re-optimization algorithm based on a statistical model accessed from a storage, and optimizing the process parameters based on the at least one of determined material of the workpiece and/or machine properties until a target objective function is achieved; and

Controlling a laser cutting process for the workpiece with the optimized process parameters, wherein execution of the re-optimization algorithm comprises:

Controlling execution of test processes with the determined preconfigured initial process parameters;

Evaluating the test processes by using a measurement-based algorithm or a preference-based algorithm for determining preferences for the test processes; and

Re-defining a parameter domain for the re-optimization algorithm based on the determined preferences.

2. The method according to claim 1 , wherein the measurement-based algorithm is based on Bayesian Optimization using Gaussian Processes.

3. The method according to claim 1 , wherein the preference-based algorithm is based on a pairwise comparison algorithm by comparing each of the test processes to select optimal test processes and wherein the determined preferences are implemented as constraints.

4. The method according to claim 1 , wherein the material properties comprise at least one of a material type dataset and a thickness dataset.

5. The method according to claim 1 , wherein evaluating the test processes is based on a standardized quality measure comprising a set of quality indices, which consists of slag residue, kerf width, burr height, perpendicularity of cutting edge, roughness, and/or robustness of the laser cutting process, in particular in dependence of a feed rate of a laser head.

6. The method according to claim 1 , wherein determining the preconfigured initial process parameters is performed by selecting from a set of process parameters, parameters which are determined to be optimal for the same or for similar material properties as the determined material properties.

7. The method according to claim 1 , wherein for re-defining the parameter domain, a physical model of the laser cutting process is used.

8. The method according to claim 1 , wherein evaluating the test processes is based on generating a standardized quality metric for different types of sensor data, including in particular optical sensor data by means of using a set of optical sensors and/or diodes.

9. The method according to claim 1 , wherein the re-optimization optimizes the target objective function is customer specific and wherein the target objective function include a maximization of at least one quality index of test processes, in particular, test cuts, a maximization of a feed rate of a laser head, and/or a minimization of power consumption of the laser machine.

10. The method according to claim 1 , wherein the re-optimization algorithm continuously trains on the statistical model and not on a neural network model.

11. The method according to claim 1 , wherein the re-optimization algorithm implements a closed loop control for automatically calculating re-optimized process parameters by learning a surrogate function, based on the determined preconfigured initial process parameters and results of the evaluated test processes.

12. The method according to claim 1 , wherein the re-optimization algorithm uses a surrogate function to optimize the target objective function.

13. An optimization tool configured to execute a calculating of optimized machine and material specific process parameters including cutting parameters, for a laser cutting process for cutting a workpiece, being a sheet metal, to be executed on a laser machine having specific machine properties, the optimization tool comprising:

a property interface configured to receive at least one of determined material of the workpiece and machine properties for which the machine and material specific process parameters are to be optimized;

a parameter interface configured to receive preconfigured initial process parameters;

a processor configured to execute a re-optimization algorithm based on a statistical model accessed from a storage, and optimize the machine and material specific process parameters based on the at least one of determined material of the workpiece and machine properties until a target objective function is minimized or maximized;

an output interface configured to output the optimized machine and material specific process parameters to a controller of the laser machine, wherein the controller of the laser machine configured to control the laser cutting process for the workpiece with the optimized machine and material specific process parameters, wherein the processor comprises:

I. a controller configured to control execution of test processes with the determined preconfigured initial process parameters;

II. an evaluation module configured to evaluate the test processes by using a measurement-based algorithm or a preference-based algorithm for determining preferences for the test processes; and

wherein the processor is further configured to re-define a parameter domain for the re-optimization algorithm based on the determined preferences of the evaluation module.

14. A non-transitory computer-readable medium storing a computer program comprising computer program code, that when executed by a processor, causes an optimization tool to:

determine at least one of material of a workpiece and machine properties for which process parameters are to be optimized by the optimization tool;

control a property interface configured to receive the at least one of determined material of the workpiece and machine properties for which the process parameters are to be optimized;

control a parameter interface configured to receive preconfigured initial process parameters;

control a processor configured to execute a re-optimization algorithm based on a statistical model accessed from a storage, and optimize the process parameters based on the at least one of determined material of the workpiece and machine properties until a target objective function is minimized or maximized;

control an output interface configured to output the optimized process parameters to a controller of a laser machine, wherein the controller of the laser machine configured to control a laser cutting process for the workpiece with the optimized process parameters, wherein the processor comprises:

I. a controller configured to control execution of test processes with the determined preconfigured initial process parameters;

II. an evaluation module configured to evaluate the test processes by using a measurement-based algorithm or a preference-based algorithm for determining preferences for the test processes; and

wherein the processor is adapted to re-define a parameter domain for the re-optimization algorithm based on the determined preferences of the evaluation module.