A control device used on a numerically controlled machine tool, and a machine tool including the control device

Abstract

The present invention relates to a control device 200 used on a numerically controlled machine tool 100, including a control unit 230 for controlling an actuator for performing a machining process of a workpiece on the machine tool, in particular based on control data, and a monitoring unit 250 for monitoring the working state on the machine tool 100. According to the present invention, the monitoring unit 250 includes a computer-implemented neural network 253 (NN), which is specifically configured to read input data from the machine control unit 230 and output output data indicating the working state of the machine tool 100.

Description

Field of the Invention

[0001] The present invention relates to a control device for use on a numerically controlled machine tool and a machine tool including such a control device, in particular for monitoring a numerically controlled machine tool, in particular via a monitoring unit of the control device. The present invention also relates to a method for monitoring a numerically controlled machine tool, in particular by means of a control device or a monitoring unit of a control device, and a corresponding computer program product for a machine tool control device. Background Art

[0002] Numerically controlled machine tools for machining workpieces, such as milling machines, lathes, milling / lathe machines, grinding machines, universal machine tools, turning centers, machining centers, gear cutting machines, etc., are known in the prior art.

[0003] Such machine tools or numerically controlled machine tools of a general type (CNC machine tools) typically include a control device for the numerically controlled machine tool (machine tool controller), which includes a machine control unit for controlling the actuators of the machine tool (in particular spindle drives and axis drives such as linear, rotational and rotary axis drives), and a machine tool for the machining process of machining the workpiece, in particular based on control data or control data containing a numerical control program (NC program).

[0004] For example, according to EP 2 482 156A1, it is known to perform process monitoring during the machining of workpieces on a machine tool, in particular for collision monitoring on the machine tool. For example, according to the teachings of EP 2 482 156A1, this can be performed based on a collision sensor that measures vibrations, wherein during the operation of the machine tool while machining the workpiece, the sensor signal is continuously compared with one or more limit values, and when the limit value is exceeded, a collision can be detected.

[0005] According to EP 2 482 156A1, a control device of a general type or a machine tool having such a control device is known, including a machine control unit for controlling the actuators of the machine tool for performing the machining process of the workpiece on the machine tool (in particular based on control data), and a monitoring unit for monitoring the machine tool based on sensor data from a collision sensor.

[0006] Regarding the process monitoring of machine tools, it is generally necessary to monitor the machining process of the machine tool to ensure that various potential problems that may occur during the workpiece machining process (such as collisions, bearing wear or bearing damage, tool wear, tool breakage, clamping device breakage, drive damage, etc.) can be quickly identified and as reliably as possible discriminated, or a quick response can be made to such problems (e.g., by automatically stopping the drive or even automatically stopping the machine), and in this case, any downtime caused by possible misidentification can also be avoided.

[0007] Accordingly, based on the above prior art, the object of the present invention is to provide process monitoring on a machine tool, which is improved compared to the prior art, in particular being able to react faster, more precisely, more variably and / or more sensitively to problems occurring during the processing, and being able to better identify these problems and / or better avoid false identification when performing reliable process monitoring. Summary of the Invention

[0008] According to the present invention, in accordance with the independent claims, there is provided a control device for use on a numerically controlled machine tool, a corresponding machine tool, a method for monitoring a machine tool, and a corresponding computer program product. The dependent claims relate to preferred exemplary embodiments of the present invention, which may be provided individually or in combination.

[0009] According to one aspect of the present invention, there is provided a control device for use on a numerically controlled machine tool, comprising: a machine control unit for controlling, in particular based on control data, an actuator for performing a machining process of a workpiece on the machine tool, and a monitoring unit for monitoring the working state of the machine tool.

[0010] According to the present invention, the monitoring unit of the control device comprises a computer-implemented neural network or an artificial neural network, which is configured to read input data from the machine control unit and output output data indicating the working state of the machine tool.

[0011] It should be noted that DE 10 2015 115 838A1 seems to have described the use of a neural network related to a machine tool. However, apparently, in the evaluation unit, it is used to evaluate temperature measurement management and (compared with the present invention) to compensate or correct temperature-related changes in the position on the machine tool in DE 102015 115 838A1. However, DE 10 2015 115838A1 cannot teach or suggest using a neural network for tool or process monitoring or for monitoring the working state of a machine tool.

[0012] Therefore, according to an exemplary embodiment of the present invention, a new and different from the prior art way is proposed to perform tool or process monitoring of a machine tool by providing an artificial or computer-implemented neural network on the control device of the machine tool.

[0013] Advantageously, such tool or process monitoring can protect the machine and / or the workpiece from damage, can ensure optimal tool use, and can provide a starting point for process optimization when necessary.

[0014] Therefore, productivity can be increased and the overall life cost of the machine tool can be reduced. At the same time, possible tool and process monitoring contributes to quality assurance and allows workpiece control and quality or program documentation.

