Automatic Parameterisation of a Laser Cutting Method

Abstract

A high current contact is disclosed having a contact pin for insertion into the high-current socket having a plurality of contact segments that are slotted in a radial direction for contacting an inner contact surface of the high-current socket; a guide sleeve surrounding the contact pin, which, by means of an at least central front pressing against the high-current socket relative to the contact pin in an axial direction from an initial position, in which the guide sleeve blocks an independent radial spreading of the contact segments in order to avoid a contact between the contacts segments protruding axially from the guide sleeve and the inner contact surface, is movable into a contact position that is set back with respect to the contact pin and in which the guide sleeve unblocks an independent radial spreading of the contact segments protruding from the guide sleeve for contacting the inner contact surface.

Claims

1.-18. (canceled)

19. A method for self-checking a deterministic process model (M) of a laser cutting process, wherein when the method is provided a parameter set as input, the method comprises the following: reading in the parameter set configured to drive a laser cutting tool; accessing a deterministic process model in order to calculate a forecasting dataset for the machining result for the read-in parameter set, wherein the deterministic process model uses at least three data structures and models physical, multidimensional relationships between the respective parameter set and the process characteristics representing the physical laser cutting process and the machining result; executing a laser cutting with the read-in parameter set; measuring a machining result as ACTUAL values of the manufactured component; comparing the measured ACTUAL values for the machining result with the forecasting dataset for the machining result by providing a comparison result and wherein when the method is provided a TARGET value for a machining result as input, executes the following parameterisation steps: reading in the TARGET value for a machining result; accessing the deterministic process model in order to calculate at least one parameter set for the read-in TARGET value for the machining result which satisfies the TARGET value, and wherein the forecasting method is a digital counterpart of the parameterisation method and mathematical inverse thereto; cutting a workpiece with the calculated parameter set; measuring an ACTUAL value for the machining result on the cut workpiece; comparing between the TARGET value and the ACTUAL value (E.sub.act) for the machining result; wherein the method considers the comparison result as a deviation and adapts the deterministic process model upon detection of a reproducible deviation; and the cutting, measuring, comparison and deterministic process model adaptation are iteratively repeated until a deviation between the TARGET value and the ACTUAL value for the cutting result falls below a preconfigurable threshold value.

20. The method according to claim 19, wherein the machining result is at least one of a data tuple with measurable information on the quality of the laser cutting process; and information on at least one of a burr height in a melt cut, a slag height in a flame cut, an edge slope, an edge waviness and an edge profile height, a surface roughness, a striation height, and a striation frequency.

21. A computer program having program code or program means, wherein, if the computer program is executed on a computer or a computer-based processing unit (V, ParM, ProgM), the computer program can be stored on a computer readable medium, wherein the program code or the program means causes a computer to execute a method for self-checking a deterministic process model (M) of a laser cutting process, wherein when the method is provided a parameter set as input, the method comprises the following: reading in the parameter set configured to drive a laser cutting tool; accessing a deterministic process model in order to calculate a forecasting dataset for the machining result for the read-in parameter set, wherein the deterministic process model uses at least three data structures and models physical, multidimensional relationships between the respective parameter set and the process characteristics representing the physical laser cutting process and the machining result; executing a laser cutting with the read-in parameter set; measuring a machining result as ACTUAL values of the manufactured component; comparing the measured ACTUAL values for the machining result with the forecasting dataset for the machining result by providing a comparison result and wherein when the method is provided a TARGET value for a machining result as input, executes the following parameterisation steps: reading in the TARGET value for a machining result; accessing the deterministic process model in order to calculate at least one parameter set for the read-in TARGET value for the machining result which satisfies the TARGET value, and wherein the forecasting method is a digital counterpart of the parameterisation method and mathematical inverse thereto; cutting a workpiece with the calculated parameter set; measuring an ACTUAL value for the machining result on the cut workpiece; comparing between the TARGET value and the ACTUAL value (E.sub.act) for the machining result; wherein the method considers the comparison result as a deviation and adapts the deterministic process model upon detection of a reproducible deviation; and the cutting, measuring, comparison and deterministic process model adaptation are iteratively repeated until a deviation between the TARGET value and the ACTUAL value for the cutting result falls below a preconfigurable threshold value.

