Method for monitoring machining process in machine tool, and monitoring device and computer program therefor

Abstract

In a method for monitoring a machining process in a machine tool, a value of a measured variable is received, the value having been determined during a machining stroke during the machining of a workpiece. At least one test value based on the value of the measured variable is compared with a tolerance limit (Ui, Li). The tolerance limits are automatically determined by performing a statistical analysis on a plurality of reference values (Ri), which have been determined by measurements during the machining of a plurality of earlier workpieces, each reference value being based on one or more values of a measurement variable, which have been determined during the machining of one of the earlier workpieces.

Description

Technical Field

[0001] The invention relates to a method for monitoring a machining process in a machine tool. The machine tool can be a gear machining machine for machining toothed workpieces, in particular a gear grinding machine. Background Art

[0002] When machining workpieces in a machine tool, production deviations inevitably occur, which manifest themselves as deviations of the actual geometry of the workpiece produced in the actual process from the preset desired geometry. Production deviations can be caused by deviations from the set process occurring during the machining process. Process deviations can occur in particular through incorrect operation, malfunctions or wear of various components of the machine tool.

[0003] For reasons of time and cost, machining defects can usually only be checked on a sample basis on the machined workpieces. If the machining process is, for example, a grinding process for fine machining of the toothing, the toothing inspection of a single workpiece usually takes significantly longer than the actual machining. It is therefore uneconomical to inspect each workpiece individually. Machining defects are therefore usually not detected until after the workpiece has already been installed in the gearbox, during the so-called end-of-line test (EOL test). At this stage, defective workpieces have high cost consequences.

[0004] It is therefore desirable to detect process deviations as early as possible during machining ("on-line") or at least immediately thereafter ("in-line"), so that defectively produced workpieces can be removed in time and the process can be intervened in a corrective manner. For this purpose, it is known to continuously determine measured values ​​at the machine tool and compare them with tolerance limits. The measured values ​​can be, for example, the power consumption of the tool or workpiece spindle or the signal of a vibration sensor. If the tolerance range defined by the tolerance limits is left, this indicates a process deviation. In this case, the workpiece that has just been machined can be removed as a defective workpiece and, if necessary, the process can be intervened in a regulatory manner.

[0005] Determining the tolerance limits is a demanding task which requires a great deal of expertise. Determining the tolerance limits is also an iterative process which is prone to errors.

[0006] WO2022100972A2 proposes that two machine parameters are measured when grinding a toothed portion with the aid of a grinding tool. If at least one of the machine parameters exceeds or falls below a preset value under the condition of taking into account the tolerance band, a signal is output. At least one of the machine parameters contains a periodic signal component. These signal components are decomposed into individual frequency components by frequency analysis, and the frequency components are used for comparison in terms of their frequency and / or amplitude. In this case, individual limit values ​​can be determined for each frequency component.

[0007] The determination of individual tolerance limits for the individual frequency components requires a particularly large amount of expertise. For example, the operator who determines the tolerance limits must be able to classify the importance of the individual frequency components for possible machining defects. Machine tool operators often lack this expertise. Due to the large number of frequency components, determining such limit values ​​is also very time-consuming. The inevitable process discreteness additionally makes the task difficult. Usually, only after the production of thousands of workpieces can a complete picture of the discreteness of all frequency components be obtained. Therefore, in practice, the limit values ​​are often set incorrectly. For example, the limit value is set too wide, so that impermissible process deviations cannot be recognized at all, or the limit value is set too narrow, so that the workpiece is removed as a defective workpiece, although the workpiece actually meets the requirements for production accuracy.

[0008] WO2021048027A1 discloses a method for monitoring a machining process, wherein a plurality of measured values ​​are detected during machining engagement between a tool and a workpiece, including the value of a power index, which indicates the instantaneous power consumption of the tool spindle during machining. A normalization operation is applied to at least a portion of these measured values ​​or the values ​​of variables derived from the measured values ​​in order to obtain a normalized value. The normalization operation is associated with at least one of the following parameters: a geometric parameter of the tool, a geometric parameter of the workpiece, and a setting parameter of the machine tool. Thus, measured values ​​obtained with the aid of different process parameters can be compared with each other. It is also proposed in this document to perform a frequency analysis on the measured values. This document does not relate to the determination of tolerance limits.

[0009] WO 2020 / 193228 A1 discloses a method for automatic process monitoring in continuous generation grinding of a workpiece with a pre-machined tooth portion, which method enables early identification of grinding wheel fracture. During machining of the workpiece, at least one measured variable is monitored. A warning indicator for grinding wheel fracture is determined based on the measured variable. If the warning indicator indicates a grinding wheel fracture, the grinding wheel is automatically checked for grinding wheel fracture. This document also does not involve the determination of tolerance limits.

[0010] Against the background of ever-increasing demands on machining quality and correspondingly decreasing error tolerances, methods are desired which specifically detect even the smallest deviations or irregularities in the machining process. For example, in generating grinding, it is desirable not only to detect complete grinding wheel fractures but also to indicate grinding wheel microcracks. In other words, methods are needed for automatic process monitoring which are able to detect process deviations in a more objective and reliable manner than methods to date. Summary of the invention

[0011] In a first aspect, the object of the present invention is to specify a method for monitoring a machining process in a machine tool which allows objective and reliable identification of process deviations without requiring special expertise for determining tolerance limits.

[0012] This object is achieved by a method according to claim 1. Further embodiments are given in the dependent claims.

[0013] That is to say, a method for monitoring a machining process in a machine tool is proposed, in which a workpiece is machined by means of a tool in one or more machining strokes, the method comprising the following steps:

[0014] receiving a value of a measured variable (measured value), which value is determined by measurement at the machine tool during a machining stroke; and

[0015] At least one test value based on the value of the measured variable is compared to a tolerance limit.

[0016] According to the invention, the tolerance limits are determined by statistically analyzing a plurality of reference values ​​determined by measurements during the machining of a plurality of earlier workpieces, wherein each reference value is based on one or more values ​​of a measured variable determined during the machining of one of the earlier workpieces, and wherein when the reference values ​​are statistically analyzed, a degree of dispersion with respect to the reference values ​​is determined and the tolerance limits are determined based on the degree of dispersion.

[0017] In the proposed method, there are a large number of reference values ​​which were obtained during an earlier machining process by measurements at other workpieces, preferably on the same machine. The reference values ​​can be stored in a database and read from the database within the scope of the method. The term "reference values" should not mean that these values ​​are particularly reliable values. Rather, the term is only used to logically distinguish data obtained in an earlier machining process from current values ​​of measured variables that have been obtained in the machining process to be monitored or test values ​​obtained therefrom. The basic idea of ​​the invention is to make the values ​​of the same measured variables during an earlier machining process available for the assessment of the current machining process in that the tolerance limits for the current machining process are calculated based on a statistical analysis of the reference values ​​based on the earlier values.

[0018] This is based on the assumption that in practice, the vast majority of earlier measured values ​​or the reference values ​​derived therefrom are determined in production processes in which there are no inadmissible process deviations. In practice, only a small number of production processes are "bad" processes, since such "bad" processes are usually quickly identified as a result of production deviations and corrective measures are taken.

[0019] For example, in practice, a certain portion of all workpieces is usually measured after the machining process (for example, every hundredth workpiece), so that after a certain number of workpieces, an inadmissible process deviation becomes noticeable through production deviations of the relevant workpiece. If, for example, the workpiece is a toothed workpiece, the workpiece is also usually installed in a gearbox after the machining process, and at least a portion of the finished gearbox is tested on a so-called EOL (EOL=end of line) test bench. Here, too, workpieces with inadmissible production deviations are quickly identified, in particular through disturbing noise development in the gearbox. The reference values ​​obtained when machining workpieces with production deviations can then be marked accordingly in the database and thus excluded from the determination of the tolerance limits or even taken into account in a targeted manner when determining the tolerance limits.

[0020] Thus, in statistical averages across a large number of machining processes, the reference values ​​essentially represent a "good" machining process, i.e. a machining process without inadmissible process deviations, and the statistical distribution of the reference values ​​represents the typical distribution that should be expected in a "good" machining process. This knowledge is used to perform an automatic determination of the tolerance limits.

[0021] Tolerance limits are therefore automatically determined based on objective criteria without requiring the operator to have in-depth knowledge of the machining process.

[0022] The tolerance limits thus determined can be used in two ways during the machining process.

[0023] On the one hand, the machining process can be directly controlled based on the comparison of the test values ​​with the tolerance limits. In particular, the machining process can be interrupted or automatically modified if the comparison shows an impermissible process deviation.

[0024] On the other hand, the tolerance limits themselves are also a valuable interpretation aid for the operator. The operator can distinguish between "good" and "bad" test values ​​based on the tolerance limits. In other words, the operator can gain an objective understanding of the quality of the current machining process for himself. Ideally, this enables the operator to directly infer the cause of the production deviations determined.

