Method for offline and / or online identification of a state of a machine tool, at least one of its tools or at least one workpiece machined therein

Description

[0001] Method for offline and / or online identification of a state of a machine tool, at least one of its tools or at least one workpiece machined therein The invention relates to a method for offline and / or online identification of a state of a machine tool, at least one of its tools or at least one workpiece machined therein, in which the machine tool has at least one, preferably rotatable clamping device for clamping a workpiece to be machined and at least one movable tool holder for positioning a tool held therein, with which the workpiece can be machined, and sensors with which at least the position of the tool and / or the tool holder can be detected in a spatially and temporally resolved manner.

[0002] Today's machine tools, such as CNC machines, are equipped with a multitude of sensors that can continuously record a wide variety of operating parameters. The resulting time series of operating parameters enable the analysis of the machine tool's operating status. For example, it is known from WO 2020 / 038815 A1 or ZHU LIDA ET AL: "Recent progress of chatter prediction, detection and suppression in milling", MECHANICAL SYSTEMS AND SIGNAL PROCESSING, ELSEVIER, AMSTERDAM, NL, Vol. 143, 31 March 2020 (2020-03-31) or GUILLEM QUINTANA ET AL: "Using kernel data in machine tools for the indirect evaluation of surface roughness in vertical milling operations", ROBOTICS AND COMPUTER INTEGRATED MANUFACTURING, ELSEVIER SCIENCE PUBLISHERS BV., BARKING, GB, Vol. 27, No. 6, 24 May 2011 (2011-05-24), pages 1011-1018, to predict the state of a device using a trained Support vector machine to determine.An operating parameter space is divided into classification volumes, of which at least one identifies a normal state and at least one other identifies a fault state of the device.

[0003] The disadvantage is that implementation requires a neural network on the one hand and extensive training data on the other, so that the state of the art provides a complex procedure.

[0004] In addition, the previously known method is difficult to determine different normal operating states of a machine tool. Normal operating states are those operating states in which the machine tool meets the predetermined, i.e.carries out the programmed process steps as intended and without errors. The normal operating states of a machine tool generally include the machining process as the primary state, and secondary states such as return travel, tool holder movements for a tool change, downtimes, or simple idle operation.

[0005] Therefore, there are currently only limited options for determining and analyzing unprofitable downtime in detail. Furthermore, individual sequences of one of the elements cannot be clearly or accurately identified. An element of a machine tool is defined as a component of the machine tool that can be moved within the machine tool. Therefore, the term "element" in this application is a collective term, which specifically includes the tool(s), their tool holders, and the clamping device for clamping the workpiece.

[0006] While it is possible to distinguish between operation and standstill of a machine tool when checking the operating status based on internal machine signals, this is achieved by measuring the power consumption of the drive motors and changing position data of the tool holder or tool. Non-productive times, however, are currently only estimated or manually recorded on the machine. A detailed and complete description of the operating status requires continuous documentation of all events.

[0007] In this respect, the object of the invention is to provide a method for the offline and / or online identification of a state or sequence of a machine tool, such as a CNC machine tool, with which the operating states or sequences can be identified in a simple and efficient manner. At the same time, the object of the invention is to provide a corresponding device.

[0008] These objects are achieved by the subject matter of the independent claims. Advantageous embodiments of the invention are the subject matter of the further patent claims and the following description.

[0009] According to the invention, the operating state of a machine tool or a sequence of a travel, of at least one of its tools, or of at least one workpiece machined therein, can be identified. The machine tool has at least one preferably rotatable clamping device for clamping a workpiece to be machined and at least one movable tool holder for positioning a tool held therein, with which the workpiece can be machined, and sensors with which at least the position of the tool and / or the tool holder can be detected in a spatially and temporally resolved manner. For identification according to the invention, the following steps are carried out.

