阅读说明：本技术 用于监控工件的激光加工过程的方法 (Method for monitoring a laser machining process of a workpiece ) 是由 E·绍尔 E·苏佩尔诺克 于 2020-10-06 设计创作，主要内容包括：本发明涉及一种用于监控用于加工工件的激光加工过程的方法,所述方法包括：对于多个同类型的加工步骤,在一个加工步骤期间针对至少一个测量参量检测测量信号；在调设阶段中,基于在调设阶段期间检测的测量信号分别确定用于至少一个监控参数的极限值；并且在监控阶段中,基于每个加工步骤的测量信号求取用于该加工步骤的监控参数值,将所求取的监控参数值与极限值进行比较,用以识别错误的加工步骤,并且基于至少一个加工步骤的测量信号求取是否满足过程变化条件,其中,在预给定的调设间隔之后或者在预给定数量的在调设阶段期间实施的加工步骤之后,从调设阶段更换至监控阶段,并且其中,当满足过程变化条件时,从监控阶段更换回调设阶段。(The invention relates to a method for monitoring a laser machining process for machining a workpiece, comprising: for a plurality of processing steps of the same type, a measurement signal is detected for at least one measurement variable during a processing step; in a setting phase, a limit value for at least one monitoring parameter is respectively determined on the basis of the measurement signals detected during the setting phase; and in the monitoring phase, a monitoring parameter value for each processing step is determined on the basis of the measurement signal of the processing step, the determined monitoring parameter value is compared with a limit value in order to identify a faulty processing step, and a determination is made on the basis of the measurement signal of at least one processing step as to whether a process change condition is satisfied, wherein a change from the setting phase to the monitoring phase takes place after a predetermined setting interval or after a predetermined number of processing steps carried out during the setting phase, and wherein the setting phase is changed back from the monitoring phase when the process change condition is satisfied.)

1. A method for monitoring a laser machining process for machining a workpiece, the method comprising:

for a plurality of processing steps of the same type, a measurement signal is detected (501) for at least one measurement variable during a processing step;

in a set-up phase (101), determining (203) a limit value for at least one monitoring parameter on the basis of the measurement signal detected in the set-up phase; and is

In a monitoring phase (102), a monitoring parameter value for a processing step carried out in the monitoring phase is determined (301) on the basis of a measurement signal of the processing step, the determined monitoring parameter value is compared (302) with a determined limit value for identifying a faulty processing step, and a determination (303) is made on the basis of a measurement signal of at least one processing step whether a process variation condition is fulfilled,

wherein, after a predefined adjustment interval or after a predefined number of processing steps carried out during the adjustment phase, the process is changed (103) from the adjustment phase to the monitoring phase, wherein, when the process change condition is met, the process is changed (104) back from the monitoring phase.

2. The method according to claim 1, further comprising, in the set-up phase (101), determining (201), based on the measurement signals detected for the measurement quantities during the set-up phase (101), a mean value curve (401) for the measurement quantities and determining (202), based on the mean value curve (401), an envelope curve (402, 403) for the measurement quantities,

wherein the envelope curves (402, 403) comprise an upper envelope curve (402) and a lower envelope curve (403), the mean value curve (401) being located between the upper envelope curve (402) and the lower envelope curve (403).

3. Method according to claim 2, wherein in the set-up phase (101) limit values for the at least one monitoring parameter are determined based on the mean value curve (401) and/or based on the envelope curve (402, 403).

4. The method as claimed in one of the preceding claims, wherein it is ascertained (303) whether a process variation condition is fulfilled on the basis of a measurement signal of a single measurement variable detected in a single processing step, or on the basis of a measurement signal of a single measurement variable detected in a plurality of processing steps, or on the basis of a measurement signal of a plurality of measurement variables detected in a single processing step, or on the basis of a measurement signal of a plurality of measurement variables detected in a plurality of processing steps.

