Method for operating and controlling a machine in a production cell by means of an operating device, and machine having such an operating device

Abstract

The invention relates to a method for operating and controlling a machine, in particular an injection moulding machine, in a production cell by means of an operating device. In the proposed method, the operating device provides a user with various function blocks for selection, and the user configures the selected function block by choosing, from among a selection of different signal outputs and signal inputs, at least one of each for connection to a function block input and a function block output. The invention also relates to a machine having such an operating device, which is designed to carry out the proposed method, and to a computer program for a processor of the operating device comprising commands for carrying out the proposed method, and to a computer-readable data carrier on which the computer program is stored, as well as to a data carrier signal which the computer program transmits to the machine.

Description

[0001] Method for operating and controlling a machine of a production cell by means of an operating device and a machine with such an operating device

[0002] Description

[0003] TECHNICAL FIELD OF THE INVENTION

[0004] The invention relates to a method for operating and controlling a machine, in particular an injection molding machine, and a production cell by means of an operating device. Furthermore, the invention relates to a machine, in particular an injection molding machine, with such an operating device, which is designed to carry out the method according to the invention, as well as to a computer program for a processor of the operating device of the machine, comprising instructions for carrying out the method according to the invention, and a computer-readable data carrier on which the computer program is stored, as well as a data carrier signal that transmits the computer program to the machine or to the operating device.

[0005] BACKGROUND OF THE INVENTION

[0006] (Production / manufacturing) machines for producing parts, such as injection molding machines for producing injection-molded parts (also called injection molded parts), have a machine control system for automatically controlling the machine and usually also an operating device with a human-machine interface for the user to enter settings (and control commands) for the machine control system. The machine control system is developed by the machine manufacturer and is therefore largely predetermined for the user of the machine. The user can typically only make limited settings on the machine using the operating device and can therefore only influence the production process for the parts desired by the user within narrow limits, for example in order to optimize or improve it.This often forces the user to have their change requests implemented by the machine manufacturer, who then incorporates additional functionality into the machine control system or integrates it into its program (software / firmware). However, this has the disadvantage that the machine user may have to share their operating knowledge with the machine manufacturer, which they would prefer to keep private (and thus confidential). Conversely, the machine manufacturer does not want to constantly implement user- or customer-specific changes to the machine control system and thus have to maintain and support a multitude of different machine control system variants.

[0007] Furthermore, production cells for manufacturing parts often comprise several machines, so that in addition to the main machine, such as an injection molding machine, additional (auxiliary / supplementary) machines must be provided for manufacturing the parts. These include conveyor machines to feed raw materials for production into the main machine and to transport the finished parts away from the main machine, but also processing machines which, for example, prepare the raw materials, and testing machines which check the finished parts for defects. It would often be useful if the main machine (or its control system) could receive inputs, such as measured values, from these supplied (auxiliary / supplementary) machines and could output specifications / set values ​​to these additional machines. This is not possible with a conventional machine control system for the main machine of a production cell with a standard operating device.

[0008] Therefore, in the field of machines with a machine control and operating device, particularly in the field of injection molding machines, and especially for (production) machines that are part of a production cell, there is a need for (alternative) methods and devices that allow the user of the machine to control it more extensively and operate it according to specific needs and requirements than is possible with currently known means, without the user having to contact the manufacturer to make the desired adjustments, which could potentially involve the undesirable disclosure of internal operating knowledge or even trade secrets from the user to the machine manufacturer. SUMMARY OF THE INVENTION

[0009] One object of the present invention is therefore to provide a method for operating and controlling a machine that allows the user of the machine to control the machine more extensively than with previously known means and to operate it according to his needs. This object is achieved by the method specified in claim 1.

[0010] A further object of the invention is to provide a use of the proposed method that allows reducing the number and frequency of user inputs. This is achieved by the use specified in claim 14.

[0011] Using the proposed method, a machine can be equipped and operated that then implements the advantages of the invention in production. Such a machine is defined in claim 15.

[0012] Furthermore, a corresponding computer program for implementing the proposed method and a computer-readable data carrier on which the computer program is stored, as well as a data carrier signal that transmits the computer program to the machine, form part of the present invention. They are specified in claims 17 and 18.

[0013] Variants of the invention are set out in the dependent claims.

[0014] In the embodiments of the invention listed below, the features enclosed in brackets are to be understood as optional.

[0015] The proposed method is used for operating and controlling a machine, in particular an injection molding machine, in a production cell. The machine comprises a machine controller for automatically controlling the machine and an operating device with a human-machine interface for the input of settings and / or control commands for the machine control by a user. The machine controller has several control inputs for transferring the settings and / or control commands from the operating device to the machine controller. In addition, various signal sources (with signal outputs) can be present, each of which can be connected to the operating device.The method comprises the following steps: making various function blocks available for selection by the operating device on a display of the human-machine interface, wherein the function blocks each comprise at least one function block input, at least one function block output and a transformation function with which at least one input signal or input value at at least one function block input is transformed into at least one output signal or output value at at least one function block output;.