[0010] First, at a series of consecutive times i, where i = 1...n, the positions P of the tool and / or the tool holder are recorded by the sensors with spatial and temporal resolution. These time series of data can then be evaluated immediately after recording, i.e., online, according to steps b) and c). It is also possible to use the method according to steps b) and c) to analyze previously completed operating phases of the machine tool. In this case, the historical data recorded during these operating phases are subsequently provided to the evaluation method described here, i.e., offline.

[0011] The data of the sensors on the positions of the tool and / or the tool holder, available as time series at the time points i, are each calculated in a first calculation step according to Δ m i = P i P i − 1 into a series of position changes Δ mi and according to Δ v i = v i v i − 1 with v i = P i − P i − 1 t i − t i − 1 und v i − 1 = P i − 1 − P i − 2 t i − 1 − t i − 2 into a series of speed changes Δv i converted.

[0012] From these data series, the condition of the tool, the tool holder, the machine tool and / or the workpiece machined in the machine tool can then be determined.

[0013] The special feature lies in the simplicity of the specified procedure, which requires neither a complex transformation of the acquired or provided sensor data nor the training of a neural network of any kind.

[0014] According to the invention, the positions P are recorded as coordinates P i (xi , yi , zi ) of a Cartesian coordinate system, stored, and made available to the evaluation process. Thus, according to the axes x, y, z of the coordinate system, the position changes Δ mi their respective components Δ m xi , Δ m yi , Δm zi and for the speed changes Δ vitheir respective components Δv xi , Δv yi , Δ v zi Such a method is then based on the coordinate system already used by the machine tool, eliminating the need to adapt the sensor data acquired by the machine tool. This supports real-time determination of the operating state and enables easier implementation of the method.

[0015] According to the invention, the identified state is an operating state, in particular a normal operating state, which represents a movement of the tool or tool holder, in particular for workpiece machining, for tool repositioning, in particular returns, for tool changing, as well as a rest phase of the tool or tool holder, in particular a downtime and / or idling, and / or a deviation from a predetermined movement speed of the clamping device, in particular override commands. Furthermore, the method allows individual operating states, when they are movements, to be divided into several sequences with different significance.This specifically means that the "cutting process" state can be divided into the sequences "approach," "cutting" (meaning "the tool reaches and contacts the workpiece," "cutting sequence" (meaning "the tool processes the workpiece," and "cutting off" (meaning "the tool loses contact with the workpiece"). This also applies to other movements of the elements.

[0016] Advantageously, additional sensors record electrical parameters of the machine tool's drive motors. For example, the additional sensors record the electrical current consumption and / or the applied electrical supply voltages of the drive motors, by means of which, for example, the clamping device is rotated and / or the tool holders are moved in space. The characteristic curves and / or time series determined in this way can then be analyzed in a further process step. For example, these characteristic curves or time series are combined with the previously determined series of position changes and speed changes, so that, in particular, an analysis of the machining processes is possible. Faulty operating states can also be determined from this. On the one hand, such findings enable the machine tool to be protected from consequential damage.On the other hand, damage to the workpiece can be detected and reported, allowing for timely inspection of the unfinished workpiece. This can prevent unnecessary further processing of a potentially unusable workpiece, improving the machine tool's utilization rate, saving costs, and accelerating the completion of a series of workpieces.

[0017] It is advisable to provide and use additional data, particularly correction factors or tool parameters, in advance to determine the conditions. Correction factors include, for example, values ​​used to precisely determine the position of a tool's cutting edge.

[0018] In an advantageous process step, the series of position changes and / or the series of speed changes are analyzed using a limit value analysis. For example, using the formula Lim v f = ∑ i = 1 n U v , f i n ± ∑ i = 1 n U v , f i − U v , f i ¯ 2 n − 1 2 , where U is a matrix describing the position of the tool or the tool holder, determine whether either a tool change or a machining process is taking place.

[0019] A simple and efficient way of identifying the states is to make case distinctions in which it is checked whether the relevant value for the position change Δm i and / or the change in speed Δv i , or one or more of its components is less than 1, equal to 1, greater than 1 or 0.

