METHOD AND DEVICE FOR OPERATING A LASER MATERIAL PROCESSING MACHINE

Abstract

A computer-implemented method for operating a laser material processing machine. An estimated result is ascertained as a function of predefined process parameters, which characterize how good an actual result of the laser material processing will be, and the process parameters are varied by means of Bayesian optimization with the aid of a data-based model, until an actual result of the laser material processing is sufficient enough.

Claims

1. A computer-implemented method for operating a laser material processing machine, the method comprising the following steps: ascertaining an estimated result as a function of predefined process parameters, which characterize how good an actual result of the laser material processing will be; and varying the process parameters by means of Bayesian optimization using a data-based model, until an actual result of the laser material processing is sufficient enough.

2. The method as recited in claim 1, wherein the process parameters are varied until the estimated result is sufficient enough, and only then is the actual result of the laser material processing detected.

3. The method as recited in claim 2, wherein the data-based model is trained as a function of actual results.

4. The method as recited in claim 3, wherein the data-based model is also trained as a function of the estimated result.

5. The method as recited in claim 1, wherein the data-based model is a Gaussian process model.

6. The method as recited in claim 1, wherein the estimated result is ascertained using a physical model of the laser material processing.

7. The method as recited in claim 6, wherein if the evaluation of the physical model with parameters were to take place outside a predefinable range, the estimated result is then ascertained using the data-based model.

8. The method as recited in claim 6, wherein the estimated result is ascertained using a physical model evaluated with the predefined process parameters and using actual results ascertained with other process parameters.

9. The method as recited in claim 1, wherein the laser material processing machine is a laser drilling machine.

10. The method as recited in claim 9, wherein variables are used for characterizing the estimated result and/or for characterizing the actual result, which characterize a geometry of a hole drilled by the laser drilling machine.

11. The method as recited in claim 1, wherein the laser material processing machine is a laser welding machine.

12. The method as recited in claim 11, wherein variables are used for characterizing the estimated result and/or for characterizing the actual result, which characterize a geometry of a weld seam welded by the laser welding machine.

13. The method as recited in claim 1, wherein after setting the process parameters, the laser material processing machine is operated using the process parameters thus set.

14. A test stand for a laser material processing machine, the test stand configured to: ascertain an estimated result as a function of predefined process parameters, which characterize how good an actual result of the laser material processing will be; and vary the process parameters by means of Bayesian optimization using a data-based model, until an actual result of the laser material processing is sufficient enough.

15. A non-transitory machine-readable memory medium on which is stored a computer program for operating a laser material processing machine, the computer program, when executed by a computer, causing the computer to perform the following steps: ascertaining an estimated result as a function of predefined process parameters, which characterize how good an actual result of the laser material processing will be; and varying the process parameters by means of Bayesian optimization using a data-based model, until an actual result of the laser material processing is sufficient enough.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Specific embodiments of the present invention are explained in greater detail below with reference to the figures.

[0062] FIG. 1 schematically shows a structure of a laser drilling machine, in accordance with an example embodiment of the present invention.

[0063] FIG. 2 schematically shows a structure of a laser welding machine, in accordance with an example embodiment of the present invention.

[0064] FIG. 3 schematically shows a structure of a test stand, in accordance with an example embodiment of the present invention.

[0065] FIG. 4 shows in a flowchart one specific embodiment for operating the test stand, in accordance with an example embodiment of the present invention.

[0066] FIG. 5 shows in a flowchart one specific embodiment for operating the test stand, in accordance with an example embodiment of the present invention.

DETAILED DESCRIPTION OF EXAMPLE EMBODIMENTS

[0067] FIG. 1 schematically shows a structure of a laser drilling machine 1. An activation signal A is provided by an activation logic 40 in order to activate a laser 10a. The laser beam strikes a material piece 12, where it produces a drill hole 11.

[0068] FIG. 2 schematically shows a structure of a laser welding machine 2. Here, too, an activation signal A is provided by an activation logic 40 in order to activate a laser 10b. The laser beam strikes two material pieces 13, 14 where it produces a weld seam 15.

[0069] A laser cutting machine (not shown) is similarly also possible.

[0070] FIG. 3 schematically shows a structure of a test stand 3 for ascertaining optimum process parameters x. Instantaneous parameters x are provided by a parameter memory P via an output interface 4 of the laser material processing machine such as, for example, laser drilling machine 1 or laser welding machine 2. This carries out the laser material processing as a function of these provided process parameters x. Sensors 30 ascertain sensor variables S, which characterize the result of the laser material processing. These sensor variables S are provided as quality characteristics y.sub.exp to a machine learning block 60 via an input interface 50.

[0071] Machine learning block 60 in the exemplary embodiment includes a Gaussian process model which, as illustrated in FIG. 4 and in FIG. 5, is trained as a function of provided quality characteristics y.sub.exp. Varied process parameters which are stored in parameter memory P, may be provided as a function of the Gaussian process model.

[0072] Process parameters x may alternatively or additionally be provided for provision via output interface 4 also to an estimation model 5, which provides estimated quality characteristics y.sub.sim instead of actual quality characteristics y.sub.exp to machine learning block 60.