22. A laser cutting tool with a parameterisation module and a with a forecasting module configured for self-checking a deterministic process model of a laser cutting process, wherein when the method is provided a parameter set as input, executes the following steps: reading in the parameter set which serves to drive a laser cutting tool; accessing the deterministic process model in order to calculate a forecasting dataset for the machining result for the read-in parameter set, wherein the deterministic process model uses at least three data structures, and models physical, multidimensional relationships between the respective parameter set and the process characteristics representing the physical laser cutting process and the machining result; executing a laser cutting with the read-in parameter set; measuring a machining result as ACTUAL values of the manufactured component; comparing the measured ACTUAL values for the machining result with the forecasting dataset for the machining result by providing a comparison result; and wherein when the method is provided a TARGET value for a machining result as input, executes the following parameterisation steps: reading in the TARGET value for a machining result; accessing the deterministic process model in order to calculate at least one parameter set for the read-in TARGET value for the machining result which satisfies the TARGET value, and wherein the forecasting method is a digital counterpart of the parameterisation method and mathematical inverse thereto; cutting a workpiece with the calculated parameter set; measuring an ACTUAL value for the machining result on the cut workpiece; comparing between the TARGET value and the ACTUAL value for the machining result wherein the method considers the comparison result as a deviation and adapts the deterministic process model upon detection of a reproducible deviation; and wherein the cutting, measuring, comparison and deterministic process model adaptation are iteratively repeated until a deviation between the TARGET value and the ACTUAL value for the cutting result falls below a preconfigurable threshold value.

23. A system for operating a laser cutting tool (L) for cutting workpieces, comprising: A laser cutting tool with a parameterisation module and a with a forecasting module configured for self-checking a deterministic process model of a laser cutting process, wherein when the method is provided a parameter set as input, executes the following steps: reading in the parameter set which serves to drive a laser cutting tool; accessing the deterministic process model in order to calculate a forecasting dataset for the machining result for the read-in parameter set, wherein the deterministic process model uses at least three data structures, and models physical, multidimensional relationships between the respective parameter set and the process characteristics representing the physical laser cutting process and the machining result; executing a laser cutting with the read-in parameter set; measuring a machining result as ACTUAL values of the manufactured component; comparing the measured ACTUAL values for the machining result with the forecasting dataset for the machining result by providing a comparison result; and wherein when the method is provided a TARGET value for a machining result as input, executes the following parameterisation steps: reading in the TARGET value for a machining result; accessing the deterministic process model in order to calculate at least one parameter set for the read-in TARGET value for the machining result which satisfies the TARGET value, and wherein the forecasting method is a digital counterpart of the parameterisation method and mathematical inverse thereto; cutting a workpiece with the calculated parameter set; measuring an ACTUAL value for the machining result on the cut workpiece; comparing between the TARGET value and the ACTUAL value for the machining result wherein the method considers the comparison result as a deviation and adapts the deterministic process model upon detection of a reproducible deviation; and wherein the cutting, measuring, comparison and deterministic process model adaptation are iteratively repeated until a deviation between the TARGET value and the ACTUAL value for the cutting result falls below a preconfigurable threshold value; a user interface for inputting and outputting data; a memory for storing a deterministic process model with: a first data structure in which a number of parameter sets is stored, wherein the first data structure represents a data interface to the laser cutting tool; a second data structure in which a number of calculated process characteristics is stored; and a third data structure in which a number of cutting results is stored, wherein the third data structure represents a data interface to a measuring device for measuring cutting results from the laser cutting tool; an electronic processing unit configured to execute a set of model algorithms on the deterministic process model; a data connection between the data structures, the memory, the electronic processing unit and the laser cutting tool; and a comparator module configured for automatic self-checking of the deterministic process model by sequentially executing inverse model algorithms for each input and checking whether a result of the sequential execution is identical to the input.

24. The system according to claim 23, wherein the input is a parameter set or a TARGET value for the machining result.

Description

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS

[0085] Further advantages features and details of the various embodiments of this disclosure will become apparent from the ensuing description of a preferred exemplary embodiment and with the aid of the drawings. The features and combinations of features recited below in the description, as well as the features and feature combination shown after that in the drawing description or in the drawings alone, may be used not only in the particular combination recited, but also in other combinations on their own, with departing from the scope of the disclosure:

[0086] FIG. 1 depicts a schematic overview of a system for parameterisation and/or forecasting for a laser machining on a laser machining tool, according to a preferred embodiment of the invention;

[0087] FIG. 2 depicts a schematic representation of a model-based forecasting method;

[0088] FIG. 3 depicts a schematic representation of a model-based parameterisation method for a laser;

[0089] FIG. 4 depicts an overview of a parameterisation module integrated into a laser machining;

[0090] FIG. 5 depicts an overview of a forecasting module integrated into a laser machining;

[0091] FIG. 6 depicts a flowchart of a forecasting method according to a preferred embodiment of the invention;

[0092] FIG. 7 depicts a flowchart of a parameterisation method according to a preferred embodiment of the invention;

[0093] FIG. 8 depicts a flowchart for an application calibration algorithm;

[0094] FIG. 9 depicts a schematic representation of a kerf of a manufactured component for illustrating a cone angle;

[0095] FIG. 10 depicts a flowchart for an application calculation algorithm according to a preferred embodiment of the invention;

[0096] FIG. 11 depicts a flowchart for an application simulation algorithm according to a preferred embodiment of the invention;

[0097] FIG. 12 depicts a flowchart for a manufacturing forecasting algorithm according to a preferred embodiment of the invention; and,

[0098] FIG. 13 depicts a graph of an averaged profile curve of the profile height of the cut surface.