[0020] The series of position changes and / or speed changes are expediently displayed in a diagram as characteristic curves and / or provided in a data array, which is used to analyze the respective state.

[0021] The method is particularly preferably used to detect an overload of one of the drives of the machine tool, to detect wear on the machine tool or on the tool, to detect a manufacturing or workpiece defect and / or to detect process instabilities, in particular chatter, ie a regenerative effect.

[0022] The method described above or its preferred embodiments are expediently computer-implemented. Accordingly, the invention also encompasses a data processing device comprising means for carrying out the method or for carrying out a preferred embodiment. Furthermore, the invention also encompasses a computer program product comprising instructions which, when executed by a computer, cause the computer to carry out the steps of the method or of a preferred embodiment. The invention also includes a computer-readable medium comprising instructions which, when executed by a computer, cause the computer to carry out the steps of the method or of a preferred embodiment.

[0023] The description of preferred embodiments of the invention given so far contains numerous features, some of which are summarized in the individual dependent claims. However, these features can also be considered individually and combined into further meaningful combinations. In particular, these features can each be combined individually and in any suitable combination with the inventive method, the inventive data processing device, and the inventive computer-readable medium. Furthermore, method features can also be viewed as a property of a corresponding device unit.

[0024] The above-described properties, features, and advantages of the invention, as well as the manner in which they are achieved, will become clearer and more readily understood in connection with the following description of the embodiments of the invention, which are explained in more detail in conjunction with the figures. The embodiments serve to illustrate the invention and do not limit the invention to the combinations of features specified therein, including with regard to functional features.

[0025] They show: Figure 1 shows a schematic representation of a machine tool, Figure 2 shows the working area of ​​the Figure 1shown machine tool, Figure 3 shows a flow chart for the method for identifying states of a machine tool, Figure 4 shows a diagram of a machining process, Figure 5 shows a matrix U with a series of position changes, Figure 6 shows a characteristic curve of the torque-generating current of the z-axis for the entire machining process of a valuable piece, and Figure 7 shows a characteristic curve for the drive torque of the main spindle during a single machining operation.

[0026] In all figures, identical features are provided with the same reference numerals.

[0027] Figure 1 shows, as an example, a turning-milling CNC machine CM as a machine tool WM. The machine tool WM comprises a work area AR which can be closed by a sliding door TR and in which two opposing spindles SP, a main spindle SP 1 and a secondary spindle SP 2 , are arranged as clamping devices AV, AV for clamping a workpiece WS ( Figure 2) are provided. The work area AR also contains two movable tool holders WH 1 , WH 2 , each of which can accommodate several tools not shown here. Furthermore, the WM machine tool includes a BE control system for programming, controlling, and monitoring its elements.

[0028] The AR workspace and the elements arranged therein then show Figure 2 in detail. The two spindles SP 1 and SP 2 are arranged concentrically to one another and can rotate about their common longitudinal axis. They each include a clamping device AV, with which a workpiece WS to be machined can be rotated. The machine tool WM also has two tool holders WH 1 and WH 2, which can be moved in all three spatial directions. The upper tool holder WH 1, as a motor milling spindle, comprises only three stations, whereas the tool holder WH 2 arranged below is designed as a tool turret with several stations.

[0029] To record the positions, the control system BE of the machine tool WM uses a virtual Cartesian coordinate system KS with the three orthogonal machine axes x , y, and z. The two spindles SP 1 and SP 2 as well as the clamping devices AV arranged on them can be rotated around the z-axis and also moved along the z-axis in such a way that workpieces can be transferred from the auxiliary spindle to the main spindle, or vice versa, without the assistance of a user.

[0030] Adjacent to the work area AR is a magazine room MG, which is closed by a magazine door in the illustration shown. The magazine room MG contains a large number of tools WZ, i.e. Drills, milling cutters and the like, which can be gripped by the upper tool holder WH 1 with the magazine door open and can be put back or returned therein.