5. The method according to any of the preceding claims, wherein the process variation condition is fulfilled above a predefined maximum value for an outlier frequency, wherein the outlier frequency is defined as: the number of measured values of the measurement signal detected during one processing step that lie outside an envelope curve (402, 403) for the measurement signal, relative to the total number of measured values of the measurement signal detected during said processing step.

6. The method of any preceding claim, wherein the at least one monitoring parameter comprises one of:

a maximum outlier spacing (503), wherein the outlier spacing is defined as the maximum spacing between a measured value of the measurement signal detected during one processing step which lies outside the envelope curve (402, 403) and the closest envelope curve (402, 403),

integration of the measurement signal of a processing step over at least one region outside the envelope curve (402, 403),

the area of the measurement signal of a processing step between the region outside the envelope curve (402, 403) and the nearest envelope curve (402, 403),

the integration of the measurement signal of one process step,

the mean square error of the measured signal of a processing step and the mean curve (401), and

frequency of outliers.

7. Method according to one of the preceding claims, wherein a processing step is identified as an error when the ascertained monitoring parameter value is above a corresponding limit value.

8. The method as claimed in one of the preceding claims, wherein, in the monitoring phase (102), limit values for the at least one monitoring parameter are additionally adapted.

9. The method according to claim 8, wherein the limit values for the at least one monitoring parameter are adapted periodically after a predetermined number of machining steps or after a predetermined time interval has elapsed.

10. Method according to claim 8 or 9, wherein the adaptation of the limit value for the at least one monitoring parameter is carried out when a predefined number of successive machining steps is identified as erroneous or when a predefined error rate is exceeded.

11. The method according to any one of claims 8 to 10, wherein the matching of the limit value for the at least one monitoring parameter comprises narrowing the limit value or increasing the limit value.

12. Method according to one of claims 8 to 11, wherein the adaptation of the limit value for the at least one monitoring parameter is carried out on the basis of monitoring parameter values which are respectively determined for a predetermined number of preceding processing steps.

13. The method of any of claims 7 to 12, further comprising outputting an error when a processing step is identified as erroneous.

14. The method of any preceding claim, wherein the at least one measured parameter comprises one of: temperature, thermal radiation and intensity of at least one process emission,

wherein the process radiation comprises, in particular, laser light reflected by the workpiece, plasma radiation generated by the processing, light in the visible spectral range generated by the processing, and light in the infrared spectral range generated by the processing.

15. The method according to any of the preceding claims, wherein the laser machining process is or comprises a laser welding process or a laser cutting process.

Technical Field

The invention relates to a method for monitoring a laser machining process for machining a workpiece. The invention can include, in particular, a method for automatically setting monitoring parameters and/or for automatically identifying process changes.

Background

In a laser processing system for processing a workpiece by means of a laser beam, the laser beam emerging from one end of a laser source or a laser guide fiber is focused or focused by means of beam guide optics or focusing optics onto the workpiece to be processed. For example, the machining can include laser cutting, laser brazingOr laser fusion welding (schwei β en). The laser processing system can also be referred to as a "laser processing facility" or can be simply referred to as a "facility". The laser machining system can include a laser machining device, such as a laser machining head, such as a laser cutting head or a laser welding head. In particular, when laser welding or laser soldering workpieces, it is important to continuously monitor the welding or soldering process in order to ensure the quality of the process. This covers the identification of machining errors.

Continuous monitoring of the laser machining process is usually carried out In real time during the execution of the laser machining process and is therefore also referred to as Online-processes berwachung or Online-processes monitoring. During monitoring, measurement signals of various measurement variables of the laser machining process, such as the intensity of the process radiation (Prozesstrahlung) or the process radiation (Prozessenission), are detected and evaluated.

An evaluation is then carried out, in which the measurement signal is checked as follows: whether certain error conditions or error criteria are met. An error is output if one or more of the measurement signals satisfies a predetermined (festgelegt) error condition during a processing step which can include processing of the workpiece or processing of a portion or region of the workpiece. Depending on whether a processing error has occurred, the corresponding processed workpiece can be marked as "good" or "acceptable" (i.e. suitable for further processing or sale) or as "bad" or "poor" (i.e. waste).