[0073] The test stand in the exemplary embodiment includes a processor 45, which is configured to run a computer program stored on a computer-readable memory medium 46. This computer program includes instructions, which prompt processor 45 to carry out the method illustrated in FIG. 4 and in FIG. 5 when the computer program is run. This computer program may be implemented in software, or in hardware, or in a mixed form of hardware and software.

[0074] FIG. 4 shows in a flowchart an exemplary method for operating test stand 3. The method starts 100 by providing initial process parameters x.sub.init as process parameters x and initializing previously recorded test data as an empty set. Process parameters x are optionally predefined using a design-of-experiment method and, as discussed in greater detail below, laser material processing machine 1, 2 activated using these process parameters x, ascertains variables y.sub.exp and trains the Gaussian process using the test data thus ascertained.

[0075] In the case of laser drilling, these parameters x in one exemplary embodiment include a pulse duration and/or a focal position time-dependently resolved via a characteristic map and/or a focal length and/or a pulse repetition frequency and/or a circular path diameter time-dependently resolved via a characteristic map and/or a circular path frequency and/or a setting angle time-dependently resolved via a characteristic map and/or a drilling duration and/or a pulse energy time-dependently resolved via a characteristic map and/or a wavelength and/or parameters, which characterize a process inert gas such as, for example, a process gas type or a process gas pressure). The aforementioned circular path in this case is a known feature in many drilling methods, for example, in twist drilling or in trepanning drilling.

[0076] In the case of laser welding, these process parameters x include laser power time-dependently and/or location-dependently resolved via characteristic maps and/or a focal diameter and/or a focal position and/or a welding speed and/or a laser beam inclination and/or a circular path frequency of a laser wobbling and/or parameters that characterize a process inert gas.

[0077] With instantaneous process parameters x, laser material processing machine 1, 2 is activated 110 and variables y.sub.exp ascertained 120, which characterize the actual result of the laser material processing.

[0078] In the case of laser drilling, these variables y.sub.exp in one exemplary embodiment include variables, which characterize the size of drill hole 11 and/or the circularity of drill hole 11 and/or the shape of a wall of drill hole 11 and/or the presence of melt deposits and/or a quantity of droplet ejection during the drilling process and/or a rounding of the edges of drill hole 11 and/or the productivity.

[0079] In the case of laser welding, these variables y.sub.exp in one further exemplary embodiment include variables, which characterize a minimum weld seam depth and/or a minimum weld seam width and/or the productivity and/or a number of weld spatters and/or a number of pores and/or a weld delay and/or weld-internal stresses and/or weld cracks along weld seam 15.

[0080] As a function of these variables, a cost function K is evaluated 130, as it may be provided, for example, by equation 1, the variables y.sub.exp being provided as features q.sub.i and corresponding target values of these variables a q.sub.i,target.

[0081] A cost function K is also possible, which penalizes deviations of the features from the target values, in particular, if they exceed a predefinable tolerance distance, and rewards a high productivity. The “penalizing” may, for example, be implemented by a high value of cost function K, the “rewarding” correspondingly by a low value.

[0082] It is then ascertained whether cost function K indicates that instantaneous process parameters x are sufficient enough; in the event a penalty resulting therefrom means a high value and a reward means a low value by checking whether cost function K falls below 140 a predefinable maximum cost value. If this is the case (“yes”), the method ends 150 with instantaneous process parameters x.

[0083] If this is not the case (“no”), data point x,y.sub.exp thus ascertained from process parameters x and associated variables y.sub.exp characterizing the result is added 160 to the ascertained test data and the Gaussian process is retrained, i.e., hyper-parameters Θ.sub.0,Θ.sub.d of the Gaussian process are adapted in such a way that a likelihood that the test data result from the Gaussian process is maximized.

[0084] An acquisition function is then 170 evaluated, as it is illustrated, for example, in formula 7, and new process parameters x′ are thereby ascertained. A branching back to step 110 then takes place.

[0085] FIG. 5 shows in a flowchart one further exemplary method for operating test stand 3. Steps 100 through 170 are the same as illustrated in FIG. 4, a separate description is therefore omitted.

[0086] After new process parameters x′ have been ascertained, however, a simulation model is called up 180 using these new process parameters x′ in order to ascertain estimated variables y.sub.sim instead of actual variables y.sub.exp.