[0015] Advantageously, direct error or problem detection (e.g., collision detection, detection of broken, worn or missing tools) can be performed, and corresponding reactions such as machine stop or tool change can be triggered automatically, immediately and without delay. This also allows for further cost reduction and scrap reduction, for example, through wear-related tool replacement or bearing maintenance.

[0016] Furthermore, for example, by continuously adapting the process parameters based on the output data of the neural network, an optimized machining speed can be achieved. Additionally, it is advantageous as historical output data and historical training data can be evaluated, for example, for recording and statistics.

[0017] In summary, the following advantages can be achieved, especially in exemplary embodiments: comprehensive protection of the machine, workpiece and tool, real-time monitoring, optimal tool utilization, part quality monitoring (e.g., through recording and process analysis), reduction of scrap and / or adaptation to complex processes or machine machining.

[0018] Moreover, in an exemplary embodiment, the advantage is that since the data or information available at the machine control (e.g., drive data and / or positioning data) can be utilized, sensorless monitoring can also be provided, or at least some or additional sensors can be dispensed with. In a further exemplary embodiment, the input data of the neural network can be supported by sensor data or provided by sensor data, for example, by additional or alternative sensors (e.g., for strain, force, active power, torque, vibration, acceleration, structure-borne sound and / or temperature, etc.).

[0019] Some exemplary embodiments of the present invention are described below, but this should not be construed as exhaustive or restrictive. It should also be noted that such exemplary aspects can be provided individually or in combination.

[0020] According to an exemplary preferred embodiment, the monitoring unit, in particular the neural network of the monitoring unit, can preferably be configured to read input data from the machine control unit, especially in real time, during the ongoing machining process of the workpiece performed on the machine tool, and / or to output output data indicating the current working state of the machine tool, especially in real time, during the ongoing machining process of the workpiece performed on the machine tool.

[0021] According to an exemplary preferred embodiment, the monitoring unit, especially the neural network of the monitoring unit, can be configured to evaluate the output data of the neural network.

[0022] According to an exemplary preferred embodiment, the monitoring unit, especially the neural network of the monitoring unit, can be configured to evaluate the output data of the neural network for tool monitoring and / or process monitoring on the machine tool.

[0023] According to an exemplary preferred embodiment, a monitoring unit, in particular a neural network of the monitoring unit, can be configured to output control data affecting the machining process to a machine control unit based on input data and / or based on an evaluation of output data.

[0024] According to an exemplary preferred embodiment, a monitoring unit, in particular a neural network of the monitoring unit, can be configured to determine when a machine tool is in an abnormal operating state based on input data and / or based on an evaluation of output data.

[0025] The abnormal operating state can be or include, for example: collisions (e.g., collisions of machine parts or tools and / or workpieces with machine parts), (possibly excessive) tool wear, tool breakage, (possibly excessive) bearing wear, bearing damage, clamping device breakage, drive device damage, etc.

[0026] According to an exemplary preferred embodiment, a monitoring unit, in particular a neural network of the monitoring unit, can be configured to determine when the likelihood of an abnormal operating state of a machine tool exceeds a predetermined limit value based on input data and / or based on an evaluation of output data.

[0027] According to an exemplary preferred embodiment, a monitoring unit, in particular a neural network of the monitoring unit, can be configured to output control data affecting the machining process to a machine control unit when it is determined that there is an abnormal operating state of the machine tool and / or the likelihood of an abnormal operating state of the machine tool exceeds a predetermined limit value.

[0028] According to an exemplary preferred embodiment, an abnormal operating state can be a collision occurring on the machine tool.

[0029] According to an exemplary preferred embodiment, an abnormal operating state can be a tool breakage occurring on the machine tool.

[0030] According to an exemplary preferred embodiment, an abnormal operating state can be a tool that is missing on the machine tool and will be used in the machining process.

[0031] According to an exemplary preferred embodiment, a monitoring unit, in particular a neural network of the monitoring unit, can be configured to output control data triggering a drive stop and / or a machining stop to a machine control unit when it is determined that there is an abnormal operating state of the machine tool and / or the likelihood of an abnormal operating state of the machine tool exceeds a predetermined limit value, and particularly preferably such that the machine control unit triggers a drive stop and / or a machine stop on the machine tool based on the control data.

[0032] According to an exemplary preferred embodiment, an abnormal operating state can be an increase in tool wear of a tool used in the machining process occurring on the machine tool or tool wear of a tool used in the machining process exceeding a limit value.

[0033] According to an exemplary preferred embodiment, the abnormal operating state may be an increase in bearing wear of the machine tool bearings occurring on the machine tool or damage to the bearings on the machine tool.

[0034] According to an exemplary preferred embodiment, the monitoring unit, in particular the neural network of the monitoring unit, may be configured to output control data for triggering an automatic tool change to the machine control unit when it is determined that an abnormal operating state of the machine tool exists and / or the probability of occurrence of an abnormal operating state of the machine tool exceeds a predetermined limit value (in particular when the abnormal operating state includes detection of excessive tool wear or tool breakage), and in particular, such that the machine control unit triggers an automatic tool change on the machine tool based on the control data.