Evaluation of the measurement signals is often complicated, since the course of the measured variables depends to a large extent on the materials used, the set laser power, the machining speed, the degree of contamination of the workpiece, etc. The setting and adaptation of error conditions is very costly and must be repeated in the event of a change in the laser machining process. Furthermore, the setting and matching of error conditions is usually performed manually.

Disclosure of Invention

The object of the present invention is therefore to specify a method for monitoring a laser machining process for machining a workpiece, in which method monitoring parameters are automatically set. The object of the present invention is therefore to provide a method for monitoring a laser machining process for machining a workpiece, in which method process changes are automatically detected in order to reset monitoring parameters. The object of the present invention is therefore to provide a method for monitoring a laser machining process for machining a workpiece, in which method operator intervention is avoided.

The object is achieved by the subject matter of the independent claims. Advantageous embodiments and further developments are the subject matter of the dependent claims.

According to one aspect of the disclosure, a method for monitoring a laser machining process for machining a workpiece, in particular a metal workpiece, is specified. The method comprises the following steps: for a plurality of processing steps, a measurement signal is detected for at least one measurement variable during one processing step. In the setting phase, a limit value for at least one monitoring parameter is determined on the basis of the measurement signal detected during the setting phase. In the monitoring phase, a monitoring parameter value for each processing step is determined on the basis of the measurement signal of the processing step, the determined value of the monitoring parameter is compared with a limit value for identifying a faulty (fehlerhaft) processing step, and whether a process change condition is fulfilled is determined on the basis of the measurement signal of at least one processing step. In this case, the process is changed from the setup phase to the monitoring phase after a predefined (first) setup interval or after a predefined (first) number of process steps carried out during the setup phase, and the setup phase is changed back from the monitoring phase when a process change condition is met.

In other words, in the commissioning phase, the assessment of the process ("good" or "bad") is preferably not active (aktiv). In the monitoring phase, it is sought whether process change conditions are met. If so, the system is replaced again into the adjustment phase. Preferably, the system remains in the monitoring phase until a process variation condition is met. In other words, in the case of a stable process (process variation conditions are not met), the system preferably remains in a monitoring phase (so-called "periodic phase"). Alternatively, in the case of a stable process in which process change conditions are not met, the system can be changed back into the set-up phase after a predefined (second) set-up interval or after a predefined (second) number of process steps carried out during the monitoring phase.

According to the method of the disclosure, a laser machining process having a plurality of identical or repeated machining steps is divided into two phases, namely a setup phase in which at least one limit value for a monitoring parameter is determined and a monitoring phase in which at least one machining step is monitored on the basis of the value of the monitoring parameter determined for this machining step. The monitoring of the laser machining process is only carried out during the monitoring phase. The change between the two phases can be carried out repeatedly according to established criteria. By means of the described method, a process change of the laser machining process can be automatically detected and automatically changed back into the setting phase in order to automatically adapt the limit values for the monitoring parameters on the basis of the changed laser machining process.

The laser machining process can include a plurality of repeated and/or identical and/or comparable machining steps. The processing of the workpiece can include one or more processing steps. In other words, the machining step can include machining of the workpiece or machining of a portion or area of the workpiece. In particular, the processing step can correspond to a weld or weld joint. Each measurement signal in turn corresponds to a processing step. The measurement signal can comprise a measurement value, which is a real number.

Both during the setting phase and during the monitoring phase, at least one measurement variable can be detected during the respective processing step. The detection of the measured variable can comprise the detection of a measurement signal or a change in a measurement signal of at least one measured variable during one processing step, i.e. preferably during each processing step, or during every nth processing step (for example during every second processing step or every third processing step, etc.). Each measurement signal can comprise a plurality of measurement values, which are each associated with the time of the respective processing step.