[0031] The WM machine tool can be in various states, i.e., operating modes, during operation. A distinction must be made here between faulty operating modes on the one hand and normal operating modes, so-called normal operating modes, on the other. An example of a faulty operating mode is the "tool breakage" state. Other operating modes that indicate a fault are conceivable. In contrast, normal operating modes can be, for example, "cutting process," "idling," "override command," or even "tool change." An override command is understood to be a manual intervention by a user of the WM machine tool that accelerates or decelerates the programmed sequence of workpiece machining. Other normal operating modes can also be sequences of the aforementioned operating modes if, as it were, partial sections of them are (should be) recognized.Using the operating conditions detected by the process, it is then possible to analyze and, if necessary, optimize the machining of the workpiece WS.

[0032] The workpiece can also be in different states. With reference to the workpiece WS, the method enables, for example, spatially resolved identification of possible manufacturing defects, thus reducing the effort required for quality assurance.

[0033] To determine the states, the WM machine tool is equipped with a variety of sensors (not shown). Some of these sensors can determine the positions of the tool and / or the tool holders or even the clamping devices with spatial and temporal resolution. The position—i.e., the spatial coordinates P(x, y, z) of the clamping devices AV or the workpiece WS clamped therein, the tool WZ, or the tool holders WH 1 , WH 2—is usually recorded separately for each machine axis using appropriate sensors.

[0034] For example, the position of the cutting edges of the tools WZ can be determined via the position of the tool holders by supplementing the tool holder position with previously provided data about the size of the tool in question. It is also possible to automatically measure the position of the cutting edges of the tools in the machine tool WM to determine correction data. Additional sensors are capable of continuously recording the electrical currents and supply voltages of the machine tool's drive motors (not shown), which can be used to drive, i.e., rotate and / or reposition, the respective rotatable and / or displaceable elements.

[0035] The signals from these sensors therefore contain data such as those recorded during machine downtimes, tool changes, idle times, rapid traverse, acceleration effects, the cutting sequence, and the actual machining process. The recorded data signals thus represent, among other things, position data, drive parameters and power, correction factors, and tool parameters.

[0036] The method proposed for identifying states is described in Figure 3 shown schematically, whereby further terms necessary for characterization are defined below: The feed rate of an element is calculated from the product of the existing feed (f) along the considered machine axis and the associated speed n, shown below only as an example for the x-axis: v f , x = f x ∗ n

[0037] For the position change of the element under consideration required for state identification, the vector Δm is introduced, defined as the ratio, i.e., the quotient, of its position at any time i to its previous position, i.e., at time i-1. This is shown as an example in Eq. 2 for the x-coordinate of the element in question. Δm x = x i x i − 1

[0038] The method offers the possibility of using the procedure offline and online.

[0039] Furthermore, the relationship shown in Eq. 3 applies to the change in feed rate in, for example, the x-direction: Δv f , x = v f , x i v f , x i − 1 where the speeds are determined according to v i = P i − P i − 1 t i − t i − 1 und v i − 1 = P i − 1 − P i − 2 t i − 1 − t i − 2

[0040] In a first method step 102 of the method 100 according to the invention, the respective current position of the tool WZ and / or the tool holder WH 1 , WH 2 is recorded or provided as data at a series of points in time. The processing of this data in a second method step 104 according to the above equations (1) - (5) can take place immediately upon its creation, which enables online identification of the state. If the calculation and state identification are staggered in time relative to the recording of the data, this is referred to as offline identification of the state. In a final method step 110, the state of the tool, the tool holder, the machine tool and / or the workpiece machined in the machine tool is then identified based on the previously determined or provided position changes and speed changes.