[0016] Selection of one of the available function blocks (with input means or (a) control element(s) of the human-machine interface) by the user; and

[0017] Configuring the selected function block (using the input means or control element(s) of the human-machine interface) by the user, whereby the configuration comprises the following steps:

[0018] • to select the various signal sources or various default settings by the control device on the display for connection to the at least one function block input;

[0019] • Selection of at least one of the signal sources available for selection (using the input means or control element(s) of the human-machine interface) by the user;

[0020] • to select the plurality of control inputs of the machine control by the operating device on the display for connection to the at least one function block output; and

[0021] • Selection of at least one of the control inputs available for selection (using the input means or control element(s) of the human-machine interface) by the user.

[0022] It should be noted here that the control device can be located directly on the machine or remotely, e.g., in a remote control room. Furthermore, remote access to the control device by the user is also possible, whether it is located directly on the machine or remotely. This means that the machine, the control device, and the user can all be located in different places. However, the control device is always considered part of the machine.

[0023] It should also be noted at this point that the user can be either a human person or a machine with artificial intelligence, which has been trained, in particular, with knowledge (e.g., training data) from trained and experienced human users. In one embodiment, the method further comprises parameterizing the transformation function of the selected function block with one or more transformation parameters (using the input means or the control element(s) of the human-machine interface) by the user, wherein the parameterization comprises assigning a value to one or more of the one or more transformation parameters.

[0024] In a further embodiment of the method, the signal sources are one or more signal outputs of the machine and / or one or more signal outputs of another machine in the production cell and / or a signal output of one or more sensors in the production cell that do not belong to the machine.

[0025] Preferably, the “further machine” is at least one piece of equipment of the production cell and / or auxiliary device of the production cell that is upstream and downstream of the machine of the production cell.

[0026] A "further machine" could also be (at least) one machine from a different production cell connected to the production cell. "Connected" refers in particular to a (particularly permanent) communication connection for data exchange and / or a connection for receiving and / or transmitting control signals. For example, two injection molding machines could be connected to optimize (smooth) energy consumption upon startup.

[0027] In a further embodiment of the method, in addition to making the plurality of control inputs of the machine control available for selection, at least one further control input of one or more machines of the production cell is made available for selection by the operating device on the display for connection to the at least one function block output in order to influence a machine control of the further machine by the user selecting the at least one further control input.

[0028] In a further embodiment, the method further comprises a physical connection of the selected signal output(s) of a further machine and / or the selected sensor(s) not belonging to the machine and / or the selected further control input(s) of the further machine to the operating device (or to the machine). In a further embodiment of the method, a function block is the following: a controller, in particular a P, a PI, a PD or a PID controller, further in particular a linear or non-linear controller, further in particular a fuzzy logic controller; a feedforward control, i.e. an element that applies a value to a manipulated variable that is independent of the states of a controlled system and the measurements resulting therefrom; a mathematical function, such asa sine or exponential function; a logic block, in particular a stateless or stateful logic block; a state machine; a lookup table; a neural network, in particular a trained or self-learning neural network.

[0029] In a further embodiment, the method further comprises executing a program code assigned to the selected function block and / or activating a processing unit assigned to the selected function block, in particular in the operating device, in particular without changing a program code (or software / firmware) of the machine control.

[0030] In a further embodiment of the method, an update period of the selected function block or a time constant of at least one of the transformation parameters ranges from one minute to several hours, in particular up to one hour, in particular up to 30 minutes, in particular up to 10 minutes. The update period refers to a period of time between the output of two consecutive output values ​​of a function block.

[0031] It should be noted that the update period is many times greater or longer than the time constant of the machine control itself, which lasts a few seconds or less than one second. Preferably, the time constant is less than one millisecond, preferably less than half a millisecond, and particularly preferably, the time constant is at most 250 ps. The time constant refers to the period between the output of two consecutive control commands / values ​​from the machine control. An important term in production / manufacturing technology is "cycle time," which refers to the time the machine, e.g., an injection molding machine, requires to produce or manufacture a part (with several parts often being manufactured in parallel during a production cycle).The update period is longer than the cycle time, which in turn is longer than the time constants of the machine control.

[0032] The control system triggered by user inputs can be viewed as a slow "superordinate" or "concurrent" control or regulation of the process cell. With this control system, the user can influence the operation of the machine, such as the injection molding machine, but also control the control of attached (auxiliary or additional) machines that form part of the process cell.