The at least one measured variable can be the temperature, the intensity of the laser light reflected by the workpiece, the intensity of the generated plasma radiation, the intensity of the light generated by the laser machining process in the visible spectral range and/or the intensity of the light generated by the laser machining process in the infrared spectral range.

In the setting phase, at least one limit value for at least one monitoring parameter is determined. The monitoring parameter can be based on only one measurement variable or can be based on a plurality of measurement variables. At least one limit value is determined (Festlegung or Bestimung) on the basis of a measurement signal detected for a measurement variable during a processing step carried out in a set-up phase, the monitoring parameter being based on the measurement variable. The determination is made, for example, on the basis of the course of the detected measurement signals and according to suitable statistical methods, for example, a boxplot analysis. The limit value can define a range of values. Preferably, the limit values for the at least one monitored parameter comprise an upper value and/or a lower value and/or a limit value range having a lower value and an upper value. The limit value can be a positive real number. The determination can also be referred to as "learning" (Einlernen) or "Teaching" (Teaching).

For example, in the setting phase, a mean value curve for the measured variable can be determined on the basis of the measured signal detected for the measured variable during the setting phase. Furthermore, an envelope curve for the measured variable (so-called "reference curve" or simply "reference") can be determined on the basis of the mean value curve. The envelope curves can include an upper envelope curve and a lower envelope curve, wherein the mean curve is located between the upper envelope curve and the lower envelope curve. The mean value curve and/or the envelope curve can be determined by means of statistical methods, for example by means of boxplot analysis. The mean value curve can comprise the mean value and/or the median value of the measurement signals detected for the measurement variable during the setting phase. For example, the measurement signals detected during the setup phase for the measurement variables can be superimposed, so that the measurement signals of the processing steps at the same processing point in time can be cancelled out (verrechen).

In the setting phase, the respective limit value for the at least one monitoring parameter can be determined on the basis of the mean value curve determined for the measured variable and/or on the basis of the envelope curve determined for the measured variable.

In particular, the upper envelope curve can be determined such that it has a predetermined first distance from the mean value curve, and the lower envelope curve can be determined such that it has a predetermined second distance from the mean value curve. The first and second pitches can be the same or different in magnitude.

The mean value curve and the envelope curve can be determined such that they lie within a predefined reliability range for the respective measured variable. The reliability range of the measured variable can be predefined, for example, by the user of the laser processing method or of the laser processing system. The reliability range of the measured variable can determine the following ranges for the respective measured variable: the measurement signal of the measurement variable must not leave this range during the processing of the workpiece by the laser processing process. If the measurement signal still leaves this reliability range, the laser machining process can be interrupted or ended.

After a predetermined set interval or after a predetermined number of machined workpieces or after a predetermined number of machining steps, in particular of the same type of machining steps, a change can be made from the set phase to the monitoring phase.

The processing steps are monitored in a monitoring phase in order to identify erroneous processing steps. The monitoring can be performed by: the value of the monitoring parameter for each carried out process step is determined on the basis of the measurement signal of the process step. Preferably, the ascertained monitoring parameter values are compared with corresponding defined limit values in order to detect whether the implemented process step is faulty. For example, a processing step can be identified as an error when the ascertained monitoring parameter value is higher than the corresponding limit value. Furthermore, when the processing step is identified as erroneous, the method can include outputting the error.

The at least one monitoring parameter can be the outlier spacing (Ausrei β erabstand), the integration, the area between the measurement signal lying outside the envelope curve and the nearest envelope curve, the integration of the measurement signal over time, the mean square error of the measurement signal and the mean value curve, or the outlier frequency. The outlier distance can be defined as the (maximum) distance between the measured values of the measured signal detected during one processing step, which lie outside the envelope curve, and the closest envelope curve. The integration may be a time integration or an integration of the number of measured values with respect to the measurement signal. The at least one monitored parameter may also be an area enclosed by the measurement signal and the reference value. That is, the at least one monitoring parameter can refer to a mean value curve and/or an envelope curve. The reference value may be a predetermined constant value, for example zero, or may correspond to a mean value curve. The values of the monitoring parameters can be determined for the respective processing steps.