[0035] According to an exemplary preferred embodiment, the monitoring unit, in particular the neural network of the monitoring unit, may be configured to adjust process parameters existing in the machine control unit based on input data and / or based on an evaluation of output data.

[0036] According to an exemplary preferred embodiment, the monitoring unit, in particular the neural network of the monitoring unit, may be configured to adjust control data existing in the machine control unit based on input data and / or based on an evaluation of output data.

[0037] According to an exemplary preferred embodiment, the monitoring unit, in particular the neural network of the monitoring unit, may be configured to adjust control data existing in the machine control unit and / or process parameters existing in the machine control unit in such a way that the machining speed of the machining process is adjusted, in particular by adjusting the feed rate during the machining process.

[0038] According to an exemplary preferred embodiment, the input data may indicate operating parameters of actuators or drives of the machine tool, in particular drive speed, motor current, and / or actuation signals output to the actuators.

[0039] According to an exemplary preferred embodiment, the input data may indicate position values of movable parts of the machine tool, in particular the actual and / or target positions of linear, rotational, and / or swivel axes of the machine tool.

[0040] According to an exemplary preferred embodiment, the input data may indicate sensor values from machine tool sensors, in particular sensor values from temperature sensors, force sensors, strain sensors, torque sensors, acceleration sensors, oscillation or vibration sensors, and / or structure-borne sound sensors.

[0041] According to an exemplary preferred embodiment, the monitoring unit may include an internal data storage device and / or may be configured to communicate with an external data storage device. The monitoring unit may preferably be configured to store the input data and / or output data of the neural network in the internal and / or external data storage device, and / or store the evaluation data generated based on the evaluation of the input data and / or output data of the neural network in the internal and / or external data storage device.

[0042] According to an exemplary preferred embodiment, the control device may further include a human-machine interface operable by an operator, in particular a graphical user interface. The monitoring unit may preferably be configured to output, on the human-machine interface, the output data of the neural network indicating the working state of the machine tool to the operator.

[0043] According to a further aspect of the present invention, preferably, a machine tool is proposed, which includes a control device according to one or more of the above aspects or embodiments.

[0044] According to a further aspect of the present invention, preferably, a method for monitoring a numerically controlled machine tool is proposed, in particular by means of a control device according to one or more of the above aspects or embodiments.

[0045] The method preferably includes controlling an actuator of the machine tool by a machine control unit, in particular based on control data, for performing a machining process of a workpiece on the machine tool, and monitoring the working state of the machine tool by a monitoring unit. More preferably, it includes reading input data from the machine control unit into a neural network implemented on a computer in the monitoring unit and / or outputting output data indicating the working state of the machine tool from the neural network.

[0046] Preferably, the method additionally or alternatively includes monitoring the working state of the machine tool by the monitoring unit during controlling the actuator of the machine tool for performing a machining process of a workpiece on the machine tool, in particular based on control data. The method preferably includes reading input data from the machine control unit into a neural network implemented on a computer in the monitoring unit and / or outputting output data indicating the working state of the machine tool from the neural network.

[0047] According to a further aspect of the present invention, preferably, a computer program product containing instructions is proposed, which, during the execution of a computer program on a computer connected to a numerically controlled machine tool or a (preferably computer-implemented) control device of the numerically controlled machine tool according to one of the foregoing aspects or embodiments, causes it to execute the method according to the above aspects or embodiments.

[0048] In the following description and explanation of the drawings, other aspects of the above aspects and features, as well as their advantages and more specific implementation possibilities, are described, and these descriptions and explanations are not restrictive. Description of the Drawings

[0049] Figure 1 Shows a schematic example diagram of a numerically controlled machine tool.

[0050] Figure 2 Shows a schematic example diagram of a numerically controlled machine tool control device according to an exemplary embodiment of the present invention.

[0051] Figure 3 Shows an exemplary flowchart of a method for monitoring a machine tool according to an exemplary embodiment of the present invention. Detailed Description of the Invention

[0052] Hereinafter, examples and exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. Identical or similar elements in the drawings may be denoted by the same reference signs, but sometimes also by different reference signs.

[0053] However, it should be emphasized that the present invention is not limited to or restricted by the exemplary embodiments and their implementation features described below, but further includes modifications to the exemplary embodiments, especially those that are included within the scope of protection of the independent claims by modifying the features of the examples or by combining one or more individual features of the examples.

[0054] Figure 1 Shows a schematic example diagram of a numerically controlled machine tool 100 configured, for example, as a milling machine.

[0055] However, the present invention is not limited to milling machines and can also be used for other types of machine tools, such as cutting machine tools for workpiece machining by, for example, drilling, milling, turning, grinding, such as milling machines, universal milling machines, lathes, turning centers, automatic lathes, milling / lathe machines, machining centers, grinders, gear cutting machines, etc.