DETAILED DESCRIPTION OF THE INVENTION

[0099] As used throughout the present disclosure, unless specifically stated otherwise, the term “or” encompasses all possible combinations, except where infeasible. For example, the expression “A or B” shall mean A alone, B alone, or A and B together. If it is stated that a component includes “A, B, or C”, then, unless specifically stated otherwise or infeasible, the component may include A, or B, or C, or A and B, or A and C, or B and C, or A and B and C. Expressions such as “at least one of” do not necessarily modify an entirety of the following list and do not necessarily modify each member of the list, such that “at least one of” “A, B, and C” should be understood as including only one of A, only one of B, only one of C, or any combination of A, B, and C.

[0100] The invention serves to improve, and in particular to simplify, the operation of laser installations. The setting or parameterisation of the laser installation requires extensive physical knowledge of the relationships between the relevant variables. Thus, the installation parameters to be set or set have a multi-dimensionally coupled correlation to the machining result. The resulting complexity makes it difficult or impossible to intuitively set the installation parameters to achieve a specific machining result. In addition, the user knows little about how the machining process will run at the time of part design. The user also knows little about this at the time of machining. Due to the lack of characterisation and simulation possibilities of the process in the prior art, a process monitoring that is understandable for the user is difficult to impossible. For the user, it is also important to know which parameters must be set when specifying a TARGET value for a machining result, for example, with a specification for the burr height in a melt cut or for a slag height in a flame cut, for the respective manufactured parts so that the desired machining result can be achieved.

[0101] The aim of the method proposed here is to control the laser material machining (application) with minimal data input so as to lead to optimal manufacturing. The minimum data input includes the desired part shape (cutting plan) and the desired material (workpiece) as well as the specification of the installation and material. The optimal manufacturing means the achievement of an optimal machining result and is determined autonomously from at least three data structures. At least two application cases, each with at least two fast algorithms, define a data flow between the at least three data structures. The data structures, application cases, algorithms, and data flow form the digital twin of the laser material machining (application). The digital twin is modular and uses, among other things, empirical values to keep the algorithms simple and fast and to ensure the comparability between digital twin and real laser material machining (application). The model includes at least three data structures:

[0102] i. Data structure 1 contains the data of the parameter set P;

[0103] ii. Data structure 2 contains the process characteristics PKG; [0104] iii. Data structure 3 contains the data of the machining result E;

[0105] The data structure 1 forms the data interface between the digital twin of the laser material machining and the real machining installation L. The data structure 3 forms the data interface between the digital twin of the laser material machining and the machining result E of the real laser material machining (application). The minimum of three data structures are dependent on the minimum data input, including cutting plan (user input) and workpiece specifications (material, thickness, size from database), empirical values (averages of collected measured values of the various categories, from database) and/or current measured values (from database) initialised as measured variables or calculation variables, and if necessary initialised further specification data (from database). An advantage over known methods from the prior art is that empirical values can now also be incorporated into a model (here, the digital twin).

[0106] The data of the at least three data structures are stored in a database system (e.g., in tabular form). The algorithms of application cases 1 and 2 read the data from the tables or write them into the tables. To do this, the algorithms use a common data interface object to execute the read and write operations on the database system. The algorithms themselves store the data for the respective calculations in the class instances of the digital twin. The database system and the classes of the digital twin are advantageously constructed to be object-oriented and codified in a class diagram.

[0107] In a preferred embodiment of the invention, at least two application cases are provided:

i. Application case 1 is the sequential execution
a. of the application calibration algorithm, which maps the values of the machining result E (data structure 3) onto the process characteristics PKG (data structure 2), and
b. of the application calculation algorithm, which maps the process characteristics PKG (data structure 2) onto the parameter set P (data structure 1).
2. Application case 2 is the sequential execution
c. of the application simulation algorithm, which maps the parameter set P (data structure 1) onto the process characteristics PKG (data structure 2) maps, and
d. of the manufacturing forecasting algorithm, which maps the process characteristics PKG (data structure 2) into the machining result E (data structure 3).

[0108] In the following, the three data structures are explained in more detail.