In the monitoring phase, it is also determined whether a process change condition is fulfilled on the basis of the measurement signal of at least one of the implemented process steps. It can be ascertained after each implemented process step or regularly after a certain number of implemented process steps whether a process change condition is satisfied. If it is determined that the process variation condition is met, the monitoring phase is switched back into the setting phase in order to re-determine the limit values and/or the mean value curve and/or the envelope curve for the at least one monitoring parameter.

The determination of whether the process variation condition is fulfilled can be carried out on the basis of a measurement signal of a single measurement variable during a single implemented process step, or on the basis of a detected measurement signal of a single measurement variable during a plurality of implemented process steps, or on the basis of a detected measurement signal of a plurality of measurement variables during a plurality of implemented process steps.

The process variation condition can be satisfied, in particular, above a predefined maximum value for the outlier frequency. The outlier frequency can be defined as: the number of measured values of the measurement signal detected during a processing step which lie outside the envelope curve for the measurement signal is compared to the total number of measured values of the measurement signal detected during said processing step. This is the case, for example, when the signal-to-noise ratio of the detected measurement signal changes, when the workpiece to be machined is not clean, or when a batch change (Chargenwechsel) of the workpiece to be machined is performed. These situations can be referred to as "process variations" and can lead to a change in the measurement signal or the process of change of the measurement signal of one or more measurement variables. For example, the mean and/or median of the measured values can vary, or the variability of the measured values, i.e. the statistically discrete parameters, can vary. For example, the predefined maximum value for the outlier frequency can be in the range of 10-100%, preferably in the range of 50-90%.

In the monitoring phase, the limit values of at least one monitoring parameter can also be adapted. The limit values can be matched on the basis of the measurement signals detected for the previously occurring processing steps. In other words, depending on the value of the monitored parameter determined for the preceding processing step, the limit value can be selected smaller or larger, or the limit value range can be selected narrower or wider. The adaptation can be used to react to whether the monitored laser machining process is operating steadily or unstably.

For example, the limit values for the at least one monitoring parameter can be adapted periodically after a predetermined number of processing steps or a predetermined time interval.

The limit values for the at least one monitoring parameter can be adapted when a predefined number of successive machining steps is identified as faulty or when a predefined fault rate is exceeded. The error rate can be defined as: relative to the predefined total number of processing steps, the number of processing steps is: in the processing step, a limit value of at least one monitored parameter is exceeded. For example, the error rate can be given in percent (i.e., with respect to 100 process steps) or in parts per thousand (i.e., with respect to 1000 process steps). The maximum error rate can be between 0.1% and 100%. The predetermined number of consecutive faulty processing steps can be between 3 and 1000, preferably between 5 and 10.

The adaptation of the limit value for the at least one monitored parameter can comprise a reduction of the limit value or an increase of the limit value. Reducing the limit value can include decreasing the magnitude of the limit value, and increasing the limit value can include increasing the magnitude of the limit value. The limit values for the at least one monitoring parameter can be adapted on the basis of the measurement signals of a predetermined number of preceding processing steps. For example, the limit values for the at least one monitored parameter can be adapted on the basis of the measurement signals of the measurement variables detected during the last n implemented process steps, where n is a natural number. For example, a trend, a slope or an average or a median of the measured variables can be calculated on the basis of the measured signals detected during the last n implemented process steps. For example, the match to the threshold value can be a relative or absolute match to the threshold value.

The matching can include determining whether the laser machining process is operating steadily or unsteadily. Further, the matching can include matching a limit value based on the determination. That is, when the limit values of the monitored parameters are matched, it can be taken into account whether the laser machining process is stable or unstable.