[0056] The machine tool 100 includes, for example, a frame including a machine body 101 and a machine base 102. For example, a movable machine slide 105 is mounted on the machine body 101 to horizontally move on the machine body 101 in the Z direction (Z-axis). A workpiece WP is, for example, clamped on the machine slide 105, and the machine slide 105 may, for example, include a workpiece table. To this end, a clamping device may also be provided on the machine slide 105 or the tool table. In addition, in a further exemplary embodiment, the machine slide 105 may include a turntable that can rotate or pivot about a vertical axis and / or another horizontal axis (optional rotary or round and / or swivel axis). In addition (or alternatively), the machine slide 105 can move in the horizontal Y direction (potentially perpendicular to the drawing plane) by the Y-axis.

[0057] For example, the machine base 102 is provided with a spindle bracket slider 103 which can move vertically in the X direction on the machine base 102, and for example, the working spindle 104 carries a cutting tool. The working spindle 104 is configured to drive a cutting tool WZ (for example, a drilling and / or milling tool) received on the working spindle 104 to rotate about the main axis SA. For example, the spindle bracket slider 103 can move vertically in the X direction by the X-axis. In addition (or alternatively), the spindle bracket slider 103 can move in the horizontal Y direction (potentially perpendicular to the drawing plane) by the Y-axis. Further, in a further exemplary embodiment, the spindle bracket slider 103 can include a rotating and / or swiveling axis to rotate or swivel the spindle 104 (optional rotating axis or circular axis and / or swiveling axis).

[0058] In addition, the machine tool 100 includes a control device 200 for example for the operation of an operator of the machine tool 100. The control device 200 includes for example a screen 210 (for example, configured as a touch screen) and an input unit 220. For example, the input unit 220 can include means for user input or for receiving user commands or command actions of the operator, such as buttons, sliders, rotary control devices, keyboards, switches, mice, trackballs, and possibly one or more touch-sensitive surfaces (for example, a touch screen that can be combined with the screen 210).

[0059] The operator can use the control device 200 to control the operation of the machine tool or the machining process on the machine tool, and can also monitor the working state of the machine tool 100 or the machining process during machining.

[0060] Figure 2 A schematic example diagram of the control device 200 of the numerical control machine tool 100 according to an exemplary embodiment of the present invention is shown.

[0061] The machine tool 100 (for example, a machine tool according to Figure 1 ) includes a plurality of actuators 110 (for example, spindle drives, axis drives, etc.) of the machine tool 100 controlled by the control device 200 and optionally a plurality of sensors 120 for outputting sensor signals related to the machine state of the machine tool 100 to the control device 200.

[0062] The actuator 110 can include for example a drive for controllable linear axes and circular axes (rotating and / or swiveling axes) for controlling the relative movement between the cutting tool and the workpiece, and also includes a drive for a cutting tool-carrying working spindle (for example, on a milling machine) or a workpiece-carrying working spindle (for example, on a lathe).

[0063] In addition, the actuator 110 may include electronic, hydraulic, and / or pneumatic control valves, pumps, or other supply devices for internal or external coolant supply or compressed air supply. The conveying device, pallet changer, workpiece changer, tool magazine, and other machine tools can also be controlled by a driver, circuit, or corresponding actuator.

[0064] The optional sensor 120 can be, for example, a sensor associated with various components or parts of the machine tool (such as axes, drivers, bearings, spindles, spindle bearings, tool magazines, tool changers, pallets or workpiece changers, internal or external coolant supply devices, chip conveyors, and / or hydraulic and / or pneumatic control devices). A large number of different sensors can be provided for each component, such as position measurement sensors, current and / or voltage measurement sensors, temperature sensors, force sensors, acceleration sensors, vibration sensors, bearing diagnostic sensors, displacement sensors, liquid level indication sensors, liquid sensors (such as sensors for measuring the pH value in the coolant, water ratio measurement sensors for oil, coolant, etc.), water content sensors in the pneumatic system, and / or filter status sensors.

[0065] The control unit 200 (control device) includes, for example, a machine control 230 with an NC control 231 and a programmable logic control 232 (also known as SPS or PLC as "programmable logic control").

[0066] In addition, the control unit 200 includes, for example, a human-machine interface 240 (also known as HMI as "human-machine interface"), which can be used by the operator of the machine tool 100 to control, monitor, and / or operate the machine tool 100.

[0067] The human-machine interface 240 includes, for example, a graphical user interface 241 (also known as GUI as "graphical user interface") that can be displayed on a monitor or touch screen (such as on the screen 210) and an input / output unit 242 (which can include an input unit 220).

[0068] In addition, the control unit 200 includes, for example, a monitoring unit 250 for monitoring the working state of the machine tool 100 or for monitoring the machining process on the machine tool 100.