[0109] The data structure 1 with the parameter set P for the laser material machining contains the parameter data, which by definition contain all influencing variables of the laser material machining (application). This means that the entirety of the data in the data structure 1 defines the digitised parameter data or the machining state before, during and after parts manufacturing. Thus, the data structure 1 is the digital representation of the realised laser material machining on the given installation. It has the function of a data interface between the machining installation and the digital twin. The influencing variables of the data structure 1 are further subdivided. They can include at least the following three sub-data structures of the parameter set (also referred to below as the “machining state”):

[0110] Specification:

the specifications of the manufacturing installation, the laser source, the machining head, the workpiece and other influencing components. Examples are specified maximum values for feed or process gas pressure, as well as the wavelength of the laser or material designations of the workpiece.

[0111] Contour:

geometric shape of the part contour as a single part or as a cutting plan. Examples are lengths, angles, curvatures, directions of rotation or coordinates of characteristic points on the contour.

[0112] Point Data:

all process parameters of the workpiece, the installation and the components thereof that can be changed over the part contour. Examples are laser power, feed, process gas pressure, nozzle spacing, focus position, but also variable workpiece properties such as thickness or temperature.

[0113] The data structure 2 (of the digital twin of the laser material machining) contains the process characteristics PKG, which comprises all essential geometric and physical properties of the process zone and the machining process. The entirety of the data in the data structure 2 represent the dynamic processes which interact in the process zone and indicate the properties of the melt and the resulting kerf in numbers. Added to this are productivity indicators, safety-relevant variables (e.g., proportion of radiation loss), and stability or tolerance values. The process characteristics are the result of calculations of the corresponding algorithms of application cases 1 and 2 over the part contour. The most important advantages of the process characteristics are:

1. as purely mathematical variables, they describe the intrinsic properties of the process zone without any influence of variations in measurement data, so they are 100% reproducible;
2. the process characteristics contain variables that either cannot be accessed, or are difficult to access, for a direct measurement, but that directly influence the machining result, which means a strong correlation with the manufacturing result;
3. the process characteristics have a theoretically formal and non-empirical relationship, both with the parameter set and with the machining result.

[0114] The process characteristics comprise at least the following six sub-data structures:

[0115] Beam Tool:

calculated variables for the laser beam focused in the workpiece. These include, for example, the Rayleigh length, the average laser power, and/or the beam radius at the workpiece top edge.

[0116] Kerf:

geometric characteristics of the kerf, such as the average angle of inclination of the cut front and/or various joint widths.

[0117] Melt Film:

geometrical characteristics and flow-mechanical characteristics of the melt, such as the ejection thickness or the Péclet number thereof, but also the ejection speed thereof, and further target values of the machining result based on the melt dynamics.

[0118] Nozzle:

contains geometric properties of the space between the nozzle and the process zone, but in particular also the pressure coupling efficiency.

[0119] Optics:

calculation results for the magnification and the power-dependent focus shift.

[0120] Workpiece:

various characteristics in connection with the given material, e.g., the required cutting power, path energy, process efficiency and/or stability parameters.

[0121] The third data structure 3 includes the machining result. The machining result includes all data belonging to the result of the machining. From a technical point of view, the values or variables of the machining result are the result of the response function of both digitised and realised parts manufacturing. This includes:

1. the quality measured on the finished component (e.g., burr or slag);
2. the measured kinetics variables measured during machining (installation-internal measurement of the effectively driven parameter dataset);
3. the process signals recorded during machining;
4. the quality forecasts calculated by the manufacturing forecasting algorithm.

[0122] Analogous to the data structure 1, the data structure 3 is also a digital image, namely that of the finished component and the quality thereof (machining result). Furthermore, and analogously to the data structure 1, the machining result also functions as a data interface between the measured values and signals of the parts manufacturing and the digital twin.

TABLE-US-00001 Quality variable Melt cut Flame cut Burr amplitude X Slag amplitude X Cutting edge profile height X X Surface roughness X Striation amplitude X Striation frequency X Striation trailing effect X X

[0123] According to the invention, a deterministic process model is provided which is identified in the figures by the reference symbol M and which can be stored in a memory, such as in a database DB. On the basis of this process model, a type of digital twin can be provided for the real laser machining or the associated physical processes.

[0124] The invention will be explained in more detail below with reference to the figures and with several exemplary embodiments.

[0125] FIG. 1 shows a laser installation L, which is controlled via an electronic processing unit V, e.g., a control computer or a microprocessor-based circuit (e.g., FPGA or ASIC or other integrated circuits). In this processing unit V, the process model M is stored in executable form. The processing unit V can also be designed to execute various algorithms, which will be described in more detail below.

[0126] In a first application shown in FIG. 3 and FIG. 1, a TARGET value for a desired machining result E.sub.targ is entered by the user, or the target value is acquired from a data structure or a message. From this target value for the machining result E.sub.targ, a parameter dataset P.sub.calcd is calculated, as shown in FIG. 3, by accessing the model M, which parameter dataset is to be set on the laser installation L so that the desired value can be achieved.