[0033] In a further embodiment of the method, an actual value or actual state is present at the function block input and / or a setpoint or target value or setpoint or target state is present at the function block output, in particular the default setting is an actual value or actual state at the function block input.

[0034] In a further embodiment of the method, the transformation function comprises at least one scaling and / or an offset, in particular an input scaling of the at least one input signal or input value and / or an addition of an input offset and / or an output scaling of the at least one output signal or output value and / or an addition of an output offset. The offset can be positive or negative. The scaling and the addition of an offset serve to optimally adapt the input signal to the input range of the transformation function and to optimally adapt the output signal of the transformation function to the input range of the subsequent unit.

[0035] In a further embodiment of the method, the at least one output signal or the at least one output value is validated to ensure that the at least one output signal or the at least one output value does not lead to any malfunction or damage to the machine or the further machine.

[0036] In a further embodiment of the method, the at least one output signal or the at least one output value is limited, in particular depending on the result of the aforementioned validation. Furthermore, the at least one input signal or input value can also be limited.

[0037] The aforementioned scaling and / or addition of an offset can also be performed depending on the result of the aforementioned validation. In a further embodiment of the method, the at least one output signal or the at least one output value is logged, in particular by periodically saving a sample of the at least one output signal or the at least one output value, in particular together with a time specification, in particular once per update period. The log containing this data can be used, in particular in the event of a fault, to determine the cause.

[0038] The above-mentioned machine with artificial intelligence can be a setting device, in particular a processor-based one, which is preferably a (permanent) component of the machine and / or the production cell and / or the operating device and / or is (at least temporarily or always) in communication with the operating device and / or can be brought into communication with it.

[0039] The setting device is (at least partially automatically and preferably automatically) suitable, determined and / or configured to select (within the scope of a computer-implemented method step) at least one functional block available for selection, and / or (within the scope of a computer-implemented method step) at least one signal source available for selection and / or to select at least one control input available for selection.

[0040] Preferably, the setting device is suitable and intended to carry out (within the scope of a computer-implemented method step) a parameterization of the transformation function of the selected function block with one or more transformation parameters.

[0041] Preferably, the setting device is suitable and intended (within the scope of a computer-implemented method step), in particular additionally, to select at least one further control input of one or more machines of the production cell, whereby a machine control of the further machine can be effected.

[0042] It is conceivable that the setting device implements the selection itself. Alternatively, the setting device can suggest and / or output the selection made to an operator and request approval for implementation before the setting device implements the selection. Preferably, the setting device makes at least one of the above-mentioned selections depending on a, in particular trainable, machine learning operating setting model, which comprises a set of, in particular, trainable parameters that are set to values ​​learned as a result of a (in particular completed) training process. The training process is carried out on the basis of training data or a plurality of training data sets.

[0043] Preferably, the setting device determines user inputs entered by an operator as a user on the operating device for operating and / or controlling the machine (in particular on the human-machine interface for entering settings and / or control commands for the machine control and / or on an input device) and preferably an associated time stamp (which is characteristic of a time of the user input).

[0044] To determine the user inputs, the setting device for the user inputs can receive characteristic user input variables (by transmission from the operating device) and / or retrieve them (from a storage device) and / or record them (even when they are entered).

[0045] Preferably, the training data comprises a plurality of training data sets, wherein a training data set comprises user inputs (determined by the setting device) or user input variables characteristic thereof and preferably a respective time stamp (associated with the user input).

[0046] Preferably, the training data sets (each) comprise at least one production parameter characteristic of at least one production step of the machine and / or another machine and / or the production cell, and preferably a plurality of production parameters.

[0047] Preferably, the training data sets (each) comprise a plurality of (at least one and preferably several) production parameters, which are each characteristic of at least one production step of the machine and / or another machine and / or the production cell.

[0048] The production parameter is preferably a variable that can be specified by an operator (in particular on the operating device) of the machine and / or the production cell, depending on which the at least one production step can be controlled and / or regulated.