[0069] For example, the monitoring unit 250 can be implemented on a computer connected to the NC control 231. In addition, the monitoring unit 250 and the NC control 231 can be implemented together on the computer of the control unit 200.

[0070] For example, the monitoring unit 250 of the control unit 200 includes a processor 251 (CPU) for data processing and application execution and a data storage device 252 for storing data and programs executable by the processor 251 and application programs.

[0071] In addition, the monitoring unit 250 includes, for example, a communication interface 256 for communicating or exchanging data with an external data processing device such as a server (for example, via a local or global communication network or possibly also via a WLAN).

[0072] For example, an artificial or computer-implemented neural network NN is formed on the monitoring unit 250. For example, the data storage device 252 includes a monitoring application 253 executable by the processor 251 and includes the neural network NN.

[0073] For example, the data storage device 252 further includes a storage unit for storing configuration data 254 for the monitoring unit 250 and a storage unit for storing training data 255 of the neural network NN for the monitoring application 253.

[0074] For example, via the communication system of the control unit 200, the monitoring unit 250 is configured to read or receive data from the machine control 230, in particular to read or receive data from the NC control 231 and / or the programmable logic control 232.

[0075] In a preferred exemplary embodiment, the monitoring unit 250 can provide the data read or received from the machine control 230 as input data to the neural network NN of the monitoring application 253. For example, this can be performed continuously, regularly or periodically and preferably in real time.

[0076] Here, the drive data (for example, motor current and drive signal) and / or position data from the machine control 230 are preferably provided as input data to the neural network NN of the monitoring application 253.

[0077] In addition, sensor signals from the machine tool sensors (for example, from drive control, current control, speed control, position control and / or position detection) optionally available at the machine tool control can be provided as input data to the neural network NN of the monitoring application 253. The advantage of this is that no special sensors need to be provided for process monitoring, but rather some or all of the data and information available on the machine control can be used directly.

[0078] The input data is preferably fed in real time (i.e., particularly preferably during workpiece machining, especially regularly, periodically or continuously) to the neural network NN of the monitoring application 253.

[0079] The neural network NN of the monitoring application 253 is configured to output output data indicating the working state of the machine tool 100, particularly during workpiece machining, based on the input data (for example, input data read from the machine control or input data provided by the machine control).

[0080] The neural network may have been pre-trained based on a simulated machining process on a virtual machine tool bed and / or based on training on a machine tool of the same type or the same structure, and / or may have been further trained during the operation of the machine tool.

[0081] In a preferred exemplary embodiment, the monitoring unit 250 of the monitoring application 253 or the neural network NN is configured to read input data from the machine control 230 during the ongoing machining process of machining a workpiece on the machine tool 100, particularly in real time, and output output data indicating the current working state of the machine tool.

[0082] In a preferred exemplary embodiment, the monitoring unit 250 of the monitoring application 253 or the neural network NN is configured to evaluate the output data of the neural network NN.

[0083] In a preferred exemplary embodiment, the monitoring unit 250 of the monitoring application 253 or the neural network NN is configured to evaluate the output data of the neural network NN for tool monitoring and / or process monitoring on the machine tool 100.

[0084] In a preferred exemplary embodiment, the monitoring unit 250 of the monitoring application 253 or the neural network NN is configured to output control data affecting the machining process to the machine control based on the input data and / or based on the evaluation of the output data.

[0085] In a preferred exemplary embodiment, the monitoring unit 250 of the monitoring application 253 or the neural network NN is configured to determine when an abnormal working state of the machine tool 100 exists based on the input data and / or based on the evaluation of the output data.

[0086] In a preferred exemplary embodiment, the monitoring unit 250 of the monitoring application 253 or the neural network NN is configured to determine when the probability of an abnormal working state of the machine tool 100 occurring exceeds a predetermined limit value based on the input data and / or based on the evaluation of the output data.

[0087] In a preferred exemplary embodiment, the monitoring unit 250 of the monitoring application 253 or the neural network NN is configured to output control data affecting the machining process to the machine tool control 230 when it is determined that there is an abnormal working state of the machine tool 100 and / or the probability of an abnormal working state of the machine tool occurring exceeds a predetermined limit value, for example to trigger a process interruption, a machine stop (e.g., if a collision occurring on the machine tool is identified or determined to be an abnormal working state), a drive stop, a tool change (e.g., if a tool breakage occurring in the machine tool is identified or determined to be an abnormal working state, or if an increase in tool wear is identified or determined, or if a tool absence is identified).

[0088] Specifically, the monitoring unit 250 of the monitoring application 253 or the neural network NN may preferably be configured to send control data for triggering a machine stop to the machine control when it is determined that a collision exists and / or the possibility of a collision occurring on the machine tool 100 exceeds a predetermined limit value 230, in particular such that the machine control 230 triggers a machine stop on the machine tool 100 based on the control data.