[0127] This first application is shown in FIG. 1 with dashed lines. Starting at the bottom right, the target value E.sub.targ is detected and transmitted to the model M, which calculates a parameter set P.sub.calcd therefrom using algorithms. This calculated parameter set P.sub.calcd can be transmitted to the laser installation L for control in a subsequent method. After further relevant variables (e.g., cutting plan and workpiece-related data) have been entered, the laser L is operated. Subsequently, the machining result is measured on the manufactured component or workpiece. This is recorded in the measured machining result E.sub.act. Optionally, an electronic analysis, in particular event-controlled and/or time-controlled further analysis, can be carried out. In particular, a comparison can now be made between the actually measured machining result E.sub.act and the target value E.sub.targ originally specified or entered by the user. The comparison result VGL1 is output, preferably on a monitor or a terminal T of the laser L and/or used for initiating further steps. In particular, the comparison value can be used to regulate the laser machining method. For example, a new target value for the desired machining result can be determined with the determined comparison value VGL1:

E.sub.target,new=E.sub.target+VGL.sub.1

[0128] This adaptation of the TARGET value can be controlled by a threshold of VGL1.

[0129] A second application case is shown schematically in FIG. 2. Here, the opposite case should be calculated. From a given parameter dataset P, a forecasting dataset for the machining result E.sub.fcst is created while accessing the model M.

[0130] In FIG. 1, this is represented starting with the input of the parameter set P, which is fed to the model M, so that a forecasting E.sub.fcst for the machining result E.sub.act can be generated, given the parameters P. According to the comparison process mentioned above, in case of application 1, even with the application 2, a comparison value VGL2 can then be calculated between the measured ACTUAL value for the machining result E.sub.act and the forecast value E.sub.fcst. This comparison value VGL2 is output on the terminal T.

[0131] With regard to the comparison value VGL2, the technical advantage is that a reproducible deviation from a refinement or calibration of the model can be used. Preferably, upon detection of such a reproducible deviation, a version step is automatically initiated to produce a calibrated model M′. This is represented in FIG. 1 by the solid line from the comparison VGL2 to the model M drawn in thinner line width to indicate that this is optional but a preferred embodiment of the invention. Thus, in this preferred embodiment of the invention, the comparison value VGL2 can be used to automatically calibrate the model M. Furthermore, a temporal and/or statistical evaluation can be carried out to evaluate in which machining phases of the laser there is no deviation and in which phases a deviation has been identified. If, for example, there was no TARGET/ACTUAL deviation in an initial phase, but a deviation has been determined at a later stage, this can possibly indicate an error and/or a gradual soiling of the focusing optics and/or other wear (e.g., on the tip of the process gas nozzle).

[0132] In conjunction with FIG. 4, the mode of operation of a parameterisation module ParM is explained below in an exemplary embodiment. The parameterisation module ParM can be implemented on the electronic processing unit V and serves to implement the first application case, which has been described above with reference to FIG. 3. This module is intended to support the user at the laser installation L during the setting and parameterisation thereof. For this purpose, the user enters a TARGET specification for the machining result E.sub.targ on a connected terminal T, on which a user interface UI can be provided. This data record is forwarded to the parameterisation module ParM via a corresponding data connection. The parameterisation module ParM then accesses the model M stored in a database DB with the corresponding algorithms, which are explained in more detail below, in order to calculate the parameter set P.sub.calcd for the TARGET value E.sub.targ. This can be output on the user interface UI and, upon detection of a verification signal, this calculated parameter set P.sub.calcd can be transmitted directly to the laser L in order to control the laser process.

[0133] Optionally, an additional checking of the model M can be performed in a subsequent period. This can be done by a comparator module KOMP by means of a comparison. For this purpose, a machining result E.sub.act is measured on the components manufactured with the parameter set and transmitted to the comparator KOMP. The comparator KOMP also has access to the user-entered command E.sub.targ and can compare these two data records. The comparison result is output, for example on the user interface UI. If the comparison result signals a deviation, this can automatically trigger a corrective action.

[0134] As indicated in FIG. 4 by the curved line CL, the above-mentioned components need not necessarily be implemented directly on the laser installation L. Preferably, the user interface UI is arranged on or at the laser installation L and the electronic processing units, such as the parameterisation module ParM, the comparator KOMP and/or the model M that can be provided as a distributed system and on different platforms in the database DB. Furthermore, all or selected components can also be provided as a service on a central server S or in a cloud CL, which can be accessed via a network connection.