According to a further aspect of the disclosure, a laser machining system for machining a workpiece by means of a laser beam is specified, which is provided for carrying out the above-described method.

Drawings

The present invention is described in detail below with reference to the accompanying drawings. Shown in the drawings are:

fig. 1 shows a flow diagram of a method for monitoring a laser machining process for machining a workpiece according to an embodiment of the disclosure;

FIG. 2 shows a flow diagram of a setup phase of a method according to an embodiment of the present disclosure;

FIG. 3 shows a flow diagram of a monitoring phase of a method according to an embodiment of the present disclosure;

FIG. 4A shows a schematic diagram of a mean curve and an envelope curve determined by a method according to an embodiment of the present disclosure; and

fig. 4B shows a schematic representation of a measurement signal detected for a measurement variable during a processing step and a monitored parameter according to an embodiment of the disclosure.

Detailed Description

In the following, the same reference numerals are used for the same and identically functioning elements, unless otherwise specified.

Fig. 1 shows a flow diagram of a method for monitoring a laser machining process according to an embodiment of the disclosure.

For example, the laser machining process can include a laser cutting process and/or a laser welding process. The method for monitoring is carried out during the laser machining process. During laser machining, a plurality of repeated machining steps are carried out in sequence. In this case, one machining step can be carried out per workpiece, or a plurality of machining steps can be carried out per workpiece.

As shown in fig. 1, the method comprises a setup phase 101 and a monitoring phase 102, wherein repeated processing steps are carried out continuously both in the setup phase and in the monitoring phase. After a predetermined set interval or after a predetermined number of processing steps carried out during the set phase 101, a change can be made from the set phase 101 into the monitoring phase 102 (arrow 103). As described below, when a process change condition is met (arrow 104), it is possible to change from the monitoring phase 102 into the commissioning phase 101.

Both in the setting phase 101 and in the monitoring phase 102, signal profiles or measurement signals of at least one measurement variable are detected or received during each processing step. During the machining of the workpiece, the measurement variables are continuously detected or measured in order to obtain corresponding measurement signals. Subsequently, for each processing step, the measurement signal of the measured variable is evaluated.

The measured variable may relate to the temperature, the wavelength range of the process radiation, the intensity of the plasma radiation generated by the process, the intensity of the process radiation (e.g. thermal radiation) in the infrared spectral range of the light, the intensity of the process radiation in the visible spectral range of the light, or the intensity of the scattered-back or reflected-back part of the laser processing beam. The measurement signal can comprise a plurality of measurement points or measurement values which are associated with different points in time of the processing step. In other words, the measurement signal is a set of measurement values received over the duration of one processing step. In general, in each receiving step, a measurement signal is received for each measurement variable.

FIG. 2 shows a flow diagram of a setup phase of a method according to an embodiment of the present disclosure.

In fig. 2, according to an embodiment, the tuning phase 101 comprises determining a mean value curve (step 201) and determining an envelope curve for each measured quantity (step 202). The determination of the mean value curve is based on the measurement signals detected in a plurality of processing steps during the setting phase 101 for the respective measurement variable, which is described below with reference to fig. 4A and 4B. Subsequently, an envelope curve is determined from the mean value curve.

Furthermore, the method comprises, in the commissioning phase 101, determining a limit value for each monitored parameter (step 203). In one example, for each of the implemented process steps, a measurement signal can be detected for one or more measurement variables. One or more monitoring parameters are defined for each of these measurement variables. On the basis of the measurement signals detected for a plurality of processing steps during the set-up phase 101, limit values are determined for each monitored parameter in the set-up phase 101. According to one specific embodiment, the respective limit value can be determined on the basis of a mean value curve and/or on the basis of an envelope curve. If, in the monitoring phase, the predetermined limit value of the predefined monitoring parameter is exceeded, the corresponding processing step is evaluated as an error. In the setting phase, at least no monitoring of the limit values or even no process monitoring takes place.