[0089] Optionally or further, the monitoring unit 250 of the monitoring application 253 or the neural network NN may preferably be configured to output control data for triggering an automatic tool change to the machine tool control 230 when it is determined that there is a tool breakage or an increase in tool wear and / or the possibility of a tool breakage occurring on the machine tool 100 exceeds a predetermined limit value, in particular such that the machine control 230 triggers an automatic tool change on the machine 100 based on the control data.

[0090] Optionally or further, the monitoring unit 250 of the monitoring application 253 or the neural network NN may preferably be configured to adjust the process parameters and / or control data existing on the machine control 230 based on input data and / or based on an evaluation of output data, for example, in such a way that the machining speed of the machining process is adjusted, in particular by adjusting the feed rate and / or rotational speed occurring during the machining process. This can be particularly advantageous when data on excessive bearing wear and / or misalignment is detected.

[0091] In some preferred exemplary embodiments, the input data for the neural network NN obtained from the machine control 230 may specify the operating parameters of the actuator 110 or the drive of the machine tool 100, particularly preferably the drive speed, motor current, and / or the execution signal output to the actuator.

[0092] In some preferred exemplary embodiments, the input data for the neural network NN obtained from the machine control 230 may specify the position values of the movable parts of the machine tool, particularly the actual position and / or target position of the linear, circular, and / or rotary axes of the machine tool 100.

[0093] In some preferred exemplary embodiments, the input data for the neural network NN obtained from the machine control 230 may optionally specify the sensor values from the sensors 120 of the machine tool 100, particularly preferably the sensor values from temperature sensors, force sensors, strain sensors, torque sensors, acceleration sensors, oscillation or vibration sensors, and / or structure-borne sound sensors.

[0094] Figure 3 An exemplary flowchart of a method for monitoring a machine tool according to an exemplary embodiment of the present invention is shown.

[0095] In step S301, the machining process on the machine tool 100 is controlled by way of example (e.g., machining a workpiece on the machine tool, e.g., based on control data such as an NC program).

[0096] In step S302, data or information available at the machine control 230 is read as input data for the neural network NN. This is preferably done continuously or regularly or periodically during the machine machining process.

[0097] In step S303, the input data read from the machine control 230 is input into the neural network NN of the monitoring unit 250. This is preferably done continuously or regularly or periodically during the machine machining process.

[0098] In step S304, the output data of the neural network NN determined based on the input data is output.

[0099] In step S305, it is determined whether there is an abnormal operating state of the machine tool 100 based on the output data of the neural network NN output at the monitoring unit 250.

[0100] If the machine tool 100 does not have an abnormal operating state, the machining process continues without interruption (see step S301).

[0101] If, based on the output data of the neural network NN at the monitoring unit 250, it is determined or detected that an abnormal operating state of the machine tool 100 exists or has occurred, then in step S306, a corresponding machine tool response is triggered, e.g., by the monitoring unit 250 outputting a signal to the machine control 230 to trigger a corresponding machine response.

[0102] A corresponding allocated machine response can be triggered according to the output data or the detected abnormal operating state.

[0103] For example, if a collision is detected, the machine can be directly stopped immediately.

[0104] For example, if tool breakage or excessive tool wear (worn tool) is detected, the machining process can be interrupted, and an automatic tool change can be triggered, e.g., inserting a corresponding tool of the same type, and the machining process can continue using the inserted tool of the same type after a very short downtime. The advantage of this is that the replacement of the worn relevant tool can be automatically performed optimally at the right time, so that the part quality and tool utilization rate can be optimally adjusted.

[0105] In addition, when a situation where maintenance or manual inspection of machine components is required (such as detecting bearing damage or excessive bearing wear) is detected, the machine stop or drive stop can be triggered.

[0106] In an exemplary embodiment, the neural network can be designed to be learnable in the case of a fuzzy system by means of a bidirectional transformation. Furthermore, training data can be fed to the neural network, where the training data can be generated based on the process monitoring of a simulated machine tool and / or based on other machine tools of the same machine tool type that are already in operation.

[0107] In summary, according to an exemplary embodiment of the present invention, it is proposed to perform tool or process monitoring of a machine tool by providing an artificial or computer-implemented neural network on the control device of the machine tool. Such tool or process monitoring can advantageously protect the machine and / or the workpiece from damage, ensure optimal tool use, and provide a starting point for process optimization if necessary.

[0108] Therefore, productivity can be increased and the overall life cycle cost of the machine tool can be reduced. At the same time, tool and process monitoring contributes to quality assurance and allows workpiece control and quality as well as program documentation.

[0109] Advantageously, direct error or problem detection (such as collision detection, breakage, wear or detection of lost tools) can be performed, and corresponding responses, such as machine stop or tool change, can be triggered automatically and immediately without delay. This also allows further cost reduction and scrap reduction, for example, through wear-related tool changes or bearing maintenance.

[0110] Furthermore, based on the output data of the neural network, the optimal machining speed can be achieved by continuously adjusting the process parameters. In addition, there is an advantage in that historical output data and historical training data can be evaluated, for example, for recording and statistics.