[0135] In the following, the operation of a forecasting module ProgM is explained by way of example with reference to FIG. 5. The forecasting module ProgM serves to implement the second application case, which has been described above with reference to FIG. 2. The procedure for the forecasting module ProgM essentially corresponds to that which has already been described above with reference to FIG. 4. Only the input (a parameterisation P specified by the user or read in from a data source) and the output (a calculated forecast E.sub.fcst for the machining result) have been exchanged.

[0136] The forecasting module ProgM calculates a forecast E.sub.fcst by accessing the corresponding algorithms from the parameter set P entered. The calculated result E.sub.fcst is displayed on the user interface UI.

[0137] As explained above, the laser can now be operated in reality with the parameterisation P.

[0138] In a later phase, the machining result for this parameterisation P can be measured and fed to the comparator KOMP for comparison with the calculated forecast E.sub.fcst. In the event of a deviation, this can be output on the user interface UI, for example, to give the user the opportunity to find a modified parameterisation. In addition, further steps can be triggered based on the comparison. In particular, upon detection of a reproducible deviation, an adaptation of the model M can be made and tested as a new version and put into operation.

[0139] As shown in the exemplary embodiment illustrated in FIG. 5, the electronic processing units ProgM, DB can be provided on a central server S. This variant also applies to the parameterisation module ParM from FIG. 4. Likewise, the variants described in conjunction with FIG. 4 (cloud-based solution) also apply to the forecasting module ProgM.

[0140] FIG. 6 is a flowchart of a forecasting method that can be executed in different flow variants. After the start of the forecasting method, which can be carried out on the forecasting module ProgM, the data for the parameter set P which is to be used to control the laser machining tool L are read in in step S61. In step S62, the deterministic process model M is accessed to calculate in step S63 the forecasting dataset E.sub.fcst for the machining result for the read-in parameter set P. In step S64, optionally, the calculation result E.sub.fcst can be displayed on a user interface UI to allow the user to change the input directly.

[0141] In a further development, the laser machining with the read-in and possibly confirmed parameter set P can be performed in step S65. Subsequently, in step S66, the machining result E.sub.act can be measured as the ACTUAL value of the manufactured component to be compared in step S67 with the forecasting dataset E.sub.fcst for coincidence and deviation. In the event of a deviation, the calculated comparison value can be output in step S68 (e.g., onto the UI) and/or adaptation procedures can be automatically triggered on the basis of the calculated comparison value; in particular, an adaptation of the model M can be made upon detection of a reproducible deviation and tested as a new version and put into operation. Since the steps S64, S65, S66, S67, S68 are optional, these are shown in FIG. 6 outlined by dashed lines.

[0142] FIG. 7 shows a flow diagram of a parameterisation method which can be executed in different sequence variants. After the method has been started, in step S71 a TARGET value for a machining result E.sub.targ is read in. In step S72, the deterministic process model M is accessed to calculate in step S73 at least one parameter set P.sub.calcd for the read-in TARGET value E.sub.targ. Optionally, the calculated parameter set P.sub.calcd can be output in step S74, for example on the UI interface.

[0143] In step S75, a workpiece can optionally be machined using the calculated parameter set P.sub.calcd. In step S76, an ACTUAL value for the machining result E.sub.act can be measured on the manufactured workpiece. In step S77, the comparison between the TARGET value E.sub.targ and the ACTUAL value E.sub.act for the machining result can be calculated. If there is a deviation, the calculated comparison value can be output in step S78. Alternatively or cumulatively, adaptation procedures can be triggered and executed on the basis of the calculated comparison value. In particular, the read-in TARGET value E.sub.targ can be adapted so that the measured ACTUAL value for the machining result E.sub.act can be maintained. In accordance with FIG. 6, the optional steps in FIG. 7 are also shown in broken lines.

[0144] The method has at least two application cases, each of which contains at least two sequentially executed algorithms. The algorithms of the first application case are the application calibration (algorithm 1) and the application calculation (algorithm 2). The algorithms of the second application case are the application simulation (algorithm 3) and the manufacturing forecasting (algorithm 4). The two application cases form the inverse of each other, which implies that the sequential execution of Algorithms 1 through 4 provides identity mapping and allows for self-checking of the digital twin.

[0145] The algorithms of the application cases are described in more detail below. The application calibration is the functional relationship between all the variables of the machining result (data structure 3) of the laser material machining (application) and the process characteristics (data structure 2). This functional relationship is based on the measured values acquired at the time after machining (ex situ). The application calibration can include empirical (statistical) or calculated (theoretical) functions.

[0146] The application calibration algorithm is shown in a flowchart in FIG. 8. This algorithm calculates the corresponding process characteristics PKG from the TARGET machining result in data structure 3, which was entered by the user, for example.