FIG. 3 shows a flow diagram of the monitoring phase 102 of a method according to an embodiment of the present disclosure. The aim of the monitoring phase is to automatically detect changes in the welding process (process changes) and to adapt the monitoring. The monitoring in the monitoring phase takes place parameterically as a function of predefined monitoring parameters, which are each defined as a function of one or more measured variables.

The monitoring phase 102 includes evaluating each monitored parameter for the processing step performed during the monitoring phase 102 (step 301). The value of the monitoring parameter is determined on the basis of the measurement signal of at least one measurement variable (which monitoring parameter is related to the at least one measurement variable) detected during the processing step. Subsequently, the ascertained monitoring parameter value for the processing step is compared with the limit value for this monitoring parameter determined in the setting phase 101 or the monitoring phase 102 (step 302). The comparison is used to identify the wrong processing step. According to one specific embodiment, the processing step is identified or marked as an error when the ascertained monitoring parameter value is higher than the corresponding limit value. Then, the error is preferably output. For example, the error can be output or displayed to a user of the laser machining process, or can be saved in an error memory. Furthermore, workpieces that have been machined in a machining step can be marked as "bad" or "wrong".

In parallel to steps 301 and 302, in the monitoring phase 102, it is determined whether a process variation condition is fulfilled on the basis of at least one measurement signal of at least one processing step carried out during the monitoring phase. The determination of whether a process-changing condition is fulfilled can be carried out, for example, on the basis of a detected measurement signal of a single measurement variable during a plurality of implemented process steps, or on the basis of a detected measurement signal of a plurality of measurement variables during a single process step, or on the basis of a detected measurement signal of a plurality of measurement variables during a plurality of implemented process steps.

For example, changes in the laser machining process (or simply "process changes") include changes in the material of the workpiece to be machined, changes in the degree of contamination of the workpiece to be machined, matching of laser power, batch changes of the workpiece to be machined, and the like. This results in a change of the measurement signal of the respective measurement variable. According to the invention, the detection of a process change therefore leads to the resetting phase being carried out again in order to retrieve at least one of the envelope curve, the mean value curve and the limit values for the monitoring parameters. For example, when the frequency of the abnormal value is higher than a predetermined value, the process variation condition can be satisfied. An outlier is defined here as a measured value of the measurement signal lying outside the envelope curve. For example, if more than 80% of the measurement signal lies outside the envelope curve, the monitoring phase is switched back into the setting phase.

The laser machining process can be variously stably operated even if there is no process variation. Preferably, the matching limit value can be generally set or adjusted according to: the laser machining process is a stable traversal stage or an unstable traversal stage. In the case of a stably operating laser machining process, the error conditions, in particular the limit values for the monitoring parameters, can be set narrower or stricter, whereas in the case of an unstably operating laser machining process, the error conditions can be set wider or more spacious (gro β ru gig) or more tolerable. The automatic adaptation of the limit values for the monitored parameters can be carried out on the basis of values of the monitored parameters which are determined for a plurality of preceding processing steps. For example, the limit value can be reduced if the monitored parameter value of the last, for example ten, processing steps is less than, for example, 80% of the limit value. In the case of a predetermined number of machining steps, for example after 10 machining steps, the adaptation of the limit values in the monitoring phase can be triggered (i.e. in the case of a mismatch of the mean value curve or the envelope curve). This can also be referred to as periodic matching. In addition, the adaptation of the limit values in the monitoring phase can be triggered when a predefined number of consecutive errors is exceeded, i.e. when a predefined number of successive processing steps that follow one another and are evaluated as being erroneous, or when a predefined error rate is exceeded, for example when more than 1% of the processing steps that are carried out are evaluated as being erroneous.

FIG. 4A shows a schematic of a mean curve and an envelope curve determined by a method according to an embodiment of the present disclosure.