[0111] In summary, the following advantages can be achieved in particular: comprehensive protection of the machine, workpiece and tool, real-time monitoring, optimal tool utilization, part quality monitoring (e.g., by recording and process analysis), reduction of scrap and / or adaptation to complex processes or machine machining.

[0112] Furthermore, an advantage in the exemplary embodiment is that since the data or information available at the machine control (e.g., drive data and / or positioning data) can be utilized, sensorless monitoring can also be provided, or at least some or additional sensors can be dispensed with. In a further exemplary embodiment, the input data of the neural network can be supported by or provided by sensor data, for example, provided by additional or alternative sensors (such as for strain, force, active power, torque, vibration, acceleration, structure-borne sound and / or temperature, etc.).

[0113] The examples and exemplary embodiments of the present invention and their advantages have been described in detail above with reference to the accompanying drawings. It should be emphasized again that the present invention is not limited or restricted in any way to the above exemplary embodiments and their implemented features, but further includes modifications of the exemplary embodiments, especially those that are included within the scope of protection of the independent claims by modifying the features of the examples or by combining one or more individual features of the examples.

Claims

1. A control device used on a numerically controlled machine tool, comprising: - A machine control unit for controlling an actuator of the machine tool for performing a machining process of a workpiece on the machine tool based on control data in the form of an NC program, and - A monitoring unit including a computer-implemented neural network for performing process monitoring on the machine tool, wherein the monitoring unit is configured to read input data from the machine control unit and output output data indicating the working state of the machine tool, and the neural network of the monitoring unit is configured to evaluate the output data indicating the working state of the machine tool for performing process monitoring on the machine tool in order to detect errors in the machining process, characterized in that the neural network of the monitoring unit is configured to read input data from the machine control unit during the process of machining a workpiece in progress and output output data indicating the current working state of the machine tool during workpiece machining; and the monitoring unit is configured to adjust process parameters of the machining process existing on the machine control unit based on an evaluation of the output data indicating the current working state of the machine tool by the neural network, the process parameters being related to workpiece machining by a machine tool working spindle carrying a tool or a workpiece.

2. The control device according to claim 1, characterized in that the monitoring unit is configured to adjust the process parameters of the machining process in such a way that the machining speed of the machining process is adjusted.

3. The control device according to claim 2, characterized in that the monitoring unit is configured to adjust the process parameters so as to adjust the machining speed of the machining process in such a way that the feed rate and / or rotational speed occurring during the machining process are adjusted.

4. The control device according to any one of claims 1 to 3, characterized in that the monitoring unit is configured to continuously adjust the process parameters of the machining process.

5. The control device according to any one of claims 1 to 3, characterized in that the neural network of the monitoring unit is configured to read input data from the machine control unit during the ongoing machining process of the workpiece performed on the machine tool and output output data indicating the current working state of the machine tool.

6. The control device according to any one of claims 1 to 3, characterized in that the monitoring unit is further configured to evaluate the output data indicating the current working state of the machine tool for performing tool monitoring on the machine tool.

7. The control device according to any one of claims 1 to 3, characterized in that the monitoring unit is configured to output control data affecting the machining process to the machine control unit based on the input data from the machine control unit and / or based on the evaluation of the output data indicating the current working state of the machine tool.

8. The control device according to any one of claims 1 to 3, characterized in that The monitoring unit is configured to determine an abnormal operating state of the machine tool based on the input data from the machine control unit and / or based on the evaluation of the output data indicating the current operating state of the machine tool, thereby detecting a collision occurring on the machine tool between a machine part and a tool and / or a workpiece and / or a tool breakage occurring on the machine tool and / or a loss of a tool used on the machine tool during the machining process and / or an increase in tool wear of a tool used on the machine tool during the machining process.

9. The control device according to claim 8, wherein the monitoring unit is configured to output control data affecting the machining process to the machine control unit when it is determined that an abnormal operating state of the machine tool exists.

10. The control device according to any one of claims 1 to 3, wherein the monitoring unit is configured to determine when the likelihood of an abnormal operating state of the machine tool occurring exceeds a predetermined limit value based on the input data from the machine control unit and / or based on the evaluation of the output data indicating the current operating state of the machine tool, thereby detecting a possible collision between a machine part and a tool and / or a workpiece and / or a possible tool breakage occurring on the machine tool and / or a possible loss of a tool used on the machine tool during the machining process and / or tool wear exceeding the tool limit value of the tool used during the machining process.

11. The control device according to claim 10, wherein the monitoring unit is configured to output control data affecting the machining process to the machine control unit when it is determined that an abnormal operating state of the machine tool exists and / or the likelihood of an abnormal operating state of the machine tool occurring exceeds a predetermined limit value.