[0025] The steps of receiving measured values ​​and comparing test values ​​based on the measured values ​​with tolerance limits are preferably performed iteratively, ie the steps are preferably repeated continuously in the course of the machining process.

[0026] The reference values ​​can be stored in a database. The method can then include calling up the reference values ​​to be statistically analyzed from the database. The measured values ​​of the current machining process can themselves be used again for a later machining process. For this purpose, new reference values ​​can be calculated based on these measured values ​​and the database can be updated by storing the new reference values ​​in the database. By continuously replenishing the database with new reference values, the method becomes self-learning to a certain extent and continuously improves itself through more and more reference values.

[0027] In particular, in the method according to the invention, each reference value can correspond to a test value, which is determined for an earlier workpiece based on a measured value measured during the machining of the earlier workpiece, that is, the reference value is formed directly by the test value of the earlier workpiece. However, it is also conceivable that the reference value is derived from the measured value of the earlier machining process in a different way. In particular, the test value compared with the tolerance limit can be directly the original measured value, and a plurality of measured values ​​determined successively in time can be compared with the same tolerance limit. In contrast, the reference value for determining the tolerance limit can be formed by an average value (or another position parameter), a maximum value, a minimum value or other values ​​characterizing the earlier machining operation. It is sufficient to store these reference values ​​in a database. As a result, less storage space is required in the database than if all earlier measured values ​​are stored in the database and called up each time when calculating the tolerance limit.

[0028] Here, the reference values ​​for each combination of workpiece geometry and tool type can be stored in separate areas of the database, that is, the reference values ​​in a specific area of ​​the database are always specific to a specific workpiece geometry and a specific tool type. Then, for each of these areas of the database, information about the workpiece geometry and about the tool type can be additionally stored in the database. If necessary, the reference values ​​can also be specific to a specific tool geometry. In the case of a trimmable tool, the tool becomes smaller after each trimming process. In this case, it is conceivable that the reference values ​​obtained from the machining process that takes place after a specific trimming process are stored in separate areas of the database respectively. In this case, information about the tool dimensions obtained from this trimming process can be additionally stored in the database. Then, after each trimming process, the tolerance limits can be re-determined, wherein only the following reference values ​​are used for calculating the tolerance limits, which are determined using tools with the same or similar tool dimensions after trimming. Here, a "similar" tool dimension is understood as a dimension that is located in a preset band around the current dimension of the tool. However, it is also conceivable to take into account the reduced tool dimensions by performing a normalization operation, as described below. A combination of these two measures is also possible.

[0029] As already mentioned, the workpiece can be, for example, a toothed workpiece, in particular a gear, in particular a spur gear, a bevel gear or any other rotationally symmetrical workpiece with a regular sequence of teeth and tooth gaps. The machining process can in particular be a grinding method, in particular a gear grinding method, such as generating grinding or profile grinding. However, the present invention is not limited to the machining or grinding methods of toothed workpieces. Therefore, the present invention can be used, for example, in the scope of tooth machining, also in gear hobbing, in gear skiving or gear honing. The machining process can also be a dressing process, in which the tool is a dressing tool (in particular a rotating dressing wheel or a rotating modified gear) and the workpiece is a grinding tool (in particular a grinding worm or a profile grinding wheel).

[0030] The measured variable can be variable over time, so that in the method according to the invention, time-dependent values ​​of the measured variable for a plurality of time points during a machining stroke are received. These time-dependent values ​​can then be treated in different ways.

[0031] In a first particularly simple variant, a time-independent test value is calculated from the time-dependent value of the measured variable, which test value represents the value of the measured variable during the entire machining stroke. Each reference value is likewise time-independent and is based on the time-dependent value of the measured variable in a machining stroke when machining an earlier workpiece. At least one time-independent tolerance limit is determined by statistical analysis of the time-independent reference values, and the time-independent test value is compared with the at least one time-independent tolerance limit.

[0032] In a second, more complex variant, a plurality of time intervals of the machining stroke are preset, wherein the time intervals can all have the same length or different lengths and do not necessarily have to be directly adjacent to one another. For each of the time intervals, at least one associated test value is determined, which is based on the value of the measured variable determined during machining in the time interval in question. For example, the test value can be the value of the measured variable in the middle of the time interval in question, the average value of the measured variable in the time interval in question (or another position parameter in terms of descriptive statistics) or any other value characterizing the behavior of the measured variable in the time interval in question. At least one tolerance limit is determined for each of the time intervals. This is done based on reference values ​​that are specifically for the time interval in question, that is, each of the reference values ​​is based on the time-related value of the measured variable during machining of one of the earlier workpieces in the time interval in question. Then, at least one test value for each of the time intervals in the time interval is compared with at least one tolerance limit for the time interval in question.

[0033] In another variant, a frequency analysis of the time-dependent values ​​of the measured variable is performed in order to determine a plurality of frequency components of the measured variable. A plurality of frequency intervals are preset, wherein the frequency intervals can all have the same size or different sizes and do not necessarily have to be directly adjacent to each other. For each frequency interval in the frequency interval, at least one associated test value is determined, the test value being based on the frequency components in the frequency interval involved. The test value can be, for example, a selected frequency component in the frequency interval involved, a maximum value or integral of the frequency components in the frequency interval involved, or any other value characterizing the frequency content of the measured variable in the time interval involved. At least one tolerance limit is determined for each of the frequency intervals. This is based on reference values ​​specifically for the frequency interval involved, i.e., each reference value is based on a frequency component in the frequency interval involved, the frequency component being determined by a frequency analysis of the time-dependent values ​​of the measured variable during one of the earlier workpieces being processed. Then, at least one test value for each frequency interval in the frequency interval is compared with at least one tolerance limit for the frequency interval involved.

[0034] In some embodiments, the workpiece is processed successively in at least two processing strokes. It is then advantageous that the tolerance limits are specific to the respective processing strokes, ie different tolerance limits are individually determined for each processing stroke.

[0035] Preferably, the reference value is based on a measured value determined during machining of an earlier workpiece in the same machining stroke on the same machine. However, it is also conceivable to use a reference value determined during machining of an earlier workpiece in other machining strokes and / or on other machines of the same type.

[0036] The statistical analysis of the reference values ​​can be performed in different ways. The degree of dispersion can be, for example, a standard deviation or the size of a value interval in which a predetermined proportion of all reference values ​​lies (e.g., an interquartile range). In some embodiments, the tolerance limits can then be set relative to the position parameters of the reference values ​​(e.g., relative to the arithmetic mean or median) as a multiple of the standard deviation or a multiple of the size of the mentioned value interval. In other embodiments, the limits of the mentioned value interval can directly form the tolerance limits, for example as the limits of the value interval in which the lower 99% or 99.9% of all values ​​lie.

[0037] In order to reduce the sensitivity of the method to the specificity of a particular statistical analysis method, it can be provided that the tolerance limits are determined by combining at least two statistical analysis methods. Alternatively or additionally, it can be provided that at least one value of the measured variable or a test value derived therefrom is compared with at least two tolerance limits determined by different statistical analysis methods.

[0038] The time-dependent measured variable can be, for example, at least one of the following variables or can be derived from at least one of the following variables:

[0039] a power index, the power index being a measure indicative of the instantaneous power consumption of a tool spindle or a workpiece spindle of a machine tool; or

[0040] A vibration indicator is determined by means of at least one vibration sensor and represents a vibration of the machine tool.

[0041] In order to reduce the dependency of the test values ​​on process parameters, such as tool diameter, workpiece diameter and model or feed, it can be provided that a normalization operation is performed when determining the test values ​​in order to normalize the test values. The normalization operation is here dependent on at least one process parameter, wherein the process parameter is at least one geometric parameter of the tool, at least one geometric parameter of the workpiece and / or at least one setting parameter of the machine tool. The normalization operation is here performed such that the normalized test value is less dependent on the at least one process parameter than without the normalization operation. Such a normalization operation is particularly valuable if the measured variable is a measure that indicates the power consumption of a tool spindle or a workpiece spindle.

[0042] If at least one of the mentioned process parameters changes, the normalization function is preferably recalculated. The recalculation of the normalization function can in particular include the use of a model which describes the expected dependence of the reference value on the process parameter, in particular a model of the process force or process power.

[0043] For further considerations regarding normalization operations, refer to publication WO2021048027A1, the contents of which are fully incorporated into the present disclosure by reference.

[0044] The method can include outputting user information to a user of the machine tool, wherein the user information is based on a comparison of at least one test value with a tolerance limit. For example, the result of the comparison can be displayed visually, for example on a display of a machine control of the machine tool or on a display of a mobile terminal device, such as a laptop or tablet computer, wherein the mobile terminal device does not necessarily need to be arranged at the same location as the machine tool and / or an acoustic output can be performed. The visual output can be performed in a graphical manner. Of course, there are countless other possibilities for outputting the user information.