[0147] An application example for the application calibration algorithm will be described in more detail below with reference to FIG. 9.

[0148] Two laser beams, which are focused with different focus position in the workpiece are shown in FIG. 9. This results in a different inclination of the joint walls, which is designated with the average value of the angle of inclination <a.sub.f> of the cut surface. The angle of inclination is determined by the calibration function.

[00008]$\tan .Math.{\alpha }_f.Math.=\frac{u}{D-2\Delta a}$

[0149] The workpiece thickness D is specified therein, the shortening Δa of the joint height is defined in the standard, and the cutting edge profile height u is a value of the machining result (as the ACTUAL value of the associated measuring method, as the TARGET value of the user). In the model assumption, u is also the difference in the beam radii at the ends of the shortened joint height:

u=W(Δa)−W(D−Δa)

[0150] The beam radius function is given by

[00009]$W\left(z\right)=W_0.Math.\sqrt{1+{\left(\frac{z-z_0}{z_R}\right)}^2}$

with the focus radius W.sub.0 and the focal length Z.sub.R.

[0151] According to the application calibration flowchart, the following steps are now executed:

[0152] the user gives the desired cutting edge profile height u as a machining result (TARGET value) before (or u is read from the data structure 3 in the “automatic mode” as the current measured value for the application);

[0153] the application calibration algorithm calculates the process characteristic using the above calibration function a.sub.f;

[0154] the algorithm writes the calculated value of a.sub.f into the data structure 2 of the process characteristics.

[0155] The deterministic model provides a calibration function for each additional variable of the machining result, in the application calibration algorithm.

[0156] The algorithm for the application calculation is shown in FIG. 10 in a flowchart. Consequently, first the machining result variables calculated in the previously executed algorithm of the application calibration are loaded. Subsequently, the details of the cutting plan and workpiece must be entered. This is usually done by the user or by the upstream control unit in the automated manufacturing flow.

[0157] The parameter set P is calculated in the central procedure of the application calculation. In particular, the variable machining parameters (e.g., feed, power, pressure, focus position) are calculated because a large part of the data structure 2 with the parameter set P is already given by the specifications and the data from the minimum user input.

[0158] When the application calculation is complete, the data are displayed and saved, and the manufacturing is either started or changed. In a preferred development, the manufacturing or the current laser machining process can be controlled on the basis of the calculated values by changing the calculated parameters manually or automatically during machining. The change is made in response to the TARGET/ACTUAL comparison by accessing the model M.

[0159] The algorithm for the application calculation will be described below with reference to an application example: “Inclination angle of the cut surface” described in more detail.

[0160] In the section on the application calibration algorithm, it was shown as an example how the process characteristic af (average angle of inclination of the cut surface) is calculated from the machining result u (profile height as a measure of the flatness of the cut surface).

[0161] The application calculation now loads the calculated cone angle αf of the kerf and all other current process characteristics and calculates therefrom the parameter set P, in particular the variable machining parameters of the manufacturing. For the example of the average angle of inclination of the cut surface, the transcendental equation provides

[00010]$\sqrt{1+{\left(\frac{\Delta a-z_0}{z_R}\right)}^2}-\sqrt{1+{\left(\frac{D-\Delta a-z_0}{z_R}\right)}^2}-\frac{D-2\Delta a}{W_0}.Math.\tan .Math.{\alpha }_f.Math.=0$

the sought focal position z_0 as a variable of the parameter set. On the subsequently cut sample, u can be measured again (as E.sub.act) and checked.

[0162] The application case 2 of the method comprises the application simulation, followed by the manufacturing forecasting. The application case 2 is the inverse function of the application case 1, consisting of the application calibration, followed by the application calculation. The consequence of this is that application calibration and manufacturing forecasting are mutually inverse algorithms, as are the application calculation and application simulation.

[0163] The application simulation will be described in more detail below with reference to FIG. 11. In the application simulation, the data of the parameter set P are used to forecast the expected machining result E on the manufactured sample. First, the already fixed data, for example specifications, are loaded. Subsequently, the cutting plan (contour) and the workpiece (geometry, material) are entered: This is by definition the minimum user input. Finally, the variable machining parameters (data at the contour points), for example the power, the feed and the focus position, are entered or read in from a data structure.

[0164] If all data of the parameter set P are completely available, the application simulation calculates the process characteristics PKG therefrom, displays them and saves them.

[0165] The variable machining parameters of the parameter set P can be changed even if the manufacturing is already running. The manufacturing and the calculation of the process characteristics are then automatically updated each time.