12. The control device according to claim 9, wherein the monitoring unit is configured to determine when the likelihood of an abnormal operating state of the machine tool occurring exceeds a predetermined limit value based on the input data from the machine control unit and / or based on the evaluation of the output data indicating the current operating state of the machine tool, thereby detecting a possible collision between a machine part and a tool and / or a workpiece and / or a possible tool breakage occurring on the machine tool and / or a possible loss of a tool used on the machine tool during the machining process and / or tool wear exceeding the tool limit value of the tool used during the machining process, and the monitoring unit is configured to output control data affecting the machining process to the machine control unit when it is determined that an abnormal operating state of the machine tool exists and / or the likelihood of an abnormal operating state of the machine tool occurring exceeds a predetermined limit value, and the monitoring unit is configured to output control data triggering the stoppage of the machine tool to the machine control unit when it is determined that an abnormal operating state of the machine tool exists and / or the likelihood of an abnormal operating state of the machine tool occurring exceeds a predetermined limit value.

13. The control device according to claim 9, wherein The monitoring unit is configured to determine when the likelihood of an abnormal operating state of the machine tool exceeds a predetermined limit value based on the input data from the machine control unit and / or based on the evaluation of the output data indicating the current operating state of the machine tool, thereby detecting possible collisions between machine parts and the tool and / or the workpiece, and / or possible tool breakage on the machine tool, and / or possible loss of the tool used on the machine tool during the machining process, and / or tool wear exceeding the tool limit value used during the machining process, and the monitoring unit is configured to output control data affecting the machining process to the machine control unit when it is determined that there is an abnormal operating state of the machine tool and / or the likelihood of an abnormal operating state of the machine tool exceeds a predetermined limit value, and the monitoring unit is configured to output control data triggering an automatic tool change to the machine control unit when it is determined that an abnormal operating state of the machine tool exists and / or the likelihood of an abnormal operating state of the machine tool exceeds a predetermined limit value.

14. The control device according to any one of claims 1 to 3, characterized in that the monitoring unit is configured to adjust the control data present at the machine control unit based on the input data from the machine control unit and / or based on the evaluation of the output data indicating the current operating state of the machine tool.

15. The control device according to any one of claims 1 to 3, characterized in that the input data from the machine control unit indicates the operating parameters of the actuators or drives of the machine tool.

16. The control device according to claim 15, characterized in that the operating parameters of the actuators or drives of the machine tool are drive speed, motor current, and / or an execution signal output to the actuator.

17. The control device according to any one of claims 1 to 3, characterized in that the input data from the machine control unit indicates the position values of the movable parts of the machine tool.

18. The control device according to claim 17, characterized in that the position values are the actual and / or target positions of the linear, rotational, and / or swivel axes of the machine tool.

19. The control device according to any one of claims 1 to 3, characterized in that the input data from the machine control unit indicates sensor values from sensors of the machine tool.

20. The control device according to claim 19, characterized in that the sensor values from the sensors of the machine tool are sensor values from temperature sensors, force sensors, strain sensors, torque sensors, acceleration sensors, oscillation or vibration sensors, and / or structure-borne sound sensors.

21. The control device according to any one of claims 1 to 3, characterized in that the monitoring unit includes an internal data storage device and / or is configured to communicate with an external data storage device, Wherein, the monitoring unit is configured to store the input data from the machine control unit and / or the output data indicating the current working state of the machine tool in the internal and / or external data storage device, and / or Wherein, the monitoring unit is configured to store the evaluation data generated based on the evaluation of the input data from the machine control unit and / or the evaluation of the output data indicating the current working state of the machine tool in the internal and / or external data storage device.

22. The control device according to any one of claims 1 to 3, characterized in that the control device further includes a human-machine interface operable by an operator, wherein, the monitoring unit is configured to output, on the human-machine interface, the output data from the neural network indicating the current working state of the machine tool to the operator.

23. A machine tool, comprising the control device according to any one of claims 1 to 22.

24. A method for monitoring a numerically controlled machine tool, the method comprising: - Controlling, by a machine control unit, an actuator for executing a machining process for machining a workpiece on the machine tool based on control data in the form of an NC program; - During the control of the actuator for executing the machining process of the workpiece on the machine tool, monitoring, by a monitoring unit for process monitoring, the machining process of the workpiece executed on the machine tool, wherein the monitoring includes: - Reading input data from the machine control unit into a computer-implemented neural network of the monitoring unit during the ongoing machining process; - Outputting output data from the neural network of the monitoring unit indicating the current working state of the machine tool during workpiece machining; - Evaluating, by the neural network of the monitoring unit, the output data indicating the current working state for process monitoring on the machine tool; - Adjusting process parameters of the machining process based on the output data evaluated by the neural network of the monitoring unit, the process parameters being present at the machine control unit and related to machining the workpiece by a machine tool spindle carrying a tool or a workpiece.

25. A computer program product, comprising instructions that, when executed by a computer connected to the numerically controlled machine tool according to claim 23 or the control device according to any one of claims 1 to 22 of the numerically controlled machine tool, cause the computer to execute the method according to claim 24.

Images