[0045] Alternatively or additionally, the method can include influencing the machining process as a function of the result of the comparison of the test value with the tolerance limit. In particular, it can be provided that the machining process is stopped when the comparison shows that an impermissible process deviation exists, or that the workpiece is automatically removed when an impermissible process deviation has been determined during machining of the workpiece.

[0046] Based on the comparison of at least one test value with the tolerance limits, a numerical process deviation indicator can be determined. In the simplest case, the process deviation indicator is a Boolean variable, which indicates by one of its two possible values: the presence of an impermissible process deviation (e.g., TRUE=impermissible process deviation, FALSE=no impermissible process deviation). However, the process deviation indicator can also be a more complex indicator, such as an array of variables with Boolean, integer or real values, each of which indicates the degree of deviation of the test variable from the associated tolerance limit. The process deviation indicator or user information based thereon can be output.

[0047] The method can also provide for an automatic check of the machine state. This can be done temporarily if an impermissible process deviation is detected, or the machine state can be checked independently of the actual monitoring of the machining process at regular intervals, for example during machining pauses. To check the machine state, the method can include:

[0048] performing a machine test cycle in which at least a portion of the machine axis is purposefully operated and status data associated with the operation is determined by measurement; and

[0049] performing a state diagnosis, wherein the state data are compared with at least one reference state variable in order to determine at least one machine state indicator,

[0050] In this case, a fault source indicator is determined as a function of the process deviation indicator and the machine state indicator, which contains information about the type of fault source present for the inadmissible process deviation.

[0051] The method does not necessarily have to include the execution of a machine test cycle and a condition diagnosis simultaneously. It can also be provided that a machine condition indicator determined in an earlier condition diagnosis is read from a database.

[0052] For example, the fault source indicator can indicate which machine axis is affected and / or whether it may be a machine fault (eg due to a malfunctioning machine axis), a process fault (eg due to incorrect clamping of a workpiece) or an operator fault.

[0053] The inspection of the mechanical state can be carried out here by means of a method as described in the application CH070373 / 2021 (patent number CH 718264) of the applicant of the present application on October 11, 2021. The contents of the application CH 070373 / 2021 or the patent number CH718264 are fully incorporated into the present disclosure by reference.

[0054] The determination of the fault source indicator is also advantageous if the tolerance limits, on the basis of which the process deviation indicator is determined, are set in a manner other than by statistical analysis of reference values. In particular, the determination of the fault source indicator is also advantageous if the tolerance limits were previously set manually. In any case, the fault source indicator is a very strong interpretation aid with the help of which the operator can quickly identify the suspected fault source even without in-depth professional knowledge. The advantage of this approach is that information from two completely different sources is combined, namely on the one hand from the monitoring of the machining process (i.e. from the process diagnosis) and on the other hand from the inspection of the machine state (i.e. from the state diagnosis). This combination of information means that it is not possible to provide a process diagnosis or a state diagnosis separately. Only by linking the process diagnosis and the state diagnosis to each other can new insights be generated, which make it easier for the operator to identify the fault source.

[0055] In another aspect, the invention provides a monitoring device for monitoring a machining process in a machine tool, wherein a workpiece is machined by means of a tool. The monitoring device is configured so that it performs the above method. For this purpose, the monitoring device can have a computer configured to perform the method. The computer can be implemented locally at a single physical location, distributedly at multiple physical locations or in the cloud. The computer can have a non-volatile memory device in which a computer program is stored, which, when it is run, causes the computer to perform the method mentioned.

[0056] The monitoring device can have one or more of the following means, wherein these means can be implemented by a computer program executed by the aforementioned computer:

[0057] a database interface configured to read reference values ​​from a database and, if necessary, to transmit new reference values ​​to the database;

[0058] limit determination means configured to perform a statistical analysis on the reference values ​​in order to determine tolerance limits;

[0059] a measured value interface which is configured to receive the value of a measured variable, for example by reading the value of the measured variable from a memory device of the machine control device or by directly reading a sensor for determining the measured variable;

[0060] a test value determination device configured to determine a test value based on a value of a measured variable;

[0061] a comparison device configured to compare the test value with the tolerance limit; and

[0062] A user interface configured to output user information, for example in the form of process deviation indicators or fault source indicators.

[0063] These devices can be implemented in different physical units. For example, the calculation of the tolerance limits can be carried out in the cloud, i.e. the database interface and the limit determination device can be implemented by a service in the cloud. The reception of the measured values, the determination of the test values, the comparison with the tolerance limits and the output of the user information can be carried out locally in the machine control of the machine tool. A data interface can then be used for data exchange between the service in the cloud and the machine, via which in particular the tolerance limits can be transmitted to the machine and the measured values ​​and / or reference values ​​can be transmitted back to the database interface via the data interface. The data interface can be implemented wirelessly or wired.

[0064] Furthermore, the monitoring device can have a user interface configured to change at least one parameter used by the computer for automatically setting tolerance limits, such as a factor by which the fluctuation measure of the reference value is multiplied when setting the tolerance limits. The user interface can also be configured to change the automatically set tolerance limits. For example, the user interface can be implemented locally on a control panel of the machine tool or also separately on a mobile device such as a laptop or tablet, for example via a touch screen.

[0065] The present invention also provides a machine tool having the above-mentioned type of monitoring device. The machine tool can also have at least one of the following devices:

[0066] a tool spindle for driving the tool in rotation about a tool spindle axis;

[0067] a workpiece spindle for driving the workpiece in rotation about a workpiece spindle axis;

[0068] a movement device configured to move the tool spindle and the workpiece spindle relative to one another in order to perform a machining stroke;

[0069] At least one sensor for determining the value of a measured variable during a machining stroke.

[0070] In particular, the sensor can be a power sensor for determining a measure that indicates the power consumption of a tool spindle or workpiece spindle or a vibration sensor for determining a measure that indicates vibrations of a machine tool.

[0071] In another aspect, the invention provides a computer program comprising commands which, when run by a computer of a monitoring device, in particular a monitoring device as defined above, cause the computer to perform the above method. The computer program can be stored on a non-volatile storage medium.

[0072] The above-described embodiments with regard to the method according to the invention also apply to the device according to the invention and the computer program according to the invention to the same extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] The preferred embodiments of the present invention are described below based on the accompanying drawings, which are only used for explanation and should not be understood as limiting. In the accompanying drawings:

[0074] Figure 1 A schematic diagram showing a generating grinder;

[0075] Figure 2 A diagrammatic representation is shown for illustrating the time profile of a measured variable during a machining stroke;

[0076] Figure 3 A diagram illustrating the statistical distribution of the maximum values ​​of a measured variable for a plurality of machining processes of different workpieces on the same generating grinding machine is shown;

[0077] Figure 4 A diagram illustrating a possible statistical distribution of reference values ​​for a plurality of machining processes of different workpieces on the same generating grinding machine is shown;

[0078] Figure 5 A diagram showing another statistical distribution of reference values ​​for a plurality of machining processes of different workpieces on the same generating grinding machine;

[0079] Figure 6 A diagram illustrating the time profile of a measured variable during a machining process and its evaluation in the time domain is shown;

[0080] Figure 7 shows a spectrum as obtained by frequency analysis of the time variation curve of the measured variable;

[0081] Figure 8 showing a graph illustrating a comparison of test values ​​obtained from frequency components of the measured variable with frequency-dependent tolerance limits;

[0082] Fig. 9 A simplified diagram showing a network with a plurality of generating grinders of the same type communicating with a database via a service-server;

[0083] Fig. 10A shows a flow chart for determining tolerance limits based on reference values;

[0084] Fig. 10B A flow chart for monitoring a machining process using tolerance limits is shown;

[0085] Fig. 10CA flow chart for determining a fault source indicator is shown;

[0086] Fig. 10D A flow chart showing updating a database with new reference values;

[0087] Fig.11 shows an illustrative example of a user interface for modifying automatically calculated tolerance limits;

[0088] Fig.12 A schematic example of a user interface for outputting information about a machining process is shown. DETAILED DESCRIPTION

[0089] Exemplary Configuration of Generating Grinding Machine

[0090] exist Figure 1 , a generating grinding machine 1 is shown by way of example as an example of a machine tool, which will also be referred to as a "machine" in the following text. The machine 1 has a machine base 11 on which a tool carrier 12 is guided movably in a radial feed direction X. The tool carrier 12 carries an axial slide 13, which is guided movably relative to the tool carrier 12 in a feed direction Z. A grinding head 14 is mounted on the axial slide 13, which can be pivoted about a pivot axis (the so-called A axis) extending parallel to the X direction in order to match the inclination angle of the tooth to be machined. The grinding head 14 in turn carries a displacement slide on which a tool spindle 15 is movable relative to the grinding head 14 in a displacement direction Y. A worm-shaped grinding wheel (grinding worm) 16 is clamped on the tool spindle 15. The grinding worm 16 is driven by the tool spindle 15 to rotate about the tool axis B.