As shown in fig. 4A, the diagram comprises measurement signals (dashed lines) of a measurement variable, which have been detected in each case in a processing step during the setting phase 101. That is, the measurement signals are detected for the same type of processing steps that are carried out in sequence in the setup phase and are superimposed in a graph. The duration of the processing step is denoted by tB. According to one specific embodiment, a mean value curve 401, an upper envelope curve 402 and a lower envelope curve 403 for the measured variables are determined on the basis of these measurement signals. According to one embodiment, the determination is made based on statistical methods of mean calculation (e.g., boxplot analysis, etc.).

As shown in fig. 4A, the envelope curves 402 and 403 can have the same spacing from the mean curve 401, but are not limited thereto. As shown in fig. 4A, the spacing of the envelope curves 402 is determined such that all measurement signals are located within the channel or window formed by the upper envelope curve 402 and the lower envelope curve 403. The envelope curves 402 and 403 can be determined directly on the basis of the measurement signals for the respective measurement variables. Alternatively, the envelope curves 402 and 403 are determined based on the mean value curve 401, for example, the upper envelope curve or the lower envelope curve can be determined by a deviation of + 20% or-20% from the mean value curve 401.

Fig. 4B shows a measured signal 501 which has been detected for a measured variable during a processing step in a monitoring phase and shows a plurality of monitored parameters according to an embodiment of the disclosure. Furthermore, fig. 4B shows a mean value curve 401 and envelope curves 402 and 403 determined during the set-up phase.

For example, the monitored parameters include outlier spacing. Here, according to an embodiment, the anomaly values represent measurement values lying outside the envelope curves 402, 403. Correspondingly, the outlier spacing is defined as the spacing between at least one measured value of the measurement signal 501 outside the envelope curves 402, 403 and the closest envelope curve 402, 403. As shown in fig. 4B, the measurement signal 501 has two regions 502a, 502B, in which the measurement signal 501 lies outside the envelope curves 402, 403, i.e. outside the window or corridor defined by the envelope curves 402, 403. According to an embodiment, the outlier spacing can be defined as the maximum spacing 503 of the measured values of the measurement signal 501 outside the envelope curves 402, 403 from the closest envelope curve 402, 403.

Furthermore, the monitoring parameter can comprise the area or the integral of the region 502b of the measurement signal 501 outside the envelope curves 402, 403, which belongs to the outlier spacing. The integration can involve an integration with respect to a region of the measurement signal 501 which lies outside the envelope curves 402, 403 and comprises anomalous measurement values. The area or the integral can relate to the sum of the magnitudes of a plurality of areas or the integral over a region 502a, 502b of the measurement signal 501 lying outside the envelope curves 402, 403. In the case of the measurement signal 501 shown in fig. 4B, the measured values are located in regions 502a, 502B outside the envelope curves 402, 403. The regions 502a, 502b are each assigned an area or an integral (shaded region) according to the above definition. The area or the integral can also be the area or the integral corresponding to the region 502b with the largest outlier spacing 503. In the case of the measurement signal 501 shown in fig. 4B, this corresponds to the area assigned to the region 502B, since this region contains the measured values with the largest outlier spacing 503.

Furthermore, monitoring the parameter can comprise integrating the measurement signal 501 over the duration tB of the processing step. For example, the integration of the measurement signal 501 can be the integration over time of the distance 505 of the measurement signal 501 from the reference value 504. As shown in fig. 4A, the reference value 504 can be a constant value, which can also be equal to zero, but is not limited thereto.

By the disclosed method for monitoring a laser machining process for machining a workpiece, monitoring parameters can be automatically reset or adapted according to process stability. In particular, the monitoring parameters can be automatically reset after a large process change has been detected by changing from the monitoring phase back into the setting phase. Furthermore, the limit values of the monitoring parameters in the monitoring phase can be adapted to minor process variations or to variations in process stability, i.e. whether a machining step based on the laser machining process is running more or less stably than a previously occurring machining step. Manual intervention and adaptation of the monitoring parameters is thus rendered superfluous, and the laser machining process can be carried out continuously in the sequence of a plurality of setting phases and monitoring phases.