[0087] In the case of laser drilling, this may take place, for example, as follows: for a radius r of drill hole 11 along a depth coordinate z, r(z) is numerically ascertained as a resolution of the equation

[00002]$\begin{array}{cc}\left[1-R\left(r,z,\alpha ,\theta \right)\right].Math.\cos \theta .Math.F_0\left(r,z\right)-{\tilde{F}}_{\mathrm{th}}=0\mathrm{where}& \left(8\right)\\ 1-R=\frac{1}{2}.Math.\left(\frac{4n\cos \theta }{\left(n^2+k^2\right)+2n\cos \theta +{\cos }^2\theta }+\frac{4n\cos \theta }{\left(n^2+k^2\right){\cos }^2\theta +2n\cos \theta +1}\right)& \left(9\right)\\ F_0\left(r,z\right)=\frac{2Q}{\pi w^2\left(z\right)}.Math.\exp \left(-\frac{2r^2}{w^2\left(z\right)}\right)& \left(10\right)\\ w\left(z\right)=\frac{d_{\mathrm{Fok}}}{2}\sqrt{1+{\left(\frac{z}{l_{\mathrm{Rayleigh}}}\right)}^2}& \left(11\right)\\ \tan \alpha =\frac{r}{w\left(z\right)}\frac{\mathrm{dw}\left(z\right)}{\mathrm{dz}}& \left(12\right)\end{array}$

[0088] In this case: [0089] n=n+ik is a predefinable complex refractive index of material piece 12, including refractive index n and extinction coefficient k [0090] {tilde over (F)}.sub.th is a predefinable ablation threshold fluence of material piece 12, [0091] Q is a predefinable pulse energy of laser 10a, [0092] d.sub.Fok is a predefinable focal diameter of laser 10a, [0093] l.sub.Rayleigh is a predefinable Rayleigh length of laser 10a, [0094] R is an ascertained reflectivity of material piece 12, [0095] α is an ascertained angle of the local beam propagation direction, [0096] θ is a predefinable relative angle between incident laser beam and surface normals of material piece 12, [0097] F.sub.0 is an ascertained irradiated fluence of laser 10a, [0098] w(z) is an ascertained local beam radius

[0099] The prediction of several features such as a presence of melt deposits and/or a quantity of droplet ejection during the drilling process is not possible with this model. To ascertain these features, either an empirical model may be predefined or a result may be ascertained from the values experimentally ascertained up to this point in time such as, for example, a mean value of all these values, or a weighting of these experimentally ascertained actual values may take place as a function of a separation of the instantaneous process parameters from those process parameters, for which the respective experimentally ascertained actual values have been determined. It is possible, in particular, that predictions of Gaussian processes that have been trained based on actual variables may be used as estimated values.

[0100] Alternatively or in addition, it is possible that at least some of the features are not able to be reliably calculated for all process parameters x. It is possible that it is checked whether instantaneous parameters x fall within a predefinable range, and that if this is not the case, the features are then ascertained with the aid of one of the aforementioned approaches.

[0101] In the case of laser welding, the ascertainment of estimated variables y.sub.sim may take place, for example, as follows:

[00003]$\begin{array}{cc}T\left(x,y,z\right)-T_0=\frac{1}{2\mathrm{\pi \lambda }h}\exp \left(-\frac{v\left(x-x_0\right)}{2a}\right)\left(q_{\mathrm{net}}{K}_0\left(\frac{\mathrm{vr}}{2a}\right)+2\underset{m=1}{\overset{\infty }{.Math.}}\cos \left(\frac{m\pi z}{h}\right)K_0\left(\frac{\mathrm{vr}}{2a}\sqrt{1+{\left(\frac{2m\pi a}{\mathrm{vh}}\right)}^2}\right)I_m\right)\mathrm{with}& \left(13\right)\\ r=\sqrt{\left(x-x_0\right)+y^2}& \left(14\right)\\ I_m={\int }_0^h{q}_{1\mathrm{net}}\left(z\right)\cos \left(\frac{m\pi z}{h}\right)\mathrm{dz}& \left(15\right)\end{array}$

[0102] and the parameters

[0103] T.sub.0—a predefinable ambient temperature

[0104] x.sub.0—a predefinable offset of the beam of laser 10b relative to the origin of a coordinate system movable with laser 10b

[0105] λ—a predefinable heat conductivity of material pieces 13, 14;

[0106] α—a predefinable temperature conductivity of material pieces 13, 14;

[0107] q.sub.net—a predefinable power of laser 10b;

[0108] q.sub.1net—a predefinable power distribution of laser 10b along a depth coordinate of material pieces 13, 14;

[0109] v—a predefinable speed of laser 10b;

[0110] h—a predefinable thickness of material pieces 13, 14;

[0111] and Bessel function

[00004]$K_0\left(\omega \right)=\frac{1}{2}{\int }_{-\infty }^{\infty }\frac{e^{i\omega t}}{\sqrt{t^2+1}}\mathrm{dt}$

as well as an ascertained temperature distribution T(x,y,z). A width and a depth of the weld seam may be ascertained from the temperature distribution (for example, via the ascertainment of isotherms at a melting temperature of one material of material pieces 13, 14).

[0112] Cost function K is subsequently 190 ascertained similar to step 130, simulatively estimated variables y.sub.sim being used instead of experimentally ascertained variables y.sub.exp.

[0113] It is then 200 checked similar to step 140 with the aid of cost function K whether or not instantaneous process parameters x are sufficient enough, instead of the predefinable maximum cost value, a second maximum cost value being capable of being used, which is greater than the predefinable maximum cost value.

[0114] If the check has indicated that instantaneous process parameters x are sufficient enough, then a branching back to step 110 takes place. Otherwise, a branching back to step 160 takes place.