[0091] The machine base 11 also carries a pivotable workpiece carrier 20 in the form of a turret, which can be pivoted about a pivot axis C3 between at least three positions. Two identical workpiece spindles are mounted diagonally opposite one another on the workpiece carrier 20, of which one is located at the bottom. Figure 1 Only one workpiece spindle 21 with an associated tailstock 22 can be seen in FIG. A workpiece can be clamped on each workpiece spindle and driven to rotate about a workpiece axis C1 or C2. Figure 1 The workpiece spindle 21 visible in the figure is in a machining position in which a workpiece 23 clamped thereon can be machined by means of the grinding worm 16. Another workpiece spindle 21, which is arranged offset by 180° and is Figure 1 The workpiece spindle, which is not visible in the drawing, is in a workpiece changing position, in which a finished workpiece can be removed from the spindle and a new blank can be clamped in. The dressing device 30 is mounted offset by 90° to the workpiece spindle.

[0092] The machine 1 thus has a plurality of movable parts, such as slides or spindles, which are movable in a manner controlled by corresponding drives. These drives are often referred to in the industry as "NC axes", "machine axes" or simply "axes". In part, this name also includes the parts driven by the drives, such as slides or spindles.

[0093] The machine 1 also has a plurality of sensors. Figure 1 Only two sensors 18 and 19 are schematically indicated in the figure. Sensor 18 is a vibration sensor for detecting vibrations of the housing of the grinding spindle 15. Sensor 19 is a position sensor for detecting the position of the axial slide 13 relative to the tool carrier 12 along the Z direction. In addition, the machine 1 also includes a plurality of other sensors. These sensors include, in particular, other position sensors for detecting the actual position of each linear axis, rotation angle sensors for detecting the rotational position of each rotational axis, current sensors for detecting the drive current of each axis, and other vibration sensors for detecting vibrations of each driven component.

[0094] All driven axes of the machine 1 are digitally controlled by a machine control 40. The machine control 40 comprises a plurality of axis models 41, a control computer 42 and an operator panel 43. The control computer 42 receives operator commands from the operator panel 43 and sensor signals from different sensors of the machine 1 and calculates control commands for the axis models 41 accordingly. The control computer also outputs operating parameters to the operator panel 43 for display. The axis models 41 provide control signals at their outputs for the machine axes respectively.

[0095] The monitoring device 44 is connected to the control computer 42 .

[0096] The monitoring device 44 can be a separate hardware unit associated with the machine 1. The monitoring device can be connected to the control computer 42 via an interface known per se, for example via the known Profinet standard, or via a network, for example via the Internet. The monitoring device can be physically part of the machine 1, or it can also be arranged physically remote from the machine 1.

[0097] During operation of the machine, the monitoring device 44 receives a plurality of different measurement data from the control computer 42. The measurement data received by the control computer include sensor data detected directly by the control computer 42 and data read by the control computer 42 from the axis model 41, such as data describing the expected positions of the various machine axes and the expected current consumption in the axis model.

[0098] The monitoring device 44 can optionally have its own analog and / or digital sensor inputs to receive sensor data directly from other sensors as measurement data. Other sensors are usually sensors that do not need to be used directly to control the actual machining process, such as acceleration sensors for detecting vibrations, or temperature sensors.

[0099] Alternatively, the monitoring device 44 can also be implemented as a software component of the machine control device 40, which is executed, for example, on a processor of the control computer 42, or the control device can be designed as a software component of a service server 45 described in detail below. Figure 1 , the processor 451 and the memory device 452 of the service-server 45 are indicated correspondingly.

[0100] The monitoring device 44 communicates with a service server 45 directly or via the Internet and a web server 47. The service server 45 in turn communicates with a database server 46 having a database DB. These servers can be arranged remotely from the machine 1. The server does not need to be the only physical unit. In particular, the server can be implemented as a virtual unit in a so-called "cloud".

[0101] The service server 45 communicates with a mobile terminal device 48 via a web server 47. In particular, the terminal device 48 can execute a web browser, by means of which the received data and its evaluation can be visualized. The terminal device does not need to meet any special requirements for computing power. For example, the terminal device can be a desktop computer, a laptop, a tablet computer, a mobile phone, etc.

[0102] Processing of batches of workpieces

[0103] For the sake of completeness, it is described below how a workpiece is generally machined with the aid of the machine 1 .

[0104] In order to process a workpiece (blank) that has not yet been processed, the workpiece is clamped to the workpiece spindle in the workpiece change position by an automatic workpiece changer. The workpiece change takes place simultaneously with the processing of another workpiece on another workpiece spindle in the processing position. When the new workpiece to be processed is clamped and the processing of the other workpiece is completed, the workpiece carrier 20 is rotated 180° about the C3 axis so that the spindle with the new workpiece to be processed reaches the processing position. Before and / or during the pivoting process, a centering operation is performed with the aid of an associated centering probe. For this purpose, the workpiece spindle 21 is rotated and the orientation of the tooth gap of the workpiece 23 is measured with the aid of a centering probe 24. On this basis, the rolling angle is set.

[0105] When the workpiece spindle carrying the workpiece 23 to be machined has reached the machining position, the workpiece 23 is brought into collision-free engagement with the grinding worm 16 by moving the tool carrier 12 along the X axis. The workpiece 23 is now machined by the grinding worm 16 in rolling engagement. This is carried out by one or more machining strokes, for example one or more roughing strokes, followed by one or more finishing strokes. Optionally, one or more polishing strokes can follow. During each machining stroke, the grinding worm 16 is continuously moved forward along the Z axis relative to the workpiece 23 in a constant or variable radial X feed (so-called axial stroke). At the same time, the tool spindle 15 is slowly and continuously moved along the displacement axis Y in order to continuously use the unused areas of the grinding worm 16 in machining (so-called displacement movement). Between two machining strokes, a so-called displacement jump may occur, which causes: in the next machining stroke, an area of ​​the grinding worm that is not directly connected to the previous area is used. Usually, the radial X feed is different between the machining strokes. As a result, the machining forces are also different between the individual machining strokes.

[0106] In parallel with the workpiece processing, the completely processed workpiece is removed from the other workpiece spindle and another blank is clamped on the spindle.

[0107] If, after machining a certain number of workpieces, the use of the grinding worm 16 has progressed to the point where the grinding worm is too blunt and / or the flank geometry is too inaccurate, the grinding worm is dressed. To this end, the workpiece carrier 20 is pivoted by ±90° so that the dressing device 30 reaches a position in which it is opposite the grinding worm 16. The grinding worm 16 is now dressed with the dressing tool 33.

[0108] Monitoring of measured variables

[0109] exist Figure 2 , the time profile of a time-dependent measured variable S(t) during a machining stroke of approximately 2 seconds duration is shown schematically in arbitrary units (au). The measured variable can be, for example, the current consumption of the tool spindle 15 or the signal of the vibration sensor 18. The exact time profile of the measured variable during the machining stroke is strongly dependent on the type of measured variable; in this respect, Figure 2 It should be understood as an example only.

[0110] The measured variable is detected digitally by the monitoring device 44. For this purpose, the measured variable is scanned (sampled) in a known manner with a preset sampling frequency and digitized with the aid of an analog-to-digital converter (ADC). This results in a sequence of discrete sampled values ​​of the measured variable for successive points in time. These digitized values ​​of the measured variable are referred to below as measured values.

[0111] exist Figure 2In the example of , the measured value fluctuates over time. During the machining stroke, the measured value has a maximum value S max and the minimum value S min The arithmetic mean of the measured values ​​during the machining stroke is S avg Variable S max , S min and S avg is an example of a time-independent test value which is derived from the value of the time-dependent measured variable. Alternatively or in addition, other time-independent values ​​can also be determined as test values, for example a measure which indicates fluctuations of the measured value during the machining stroke. As will be explained in more detail below, separate test values ​​can also be determined for different time intervals of the machining stroke, or the test values ​​can be determined based on frequency components which are determined by a frequency analysis of the time profile of the measured value.

[0112] The test values ​​determined in this way often fluctuate between workpieces. Figure 3 The successive workpiece numbers n are plotted along the horizontal axis and the corresponding test values ​​S are plotted along the vertical axis. i (n). The test value S i (n) For each workpiece and the upper tolerance limit U i And the lower limit of tolerance L i In this example, the test value S for almost all workpieces i (n) are all located in the tolerance band between these two tolerance limits, except for the values ​​for workpieces with workpiece numbers n=34 and n=44. The deviations from the tolerance band indicate inadmissible process deviations when machining these workpieces. The corresponding workpieces can be removed from the process accordingly and measures can be taken to bring the process back within the tolerance limits.