[0041] In principle, a distinction can be made between different cases for position changes and speed changes per machine axis and per element. First, it is only necessary to determine the direction in which the element is moving. This can be determined using Table 1 below. The position changes and speed changes represent nothing more than the quotients of the raw data. Table 1: Meanings based on the Quotients case Meaning Δm x < 1 Element moves in negative x-direction Δm x = 1 Element moves at a constant x-level Δm x > 1 Element moves in positive x-direction ... ... Δm z < 1 Element moves in negative z-direction Δm z = 1 Element moves at a constant z-level Δm z > 1 Element moves in positive z-direction Δv f, x < 1 Feed rate in x-direction decreases Δv f,x = 1 Feed rate in x-direction is constant Δv f,x > 1 Feed rate in x-direction increases ... ... Δv f, z < 1 Feed rate in z-direction decreases Δv f,z = 1 Feed rate in z-direction is constant Δv f, z > 1 Feed rate in z-direction increases

[0042] Subsequently, by means of a combined analysis of two or more cases, the condition, in particular the operating condition, of the machine tool at the time in question in the series of time points can be determined.

[0043] Below, some conditions are listed schematically as examples, based on which the procedure can identify different operating states. a) Identification of return runs: If more than one machining process is performed on a workpiece, the tool must be returned to the starting point. To identify this state, one of the three conditions listed below must be met: Δm z > 1 & Δm x > 1 Δm z = 1 & Δm x > 1 Δm z > 1 & Δm x = 1 b) Identification of downtimes and idle times: In the event that the machine tool is manually put into idle mode during the machining process, for example, by a user, so that no machining takes place despite the main spindle being rotated, the condition below applies. This also applies to downtimes when the main spindle is stationary. Δv f , x = Δv f , z = Δv f , y = 0 c) Identification of idle or standstill: A distinction between idle and standstill can be made using the spindle speed additionally read from the BE control system. If the speed of the clamping device AV holding the workpiece WS is not equal to 0, then idle is present. d) Identification of override commands: The BE control system of the machine tool returns manually executed override commands. These can be read and subsequently processed. If the conditions from Figure 4 they are assigned to the machining process. Override commands change the conditions from Figure 4Not applicable. It should be noted that, depending on the command, the conditions for identifying downtimes and idle times may also apply. e) Identification of the actual machining process and determination of the individual cutting sequences: To identify the individual cutting sequences during continuous machining, the first step is to consider the axis feed rate signal. To ensure that even low feed rates are identified as such during machining, it is possible to square the values ​​and then round them to the nearest natural number, including 0.

[0044] In the next step, the position change and the change in feed rate of the respective axes are determined with high accuracy. This allows even small deviations to be identified.

[0045] Subsequently, a further case differentiation must be carried out so that the currently performed cutting sequence can be identified. The possible cases and the respective conditions are listed in the Figure 4 The conditions set out also apply to grooving and internal machining of the workpiece.

[0046] In Figure 4 In the embodiment shown, for example, the first tool holder WH 1 moves from its rest position in the negative z-direction. The method is able to detect this condition using the specified quotients. For this to happen, all of the conditions listed in B1 must be met: Δv f , x = 1 Δv f , z = 1 Δm x = 1 Δ m z < 1

[0047] As time progresses, the tool holder WH 1 is also moved in the positive X direction, ensuring that all conditions B2 are met. When sensor data that satisfy conditions B4 first occurs, the process detects the start of a machining process: the "lead-in" sequence. The end of the machining process, the "cut-off" sequence, is detected when conditions B5 are no longer met, but rather conditions B4.

[0048] To implement the method according to the invention, the data points of the raw signal applicable to the respective cases for each element for each machine axis are stored in a matrix U, as shown in Figure 5for an element for one axis, shown as an example. In the illustrated embodiment, the corresponding seventeen quotients are shown for seventeen points in time from a series of points in time. The quotients are preferably the feed rate in the negative z-direction.

[0049] If, after a selection of operating conditions, only machining processes are to be further analyzed, tool changes must be identified and eliminated, since some of these conditions could also apply to the tool change.