[0166] Application example: “Inclination angle of the cut surface”

[0167] In the application calculation, the focal position z_0 had to be determined numerically from a transcendental equation. Here, in the application simulation, the inverse procedure is used, wherein the focus position is specified as a value in the data structure 1 of the parameter set P. From this, the inclination angle of the cut surface is calculated with

[00011]$\tan .Math.{\alpha }_f.Math.=\frac{W_0}{D-2\Delta a}.Math.\left(\sqrt{1+{\left(\frac{\Delta a-z_0}{z_R}\right)}^2}-\sqrt{1+{\left(\frac{D-\Delta a-z_0}{z_R}\right)}^2}\right)$

and stored as the value of the process characteristics PKG in the data structure 2.

[0168] The manufacturing forecasting is the inverse algorithm or the inverse function of the application calibration, and will be described below with reference to FIG. 12.

[0169] First, the process characteristics calculated with the application simulation are loaded. Then the functions of the algorithm for the manufacturing forecasting are applied; as a result of these calculations one obtains the calculated or forecast data for the (expected) machining result E.sub.targ. If, for example, it is decided in the course of an experiment or a random check that the manufacturing forecasting (and thus the application calibration) must be remeasured, the corresponding measuring methods are carried out, whereby the measured values of the machining result E.sub.act are obtained.

[0170] The comparison between the calculated and the measured machining result E.sub.act−E.sub.targ, starting from the same parameter set P, provides the correlation between ACTUAL and TARGET and allows the determination of the correction data for the two algorithms of the parameterisation.

[0171] Application example: Inclination angle of the cut surface

[0172] The process characteristic PKG of the inclination angle of was calculated and stored from the parameter set P. This value is now read by the manufacturing forecasting (together with other values of the process characteristics) and thus with:

u=(D−2Δa).Math.tana.sub.f

the profile height u is calculated. As a calculated value, this machining result dataset represents a nominal value, which can now be determined with the corresponding measurement.

[0173] The following describes how the profile height of the interface can be measured and fed as an input dataset into the algorithms described above.

[0174] With a surface measuring device, a section of a cut surface on the sample is recorded (preferably in a non-contact manner). With an optical detection device (e.g., CCD camera—macroscope or microscope with a variable focal plane, white light interferometer) an image of the sample is detected, in which the cut surface is shown, for example, of a structural steel of thickness 15 mm. The cut surface has the typical characteristics of a flame cut (periodic striation formation at the top, at the bottom wide and irregular striations with trailing effects). Above the sample, a reference surface with high flatness is clamped together to compensate for the clamping angle. Basically, the sample to be measured in the measuring space of the surface measuring device is clamped together with a reference surface in the sample holder. The flat reference surface is at right angles to the cutting surface the orientation of which can be attributed to the reference surface. The result of such a measurement can be displayed and output in the form of a false colour image.

[0175] The actual measuring range affects only one image section, so that only a smaller rectangular or polygonal section of the cut surface is considered. This section can be approximately 13 mm long and 8 mm wide, resulting in a Δa of approximately one millimetre. The averaged profile curve of the cut surface in the vertical direction is now recorded over this rectangular measuring range. The averaged profile curve is shown graphically in FIG. 13. In the profile curve, the minimum (dashed line) and the maximum (dotted line) can now be distinguished. The difference between maximum and minimum results in the profile height u as a measured variable for the machining result E.sub.act. The sign of u is positive if the depth of cut z (horizontal axis of the profile curve) of the minimum is smaller than that of the maximum.

[0176] After the profile height u has been determined, the measured value is written into the data structure 3 of the machining result E. In the preceding text sections on the application example of the cone angle of the kerf, it is described how this value can be used for comparison with the calculated or user-entered TARGET values.

[0177] Finally, it should be noted that the description of the invention and the exemplary embodiments are not to be understood as limiting in terms of a particular physical realisation of the invention. All of the features explained and shown in connection with individual embodiments of the invention can be provided in different combinations in the subject matter according to the invention to simultaneously realise their advantageous effects.

[0178] In particular, it will be apparent to a person skilled in the art that the invention can be applied not only to laser installations L for cutting, but also to other applications such as welding and/or labelling and/or engraving and/or drilling. Furthermore, the components of the terminal T, of the comparator and of the electronic processing unit V can be incorporated in a unit of hardware or can also be realised distributed over a plurality of physical products. In addition, the above-mentioned components can be arranged directly on the laser. Likewise, the parameterisation module ParM and/or the forecasting module ProgM can be formed locally on or at the laser. Alternatively, it can also be designed as a cloud system or on a server and be in data exchange with the laser L.

[0179] The scope of protection of the present invention is given by the following claims and is not limited by the features explained in the description or shown in the figures. Having described some aspects of the present disclosure in detail, it will be apparent that further modifications and variations are possible without departing from the scope of the disclosure. All matter contained in the above description and shown in the accompanying drawings shall be interpreted as illustrative and not in a limiting sense.