[0113] Automatic setting of tolerance limits

[0114] Tolerance limit U i , L i The reference value S is determined automatically by statistical analysis of a plurality of values ​​of a reference variable (reference values). These reference values ​​are determined by measurements during the machining of a plurality of earlier workpieces. Each reference value is based on the value of a measured variable determined during an earlier machining process on the same machine in the same machining stroke at one of the earlier workpieces. In particular, a test value S can be used as a reference value. i , which test values ​​are determined during the relevant machining stroke when machining a corresponding earlier workpiece.

[0115] This is based on the consideration that, on average, in a large number of machining processes on a large number of workpieces, the vast majority of the machining processes will not have inadmissible process deviations, since otherwise inadmissible process deviations would sooner or later be recognized based on production deviations of the workpieces machined in this way. If the number of workpieces for which reference values ​​have been determined is sufficiently large (e.g., greater than 1,000 or even greater than 10,000), then the distribution of the reference values ​​over the entirety of the workpieces closely approximates the expected distribution for a fault-free production process. Therefore, by means of a statistical analysis of these values, the tolerance limits can be set automatically based on objective criteria.

[0116] This is exemplified by Figure 4 Explain, in Figure 4 The value of the reference variable (reference value) R in arbitrary units (au) is shown in i The empirical frequent distribution of . Plot the reference value R on the horizontal axis i , the relative frequency for value intervals ("bins") of equal size is plotted on the vertical axis as a bar graph. In the present case, the frequency distribution corresponds approximately to a Gaussian normal distribution, the density function of which is also plotted in the dotted line Figure 4 For this frequent distribution, the arithmetic mean μ of the reference values ​​can be calculated i and empirical variance (defined as the reference value and the arithmetic mean μ i The mean square error) and the empirical standard deviation σ i (defined as the square root of the empirical variance).

[0117] Now it is possible to automatically set tolerance limits based on these statistical variables. For example, in this example, the upper tolerance limit is set to U i =μ i +z i σ i , the lower tolerance limit is set to L i =μ i -z i σ i Here, the factor z i is a freely selectable positive real number that indicates the difference between the tolerance limit and the mean value μ i According to the known 6σ design (which is however usually used for another purpose), for example z can be chosen i = 6. Depending on the customer's tolerance sensitivity, another factor z can also be selected i . Factor z i This can optionally be specified by an operator.

[0118] exist Figure 4 In the example, the reference variable R iThe values ​​of are approximately normally distributed. However, in many cases this is not the case. For example, in Figure 5 The reference value R is shown in i The empirical frequent distribution of deviates greatly from the normal distribution. Therefore, Figure 5 The frequent distribution of is bimodal and very asymmetric. This distribution is only inadequately characterized by the arithmetic mean and standard deviation. In these cases, other statistical analysis methods may be more promising. Two such methods are described below.

[0119] Instead of the arithmetic mean, the median Q(0.5) of the reference values ​​(i.e. the p-quantile Q(p) for p=50%) can be calculated, for example, as a more robust form of the location parameter. Instead of the standard deviation, which can be strongly influenced by individual "outliers", for example the interquartile range IQR of the reference values ​​can be calculated as a more robust measure of fluctuations, i.e. the value interval in which the middle 50% of all values ​​of the reference value lie, or in other words the distance Q(0.75)-Q(0.25) (also called the upper and lower quartile) between the p-quantiles of the frequent distribution for p=75% and p=25%.

[0120] The term “p-quantile” is understood here as follows, where it is common in descriptive statistics for sample quantiles to have as a minimum value a predetermined proportion p of all values ​​of the sample lies below said minimum value, where p is a real number between 0 and 1 and is referred to as the off-target proportion.

[0121] Correspondingly, the tolerance limit can be set according to the p-quantile of the distribution of the reference value.

[0122] In particular, the tolerance limits can be set based on the median and the interquartile range. Thus, for example, the upper and lower tolerance limits can be set as follows:

[0123] U i =Q(0.5)+z i (Q(0.75)-Q(0.25)),

[0124] L i =Q(0.5-z i (Q(0.75)-Q(0.25)),

[0125] The factor z is again a positive real number that can be freely chosen.

[0126] Based on the median and its distance from the upper and lower quantiles, alternative settings can be made as follows:

[0127] U i =Q(0.5)+z i(Q(0.75)-Q(0.5)),

[0128] L i =Q(0.5-z i (Q(0.5)-Q(0.25)),

[0129] A different determination can be made based on the upper and lower quantiles alone as follows:

[0130] U i =Q(0.75)+z i (Q(0.75)-Q(0.25)),

[0131] L i =Q(0.25-z i (Q(0.75)-Q(0.25)),

[0132] In this case, no location parameter such as the median is needed.

[0133] In a simpler embodiment, the preset quantile of the frequent distribution of the reference value can be used directly as a tolerance limit, for example. For example, the 99%-quantile Q (0.99) (also commonly referred to as the "last percentile") can be set as the upper tolerance limit, and the 1%-quantile Q (0.01) ("first percentile") can be set as the lower tolerance limit. In this case, the miss-target share p can be preset by the operator, and the corresponding quantile is determined by the miss-target share.

[0134] When determining the tolerance limits, a plurality of statistical methods can also be combined. For example, the upper tolerance limit can be defined as a weighted average of the upper tolerance limits calculated using two different methods.

[0135] Alternatively, two or more upper and / or lower tolerance limits can also be defined for the test value, wherein the tolerance limits are determined by different statistical methods. i =μ i +z i σ i The first tolerance limit calculated is compared with the second tolerance limit calculated as U i =Q(0.75)+z i The calculated second upper tolerance limit (Q(0.75)-Q(0.25)) is compared. If only one of the two upper tolerance limits is exceeded, an inadmissible process deviation can be inferred.

[0136] Implemented with the help of database

[0137] The reference values ​​can be predetermined and stored in a database 46. The tolerance limits are then automatically set by the monitoring device 44 in that it accesses the database 46, reads the reference values ​​from it and statistically evaluates them.

[0138] After a workpiece has been successfully machined, the monitoring device 44 can calculate new values ​​for the reference variables from the measured values ​​of the workpiece and store them in the database 46. In this way, the database is continuously replenished with new reference values, which are then available for setting tolerance limits for subsequent workpieces. As a result, the monitoring device becomes self-learning to a certain extent.

[0139] Time-varying tolerance limits

[0140] exist Figure 3 In the example of , a time-independent test value is determined for the time-dependent measured variable and compared with an upper tolerance limit and a lower tolerance limit. Alternatively or in addition, it is also possible to compare a test value determined specifically for a specific time interval during a machining stroke with the tolerance limits. In this case, these tolerance limits themselves can be time-dependent, i.e. different tolerance limits can be set for different time intervals.

[0141] This is Figure 6 In the description. Figure 6 The solid line shows the time-dependent course of the measured variable S(t) for a specific workpiece (in this example, a workpiece with workpiece number n=12). In this example, the measured value first increases during the machining stroke, then reaches a plateau, and then falls again. The power consumption of the tool spindle, for example, can show this course if the tool first gradually engages with the workpiece, machines the workpiece when fully engaged, and then gradually disengages again. Usually, this course is also superimposed with noise components, which, however, are not shown for the sake of clarity. Figure 6 Shown in.

[0142] Also in Figure 6 The upper tolerance limit U i And the lower limit of tolerance L i , the upper and lower tolerance limits are set separately for a plurality of time intervals i. In the present example, there are a total of 8 such time intervals with associated upper and lower tolerance limits. During the machining stroke, the measured variable S(t) is detected digitally and the test value S derived from the value of the measured variable is i These tolerance limits are constantly compared to be able to define inadmissible process deviations. Figure 6 In the example of FIG. 1 , the corresponding value of the measured variable S(t) in the middle of the time interval in question is simply used as the test value S i .

[0143] Figure 6 The circle in the figure shows: Test value S in interval i i What kind of fluctuations are experienced between the workpieces. In this example, the values ​​are mostly within the associated tolerance limits U i and L i Only for one of the workpieces (here the workpiece with workpiece number n=87), the value of the measured variable S(t) during the time interval i=2 exceeds the upper tolerance limit U i The complete curve of the measured variable S(t) for this workpiece, drawn with a dashed line, shows that the measured variable increases unexpectedly quickly. Exceeding the upper tolerance limit indicates an impermissible process deviation. Accordingly, the workpiece in question can be removed and inspected in detail or discarded, and if necessary, the machine status can be checked or the process can be adjusted.