[0050] In the method according to the invention, a tool change can be detected by considering a limit value. The limit value is defined as the sum of the empirical mean of the feed rates stored in U (U v,f (i)) plus a standard deviation of 50%: Lim v f = ∑ i = 1 n U v , f i n ± ∑ i = 1 n U v , f i − U v , f i ¯ 2 n − 1 2

[0051] As soon as a feed rate data point in matrix U exceeds this limit, it is identified as belonging to the tool change and is set to zero in matrix U. The indices of the values ​​in matrix U that are not identified as tool changes and are therefore identified as belonging to the machining process are written into a vector t in the next step to select the machining processes. Thus, vector t contains only those times at which the feed rate in the negative z-direction is not equal to 0. By transferring this value to vector t, the distances (d) between the individual measurement points can be determined: d i = t i + 1 − t i i ∈ ℕ 1 ≤ i ≤ n − 1

[0052] This determines the values ​​stored in d. For a continuous measurement signal, d = 1. Therefore, the condition b = e + 1 can be used to determine the starting value (b) of the subsequent machining process and the final value (e) of the current machining process for each individual machining process, which are not directly identifiable as such in the matrix U shown.

[0053] It is also possible to quantitatively evaluate the temporal progression of position and speed changes and draw conclusions about the operating status for freely selectable or predetermined time periods. For example, pulsating variations in speed changes within a specific time period can indicate a regenerative effect. Using this method, it is also possible to detect differences in speed changes that occur during the same machining step of two identical workpieces manufactured consecutively. This can provide information about the wear of the tool used during the process.

[0054] In addition, it is possible in a further process step 106 ( Figure 3) to record further internal machine signals, in particular electrical current and voltage signals for the drive motors of the machine tool. For this purpose, Figure 6 The torque-generating current for the drive motor responsible for adjustment along the z-axis for the entire machining process of the workpiece. These include the individual machining times BZ as well as the resulting non-productive times NZ.

[0055] Figure 7 shows the section DL from the diagram according to Figure 6in a higher time resolution. The current curve for the drive of the main spindle is shown over time and can be divided chronologically into several time periods. During a first time period K1 between time t0 and time t1 the drive is at a standstill. Then a needle-shaped current pulse occurs at time t1 as time period K2, followed by a time period K3 for the drive control of the motor, during which the main spindle is moved to the programmed position. Having arrived there at time t2, time period K4 begins, in which the tool WZ reaches the workpiece WS and makes contact. This is the lead-in sequence. Time period K4 is also comparatively short and ends at time t3, followed by time period K5 for the main machining of the workpiece, i.e. the cutting process. It ends at time t4.Then a second drive control K3 takes place again between the times t 4 and t 5 and a second acceleration effect K2 between t 5 and t 6 .

[0056] Thus, in a further method step 106 ( Figure 3 ) such machine signals are recorded or provided and taken into account when identifying the state in method step 110. This opens up the possibility of identifying unprofitable process states, such as downtimes, idle times, tool changes, rapid traverses or the aforementioned override commands. In particular, the consideration of the additional machine signals makes it possible to detect faulty states, for example, when an unexpected change in the current flow ( Figure 7 , arrow PF). This allows the production process to be adapted as needed.

[0057] Overall, the individual machining processes can be identified and determined from the continuous manufacturing process. Furthermore, the method represents a simple and cost-effective method for online / offline process monitoring throughout the entire manufacturing process.