[0144] Tolerance limit U i and L i can be determined automatically by statistical analysis. For this purpose, the values ​​of the measured variable S(t) determined earlier in the processing are used again. For each time interval i, a characteristic value of the measured variable S(t) is determined as a reference variable, for example the average value of the digitized values ​​of the measured variable S(t) during the time interval, or the value of the measured variable S(t) in the middle of the time interval. The distribution of these reference values ​​is now described in conjunction with the above Figures 3 to 5 The statistical analysis is performed in a similar manner as described. In particular, the statistical characteristic variables for these reference values, in particular the position parameter and the degree of dispersion, can be determined again, and the tolerance limits can be calculated therefrom. In this way, for each of the time intervals i, a separate upper tolerance limit U is determined. i and a separate lower tolerance limit L i .

[0145] In a further development, it is also conceivable to determine a test value for each time interval, which test value characterizes the behavior of the measured variable S(t) in this time interval in a different way than only by the current value in the middle of the time interval. For example, the average value of the measured values ​​can be determined as the test value for each time interval, or a regression analysis can be performed on the measured values ​​in order to determine a test value that characterizes the time variation of the measured values ​​in the relevant interval. For example, the slope of the measured values ​​in the relevant time interval can be determined as the test value. The tolerance limits can then be automatically determined for this test value by statistically analyzing the test values ​​calculated accordingly during the machining of an earlier workpiece as reference variables.

[0146] Frequency-dependent tolerance limits

[0147] For time-dependent measured variables, the comparison with the tolerance limits can also be carried out in the frequency domain instead of in the time domain. For this purpose, the digitized time-dependent values ​​of the measured variable are first subjected to a frequency analysis in order to determine a plurality of frequency components of the measured variable. This can be done by applying a suitable transformation, in particular a discrete Fourier transformation (DFT), which can be implemented in particular as a fast Fourier transformation (FFT), to the time-dependent values ​​of the measured variable. However, other methods can also be used for the frequency analysis, for example wavelet analysis.

[0148] The results of this frequency analysis are Figure 7 Schematically illustrated in FIG. Figure 7 The spectrum of the measured variable in the form of a plurality of frequency components (spectral values) T(f) as a function of the frequency f is shown. The spectrum is obtained by filtering and DFT of the time-dependent measured variable. A plurality of peaks can be identified in the spectrum, which are characterized by a circular shape. The frequency components in the range of these peaks are of particular interest for subsequent analysis.

[0149] Different frequency intervals are now preset for the spectrum. For each frequency interval, at least one test value is compared with one or more frequency-dependent tolerance limits. Figure 8 In this figure, the upper tolerance limit U is shown for each of the multiple frequency intervals i. i (solid line) and lower tolerance limit L i (dashed line). In addition, for each frequency interval, a plurality of test values ​​are shown as circles, which are determined from the frequency components of the measured variable in the corresponding frequency interval according to measurements at different workpieces. For example, the corresponding test value can be the integral or maximum value of the frequency components in the frequency interval in question.

[0150] Therefore, the frequency interval can be selected narrowly, for example, so that there is just one peak in each frequency interval, and the intensity of the peak involved can be used as a test value. Then, the associated tolerance limit defines a reliable range in which the intensity is allowed to move. The intensity of the peak can be determined by integrating the spectrum in the frequency interval involved or as the maximum value of the frequency component in the frequency interval involved. However, the frequency interval can also be selected to be wider, so that there are multiple peaks in the frequency interval. Correspondingly, the test value can also be defined more complexly. The frequency intervals do not necessarily have to be directly adjacent to each other. For example, it is feasible to compare only the intensity of the selected peak with the tolerance limit, such as the peak at a specific multiple of the workpiece or tool rotation speed. Therefore, the intensity of the peak at a certain multiple of these frequencies allows direct inference of a specific type of process deviation, which will be discussed in more detail below.

[0151] exist Figure 8In the example of , for each frequency interval, the integral of the frequency components in the frequency interval involved is selected as the test value. In this example, most of the test values ​​are in the range between the associated tolerance limits. For the frequency interval i=6, only the test value of the workpiece with workpiece number n=27 is lower than the lower tolerance limit L6, while for the frequency interval i=14, the test value of the workpiece with workpiece number n=51 exceeds the upper tolerance limit U 14 This again indicates a certain process deviation. In some cases, the frequency interval in which the deviation occurs even allows direct inferences about the type of process deviation.

[0152] Each upper tolerance limit U i and / or lower tolerance limit L i Again, it can be determined by statistical analysis of reference values ​​which were determined for an earlier machining process. For example, the relevant test value which has been determined for an earlier workpiece for the relevant time interval can be used as a reference value.

[0153] By continuously determining new test values ​​and comparing them with the tolerance limits, the comparison with the tolerance limits can be performed repeatedly (cyclically) for each machining process. For example, a frequency analysis can be performed continuously during machining and the resulting frequency components or the test values ​​obtained therefrom can be compared continuously with the tolerance limits.

[0154] Normalization operation

[0155] One difficulty in process monitoring, especially in gear machining, is the fact that the monitored measured variables are related in a highly complex manner to a number of geometrical properties of the tool (for example, diameter, form, number of threads, lead angle, etc. in the case of grinding worms), geometrical properties of the workpiece (for example, form, number of teeth, helix angle, etc.) and setting parameters at the machine (for example, radial feed, axial feed, speed of tool and workpiece spindle, etc.). Due to these diverse and complex dependencies, on the one hand, it is extremely demanding to draw direct inferences from the monitored measured variables to specific process deviations and the resulting machining defects. On the other hand, it is extremely difficult to compare measured variables from different machining processes with one another. Additional challenges arise when using dressable tools. Through dressing, the diameter of the tool changes during the machining of a series of workpieces, and thus the machining conditions also change. Therefore, even if all other framework conditions remain unchanged, the monitored measured variables from different dressing cycles within the same series of workpieces cannot be directly compared.

[0156] In order to compensate for differences in the machining conditions of different workpieces, a normalization operation is applied to the measured values ​​or the test values ​​determined therefrom. The normalization operation takes into account the influence of one or more process parameters on the measured values ​​or test values, in particular the influence of geometrical parameters of the finishing tool (in particular its dimensions, in particular its outer diameter), geometrical parameters of the workpiece and / or setting parameters of the finishing machine (in particular radial feed, axial feed and rotational speed of the tool and workpiece spindle). As a result, the normalized test values ​​generated are independent of the mentioned process parameters or at least significantly less dependent than without normalization. Due to the normalization operation, the normalized values ​​between different machining processes are comparable even if these process parameters are different. In particular, there is no need to define tolerance limits related to the process parameters.

[0157] The normalization operation is preferably based on a model that describes the expected correlation of the measured variable with the mentioned parameters. If the measured variable is a power indicator, the model preferably describes the correlation of the process power (i.e. the mechanical power or electrical power required for the machining process performed) with the mentioned parameters. The model of the process power can be based in particular on a force model that describes the expected correlation of the cutting force acting at the location of the contact between the finishing tool and the workpiece with the geometric parameters of the finishing tool, the geometric parameters of the workpiece and the setting parameters of the finishing machine. The model of the process power can also take into account the length of the lever arm that acts between the tool axis and the contact point between the finishing tool and the workpiece. The lever arm length can be approximated in particular by the outer diameter of the finishing tool. In addition, the model of the process power can take into account the rotational speed of the tool spindle.

[0158] The normalization operation can, for example, include multiplying the detected measured values ​​or variables derived therefrom by a normalization factor. However, more complex normalization operations are also conceivable. If the measured values ​​include the value of a power indicator, the normalization factor can in particular be an inverse power variable or a variable derived therefrom, which is calculated based on a model of the process power for the specific existing processing conditions.

[0159] The normalization operation is preferably applied directly to the detected values ​​of the measured variable, if necessary after filtering. The normalization operation is advantageously performed in real time, i.e. still during the machining process, in particular still during the machining of the corresponding workpiece, i.e. still during the machining engagement of the tool and the workpiece. The normalized values ​​are thus provided directly during the machining process and can be used in real time to monitor the machining process.

[0160] The normalization operation can be recalculated each time at least one of the process parameters changes. The recalculation of the normalization operation then preferably comprises applying the mentioned model with the changed process parameters.

[0161] For further considerations on the normalization operation, refer to publication WO2021048027A1, the contents of which are fully incorporated into the present disclosure by reference.

[0162] Perform status diagnostics in the event of impermissible process deviations

[0163] If a process deviation is determined, it can be sensible to investigate the cause of the process deviation. For this purpose, an automatic diagnosis of the machine state can be carried out, or already existing data determined in such a diagnosis can be used. In order to diagnose the machine state, a test cycle is carried out in which at least one part of the machine axis is operated in a targeted manner and condition data associated with this operation are determined by measurement. Based on these status data, a status diagnosis can then be carried out, wherein the status data are compared with at least one reference status variable in order to determine at least one machine status indicator. Based on the process deviation indicator and the machine status indicator, a fault source indicator can then be determined, which indicates, for example, whether a machine error, a pre-processing error or an operating error is present.

[0164] For example, in gear machining, a determined process deviation can be that the intensity of the frequency component of the drive power of the tool spindle in the rotational speed of the workpiece spindle exceeds the upper tolerance limit. This process deviation can have various causes. Thus, one cause can be an inadmissible total pitch error of the workpiece blank due to faulty pre-machining. However, the process deviation can also be the result of an imbalance due to faulty workpiece clamping or the result of a faulty workpiece spindle.

[0165] In order to investigate the cause of such a process deviation, a diagnosis of the workpiece spindle with the workpiece clamped thereon but not in machining engagement with the workpiece can now be performed. If this does not yield anomalies, it allows the conclusion that the process deviation is the result of a pre-machining defect of the workpiece. Otherwise, a condition diagnosis of the workpiece spindle can then be performed without the workpiece being clamped on the workpiece spindle. If this does not yield anomalies, it can be concluded that the process deviation is the result of a clamping error of the workpiece. Otherwise, it can be concluded that the process deviation is caused by a defective workpiece spindle.

[0166] In this way, the operator, even without in-depth specialist knowledge of his own, directly receives instructions which allow him to assess the machining process and the machine status in a differentiated manner.

[0167] This approach is also advantageous if the tolerance limits are set in a different way than described above, for example if the tolerance limits are set purely manually.

[0168] Implemented in the cloud

[0169] The calculation and monitoring of the tolerance limits can be performed locally in a monitoring device directly associated with the machine tool. However, it is also conceivable to perform at least part of the method in the cloud. An example is Fig. 9 In the description.

[0170] The machine 1 to be monitored and a plurality of further machines 2, 3, ..., N are connected to a service server 45 and a database 46 via a web server 47. The service server 45 and the database 46 are located in the cloud.

[0171] Each of these machines has a monitoring device which continuously transmits certain data to a database 46 during operation of the respective machine. These data include in particular a machine-to-machine identification, a time stamp and a plurality of test values ​​as described above. The data can optionally also include other data, such as data on measurements performed on the workpiece immediately following production, such as indicators for the achieved workpiece quality.

[0172] These data are stored in a database DB. Thus, over time, the database contains a very large amount of process data, which are obtained for a number of machines in many different machining processes. These data can be accessed for future machining processes. For example, the stored test values ​​can be used as reference values ​​when determining tolerance limits for future machining processes.

[0173] The results of the monitoring can be retrieved and visualized in a decentralized manner from any location. For this purpose, a web server 47 is used, which communicates with a separate mobile terminal 48, such as a tablet computer.

[0174] flow chart

[0175] exist Figures 10A to 10D A flow chart is shown in , which approximately summarizes the above-described method in a graphical manner.

[0176] Here, Fig. 10A The steps for determining the tolerance limits are described. In step 101, the monitoring device reads reference values ​​from a database. In step 102, the monitoring device performs a statistical analysis on the reference values ​​to determine the tolerance limits. In step 103, the monitoring device stores the tolerance limits in a memory device of the monitoring device, whereby the tolerance limits can be used later.

[0177] Fig. 10BThe steps for determining the deviation indicator are described. In step 111, the value of the measured variable is determined by measurement. In step 112, the monitoring device receives these measured values. In optional step 113, the monitoring device determines the associated process parameters, for example by reading them from the machine control device, and applies a normalization operation that takes into account these process parameters. In optional step 114, the monitoring device performs a frequency analysis. In step 115, the monitoring device calculates a test value based on the (optionally normalized) measured values ​​or their frequency components. In step 116, the monitoring device compares the test value with the previously determined tolerance limits and determines the deviation indicator. In step 117, the monitoring device outputs the deviation indicator to the machine control device or the user interface. During the processing of the workpiece, steps 111 to 117 are repeated periodically.

[0178] Fig. 10C The steps for determining the fault source indicator are described. In step 121, the monitoring device performs a test cycle. In step 122, the monitoring device performs a state diagnosis based on the measurement results of the test cycle to determine the machine state indicator. Alternatively, the monitoring device reads the machine state indicator that has been determined in an earlier state diagnosis from a database. In step 123, the monitoring device compares the machine state indicator with a previously determined process deviation indicator. In step 124, the monitoring device outputs the fault source indicator.

[0179] Fig. 10D It is described how the measured values ​​determined during the monitoring of the current machining process can be used to determine and store new reference variables. In step 131, the machining process is monitored. This is done as in Fig. 10B In the process of the machining process, the values ​​of the measured variables (measured values) are continuously determined. In step 132, new reference values ​​are calculated from the measured values. In step 133, the reference values ​​are stored in a database from which the reference values ​​are retrieved. Fig. 10A In this way, the database is supplemented with new reference values ​​during each machining process.

[0180] user interface

[0181] The monitoring device 44 can provide a user interface that enables a user to specify one or more parameters that the monitoring device needs to perform the automatic setting of tolerance limits, such as the parameters z mentioned above. i Or a specific off-target value p for which the corresponding p-quantile of the distribution of reference values ​​applies as tolerance limit. The user interface furthermore enables the user to manually change the automatically calculated tolerance limits.

[0182] exist Fig.11A highly simplified user interface is illustrated in a highly schematic manner in FIG. 20. For each frequency interval, the operator can enter the factor z in box 201. i The resulting tolerance limits are displayed graphically to the operator. By dragging arrows 202, the operator can manually change each tolerance limit.

[0183] The monitoring device 44 can also provide a user interface that enables output of user information based on the comparison of the test value with the tolerance limit. Fig.12 In the example of FIG. 1 , such a user interface is illustrated in a highly simplified form and in a highly schematic manner. The user interface illustrated here displays the quality of the current machining process and the status of the machine tools for two machines "A" and "B". This display is carried out in the manner of a traffic light system: machining processes for which all test values ​​have the smallest distance from the tolerance limits are displayed by green traffic lights, processes with impermissible process deviations are displayed by red traffic lights, and processes with test values ​​very close to the tolerance limits are displayed by yellow traffic lights. The machine status is also displayed in a similar manner.

[0184] exist Fig.12 In the example of FIG. 2 , the traffic light 212 for the process on machine A is green, and the traffic light for the machine status of machine A is also green. The user can see at a glance that everything is going well in machine A.

[0185] In contrast, the traffic light 214 for the machining process on machine B is red, i.e. an inadmissible process deviation has been determined in this machining process. The traffic light 215 for the machine status of machine B is yellow, i.e. at least one axis of machine B has been found to be critical in the machine diagnosis. In the present example, it is assumed that this is the C1 axis (i.e. one of the two workpiece spindles).

[0186] For example, the process deviation indicator can show that the frequency components of the drive power of the tool spindle or the frequency components of the vibration signal of the vibration sensor 18 are outside the tolerance limit at the rotational speed of the workpiece spindle and its multiples, and the state diagnosis may have shown that when the workpiece spindle is operated, even if no workpiece is clamped on the workpiece spindle, an increased vibration occurs. As already explained above, this together indicates a faulty C1 axis. That is, the comparison of the process deviation indicator and the machine state indicator shows here that the C1 axis is most likely the cause of the determined process deviation, i.e., the fault source indicator has been determined by comparison, and the fault source indicator indicates the C1 axis as the fault source. Correspondingly, the user interface outputs a warning "Attention: Test C1 axis!" to the user.

[0187] The user then has the possibility to investigate the indication in detail. For example, the user interface can provide a view for this purpose in which the indication is displayed in the same manner as in Figure 3 , 6 Or 8 in a similar manner graphically showing the comparison of the test values ​​with the associated tolerance limits, so that it is easy to recognize which frequency components exceed the associated tolerance limits and by what magnitude.

[0188] The user interface can be implemented, for example, in an operating panel 43 or in a mobile terminal 48 .

[0189] Of course, countless other implementations of this user interface are possible.

[0190] Variants

[0191] The invention is not limited to the embodiments described above, and various modifications are possible without departing from the scope of the invention as defined in the claims.

[0192] In particular, it is also possible to use statistical methods that differ from the methods described above to determine the tolerance limits. This also includes machine learning algorithms. For example, reference values ​​can each be used together with an associated quality indicator as a training data set for such a machine learning algorithm, wherein the quality indicator indicates a measure of the quality of the machining process with the aid of which the corresponding reference value has been obtained. For example, the quality indicator can then be determined by (tactile or contactless) measurements at a workpiece for which the reference value has been obtained for machining. Alternatively, the quality indicator can be determined by measurements in an EOL test bench. After presetting the desired machining quality, the machine learning algorithm trained in this way can, for example, automatically determine the tolerance limits, compliance with which is expected to result in the desired machining quality.