Claims

1. Method (100) for the offline and / or online identification of an operating state of a machine tool (WM), wherein the machine tool (WM) has at least one preferably rotatable clamping apparatus (AV) for clamping a workpiece (WS) to be processed, and at least one tool holder (WH1, WH2) which is able to be moved for positioning a tool (WZ) which is held therein with which the workpiece (WS) can be processed, and sensors with which at least the position of the tool and / or the tool holder can be detected in a spatially and time-resolved manner, comprising the steps: a) for a series of time points i, i = 1...n, detecting or providing (102) positions P of the tool and / or the tool holder (WH1, WH2), wherein the positions P are present as coordinates Pi (xi, yi, zi) of a Cartesian coordinate system (KS), characterised in that it additionally comprises the following steps: b) establishing (104) for the series of time points i b1) a series of position changes Δmi of the tool (WM), the tool holder (WH1, WH2) and / or the clamping apparatus (AV) in accordance with Δ m i = P i P i − 1 and b2) a series of speed changes Δvi of the tool (WM), the tool holder (WH1, WH2) and / or the clamping apparatus (AV) in accordance with Δ v i = v i v i − 1 where v i = P i − P i − 1 t i − t i − 1 and v i − 1 = P i − 1 − P i − 2 t i − 1 − t i − 2 , wherein according to the axes x, y, z of the Cartesian coordinate system (KS) for the position changes Δmi their respective components Δmxi, Δmyi, Δmzi and for the speed changes Δvi, their respective components Δvxi, Δvyi, Δvzi are determined, c) identifying (110) the operating state of the machine tool (WM) on the basis of the position changes Δmi and the speed changes Δvi, wherein the operating state represents a movement of the tool or tool holder, a rest phase of the tool or tool holder and / or represents a deviation from a predetermined movement speed of the clamping apparatus.

2. Method (100) according to claim 1, wherein the operating state represents d) the movement of the tool and / or the tool holder d1) for workpiece processing, in particular cutting sequences, d2) for tool repositioning, in particular return movements, d3) for tool changing, wherein e) the operating state represents a standstill time and / or an idling and / or f) the deviation from override commands.

3. Method (100) according to one of the preceding claims, wherein further sensors detect electrical parameters of drive motors of the machine tool.

4. Method (100) according to one of the preceding claims, wherein further data, in particular correction factors or tool parameters for determining the states are provided and / or established in advance and are utilized during the identification.

5. Method (100) according to one of the preceding claims, wherein the series of position changes Δmi and / or the speed changes Δvi are analyzed on the basis of a limit value consideration.

6. Method (100) according to one of the preceding claims, wherein the identification of the state is made on the basis of case distinctions, wherein it is tested whether the value in question for the position change Δmi and / or the speed change Δvi, or if dependent upon claim 2, one of its components is / are less than 1, equal to 1, greater than 1, or 0.

7. Method (100) according to one of the preceding claims, wherein the series of position changes Δmi and / or the speed change Δvi are represented in a diagram as characteristic lines or are provided in a data array and, on the basis thereof an analysis of the respective state takes place.

8. Method (100) according to one of the preceding claims, used g) for recognizing an overloading of one of the drives of the machine tool, h) for recognizing wear on the machine tool and / or on the tool, i) for recognizing a production or workpiece fault and / or j) for recognizing process instabilities, in particular juddering (regenerative effect).

9. Method (100) according to one of the preceding claims, which is computer-implemented.

10. Apparatus, in particular a control system (BE), for data processing, comprising means for carrying out the method according to one of claims 1 to 9.

11. Computer program product comprising commands which, on execution of the program by way of a computer cause said computer to carry out the steps of the method (100) according to one of claims 1 to 9, wherein the method (100) is embodied for the offline and / or online identification of an operating state of a machine tool (WM), and the machine tool (WM) has sensors with which at least the position of the machine tool (WZ) and / or the tool holder (WH1, WH2) can be detected in a spatially and time-resolved manner, wherein the method (100) comprises the step a), in which, for a series of time points i, i = 1...n, positions P of the tool (WZ) and / or the tool holder (WH1, WH2) are detected.

12. Computer-readable medium comprising commands which, on execution by way of a computer cause said computer to carry out the steps of the method (100) according to one of claims 1 to 9, wherein the method (100) is embodied for the offline and / or online identification of an operating state of a machine tool (WM), and the machine tool (WM) has sensors with which at least the position of the machine tool (WZ) and / or the tool holder (WH1, WH2) can be detected in a spatially and time-resolved manner, wherein the method (100) comprises the step a), in which, for a series of time points i, i = 1...n, positions P of the tool (WZ) and / or the tool holder (WH1, WH2) are detected.