[0193] Although the present invention is described with respect to tooth generation grinding, the present invention is also applicable to other types of tooth machining, such as hobbing, peeling, honing, profile grinding, etc. The present invention is also applicable to methods for machining other types of workpieces than teeth.

Claims

1. A method for monitoring a machining process in a machine tool (1), in which a workpiece (23) is machined by means of a tool (16) in one or more machining strokes, the method comprising the following steps: receiving a value of a measured variable (S(t)), wherein the value of the measured variable (S(t)) is determined by measurement at the machine tool during the machining stroke; and At least one test value (S(t)) based on the value of the measured variable (S(t)) i , T i ) and tolerance limit (U i , L i ) for comparison, It is characterized in that By comparing multiple reference values ​​(R i ) performs statistical analysis to determine the tolerance limits (U i , L i ), the reference value being determined by measurements during machining of a plurality of earlier workpieces, wherein each reference value (R i ) is based on one or more values ​​of a measured variable (S(t)) determined during machining of one of the earlier workpieces, and Wherein the reference value (R i ) is used to determine the reference value (R i )’s dispersion (σ i , IQR), and based on the dispersion (σ i , IQR) sets the tolerance limit (U i , L i ).

2. The method according to claim 1, The reference value (R i ) is stored in a database (46), and the method comprises calling the reference value (R) from the database (46) i ).

3. The method according to claim 2, The method comprises: A new reference value (R) is calculated based on one or more values ​​of the measured variable (S(t)) i ),and By setting the new reference value (R i ) is stored in the database (46) to update the database.

4. The method according to any one of the preceding claims, Each reference value (R i ) corresponds to the test value (S i , T i ).

5. The method according to any one of the preceding claims, wherein the time-dependent values ​​of the measured variable (S(t)) are recorded for a plurality of time points during the machining stroke, The test value (S i , T i ) is time-independent and is calculated from the time-dependent value of the measured variable (S(t)), Each reference value (R i ) is time-independent and is based on the time-dependent value of the measured variable (S(t)) during a machining stroke when machining one of the earlier workpieces, where the reference value (R i ) to determine at least one time-independent tolerance limit (U i , L i ),and The time-independent test value (S i , T i ) and the at least one time-independent tolerance limit (U i , L i ) for comparison.

6. The method according to any one of claims 1 to 4, wherein time-dependent values ​​of the measured variable (S(t)) are recorded for a plurality of points in time during the machining stroke, Wherein a plurality of time intervals of the processing stroke are preset, wherein at least one associated test value (S) is determined for each of the time intervals i ), said test value being based on the value of the measured variable (S(t)) in the time interval concerned, wherein at least one tolerance limit (U i , L i ), Each reference value (R i ) is specific to one of the time intervals and is based on the time-dependent value of the measured variable (S(t)) during the machining of one of the workpieces earlier in the time interval in question, and wherein for each of the time intervals, the at least one test value (S i ) and at least one tolerance limit (U i , L i ) for comparison.

7. The method according to any one of claims 1 to 4, wherein time-dependent values ​​of the measured variable (S(t)) are recorded for a plurality of points in time during the machining stroke, wherein a frequency analysis of the time-dependent values ​​of the measured variable (S(t)) is performed in order to determine a plurality of frequency components (T(f)) of the measured variable (S(t)), There are multiple frequency intervals preset. wherein at least one associated test value (T i ), the test value is based on the frequency components (T(f)) in the frequency interval involved, wherein at least one tolerance limit (U i , L i ), Each reference value (R i ) is specifically for one of the frequency intervals and is based on frequency components in the frequency interval in question, which frequency components have been determined by a frequency analysis of the time-dependent values ​​of the measured variable (S(t)) during the machining of one of the earlier workpieces, and For each of the frequency intervals, the at least one test value (T i ) and the at least one tolerance limit (U i , L i ) for comparison.

8. The method according to any one of the preceding claims, The workpiece (23) is machined in at least two machining strokes and the tolerance limits (U i , L i ) specifically for the corresponding machining stroke.

9. The method according to any one of the preceding claims, The degree of dispersion (σ i , IQR) is the standard deviation or the size of the interval between the following values, all reference values ​​(R i ) is within the value interval, and wherein the reference value (R i ) position parameter (μ i ; Q(0.5)) of the tolerance limit (U i , L i ) is set to be a multiple of the standard deviation or a multiple of the size of the value interval.

10. The method according to any one of the preceding claims, The degree of dispersion (σ i , IQR) is the following value interval, in which a predetermined proportion of all reference values ​​lies, and The limits of the value interval form the tolerance limits.

11. The method according to any one of the preceding claims, wherein the tolerance limits are determined by combining at least two statistical analysis methods, or In this case, at least one value of the measured variable or a test value derived therefrom is compared with at least two tolerance limits determined by different statistical methods.

12. The method according to any one of the preceding claims, The measured variable (S(t)) is at least one of the following variables or is derived from at least one of the following variables: a power index, the power index being a measure of the instantaneous power consumption of a tool spindle (15) or a workpiece spindle (21) of the machine tool (1), or A vibration indicator is determined by means of at least one vibration sensor (18) and represents a vibration of the machine tool (1).

13. The method according to any one of the preceding claims, The method further comprises: A normalization operation is performed to convert the at least one test value (S i , T i ) normalized, wherein the normalization operation is related to at least one process parameter, wherein the at least one process parameter is selected from geometric parameters of the tool (16), geometric parameters of the workpiece (23) and setting parameters of the machine tool (1), So that the normalized test value (S i , T i ) is less correlated with the at least one process parameter than without the normalization operation.

14. The method according to claim 13, comprising: changing at least one of the process parameters, and After the change, the normalization operation is recalculated with respect to the process parameter, wherein the recalculation of the normalization operation includes in particular the use of a model which describes the expected dependence of the measured variable (S(t)) on the process parameter, in particular a model of process forces or process performance.

15. The method according to any one of the preceding claims, The method comprises: Output information, the information is based on the at least one test value (S i , T i ) and the tolerance limit (U i , L i ); and / or According to the test value (S i , T i ) and the tolerance limit (U i , L i ) to influence the processing process.

16. A method according to any one of the preceding claims, The method further comprises: Based on the at least one test value (S i , T i ) and the tolerance limit (U i , L i ) to determine the process deviation indicator; and Optionally, the process deviation indicator or information based on the process deviation indicator is output.

17. The method according to claim 16, The method further comprises: reading in a machine state indicator, which is determined by a state diagnosis of the machine tool, wherein in the state diagnosis state data are compared with at least one reference state variable in order to determine at least one machine state indicator, and wherein the state data are determined by measurements in a machine test cycle, in which at least one part of the machine axes is operated in a targeted manner and state data associated with the operation are determined by measurements; and A fault source indicator is determined from the process deviation indicator and the machine state indicator, wherein the fault source indicator contains information about the type of fault source present for the inadmissible process deviation.

18. A method for monitoring a machining process in a machine tool (1), in which a workpiece (23) is machined by means of a tool (16) in one or more machining strokes, the method comprising the following steps: receiving a value of a measured variable (S(t)), wherein the value of the measured variable (S(t)) is determined by measurement at the machine tool during the machining stroke; and At least one test value (S(t)) based on the value of the measured variable (S(t)) i , T i ) and tolerance limit (U i , L i ) for comparison, Based on the at least one test value (S i , T i ) and the tolerance limit (U i , L i ) to determine the process deviation index, reading in a machine state indicator, which is determined by a state diagnosis of the machine tool, wherein in the state diagnosis state data are compared with at least one reference state variable in order to determine at least one machine state indicator, and wherein the state data are determined by measurements in a machine test cycle, in which at least one part of the machine axes is operated in a targeted manner and state data associated with the operation are determined by measurements; and A fault source indicator is determined from the process deviation indicator and the machine state indicator, wherein the fault source indicator contains information about the type of fault source present for the inadmissible process deviation.

19. The method according to claim 17 or 18, comprising: Output information based on the fault source indicator.

20. A monitoring device for monitoring a machining process in a machine tool (1), in which a workpiece (23) is machined by means of a tool (16), It is characterized in that The monitoring device is configured to carry out the method according to any of the preceding claims.

21. A monitoring device according to claim 20, wherein the monitoring device has a user interface, and the user interface is configured to allow a user to: change at least one parameter used by the computer to automatically set the tolerance limit; and / or change the automatically set tolerance limit.

22. The monitoring device according to claim 20, wherein the monitoring device has a user interface configured to output information based on the at least one test value (S i , T i ) and the tolerance limit (U i , L i ) comparison.

23. A computer program comprising instructions which, when the computer program is executed by a computer, cause the computer to perform the method according to any one of claims 1 to 19.

Images