[0049] The at least one production parameter and preferably the plurality of production parameters can be selected from a group of production parameters which include a closing force of a tool of an injection molding machine, a target breathing of a tool of the injection molding machine, a column extension, a heating temperature or heating power at the plasticizing of the injection molding machine, a heating power of a heating element (for heating the material used to produce the injection molded part), a processing temperature of the starting material or materials used to produce the injection molded parts to be produced by the injection molding machine.Starting materials, a (target) injection speed and / or an injection pressure and / or an injection volume of the material used to produce the injection-molded parts into the respective cavities, a (target) injection work, at least one material variable characteristic of the material used to produce the injection-molded parts (such as the composition of a plastic, recycled content), a torque and / or a rotational speed of a plasticizing screw and / or a drive motor of the plasticizing screw or a variable characteristic thereof, a conveying behavior of the material used to produce the injection-molded parts (granules, powder, flakes and the like) or a variable characteristic thereof, such as a coefficient of friction or a bulk density, a variable characteristic of a conveying behavior of the melt produced in a plasticizing screw, such as rheological variables, e.g.Viscosity of the plastic, a variable characteristic of a (target) cooling temperature and / or cooling capacity of a tool, internal tool pressures of the melt in the cavity, temperatures of the tool surface in the cavity, a system pressure of a basic supply of a hydraulic component of the injection molding machine, an inspection parameter of an optical inspection device (in particular downstream of the injection molding machine) for inspecting the produced injection molded part (in particular for determining a variable characteristic of a quality of the injection molded part and / or the production process), a drying variable characteristic of a drying process (for example of the material to be used to produce the injection molded part, such as plastic granulate) and / or a variable characteristic of a cooling / drying process of the produced injection molded part and the like as well as combinations thereof.

[0050] The production parameters can refer to parameters of the production process of the injection molding machine and / or another machine (such as a peripheral system of the injection molding machine that is in communication with the machine control system (MES, abbreviation for Manufacturing Execution System)).

[0051] Additionally or alternatively, the training data sets may comprise at least one and preferably a plurality of machine parameters which are characteristic of at least one machine and / or of a plurality of the machines of the production cell and / or of the production cell.

[0052] The at least one machine parameter and preferably the plurality of machine parameters can be selected from a group of machine parameters, which include a number of cavities of the production cell currently used and / or ready for use to produce the injection-molded parts, a maximum number of cavities of the production cell that can be used to produce the injection-molded parts, an age, the tool configuration in injection-molded machines with interchangeable inserts, an age, a shot count and / or a maintenance status of the tools and / or machines and / or machine parts used to produce the injection-molded parts, a machine type and / or tool type of the tools and / or machines and / or machine parts used to produce the injection-molded parts, in particular a type (of) plasticizing screw, the number and flow cross-section of the venting channels of a cavity of the injection-molded tool,Type of cooling medium for cooling the tool used in the injection molding machine (e.g. water / oil), number of plasticizations, number of core pulls and the like, as well as combinations thereof.

[0053] The selection of the at least one machine parameter or the plurality of machine parameters (particularly from the above group) can depend on the application. The machine parameters can vary depending on which injection-molded part is being molded. Therefore, different machine parameters can be specified (and / or can be specified) for different types of injection-molded parts to be produced. Preferably, at least two, preferably at least three different types of injection-molded parts to be produced are specified, and at least one machine parameter and preferably a plurality of machine parameters are assigned to each of these types.

[0054] Additionally or alternatively, the training data sets may comprise at least one and preferably a plurality of injection-molded part parameters which are characteristic of an injection-molded part to be produced. The at least one injection-molded part parameter and preferably the plurality of injection-molded part parameters may be selected from a group of injection-molded part parameters which include a mass and / or weight of the injection-molded part to be produced or a characteristic variable thereof, a crystallinity around the sprue of the injection-molded part to be produced or a characteristic variable thereof, the presence of black specks, a formation of threaded parts or a characteristic variable thereof, shape and / or position tolerances, damage (such as flaws), characteristic mechanical and / or physical variables such as puncture resistance, impact strength, tensile strength of the injection-molded part to be produced or a characteristic variable(s) thereof,an effective area of ​​the injection-molded part to be produced and / or an effective area of ​​the cavity suitable for producing the injection-molded part or a size characteristic thereof, a size characteristic of a geometry of the injection-molded part to be produced, a size characteristic of a weight and / or a geometric extension of the injection-molded part to be produced and the like, as well as combinations thereof.

[0055] Preferably, the training data sets (each) comprise a plurality of sensor data measured by at least one sensor device of the machine and / or the production cell and preferably sensor data measured by a plurality of sensor devices of the machine and / or the production cell or data derived therefrom.

[0056] The sensor data and preferably the plurality of sensor data can be selected from a group of sensor data which includes an (actual) temperature of at least one area and preferably a plurality of areas of a tool of the injection molding machine, an (actual) temperature of at least one area and preferably a plurality of areas along the plasticizing screw, an injection pressure and / or an injection speed or variables characteristic thereof, a closing force, a closing speed or also an acceleration when closing a tool of the injection molding machine or a variable characteristic thereof, a viscosity of the material used to produce the injection-molded parts or a variable characteristic thereof, a variable characteristic of a breathing of the tool of the injection molding machine such as the column extension, an injection work (determined based on sensor data),an (actual) torque and / or rotational speed of a plasticizing screw and / or a drive motor of the plasticizing screw or a characteristic value thereof, a resistance during the injection process (which depends, for example, on a cross-section of the venting channels of the respective cavity of the injection molding machine), a feed and / or return temperature of a cooling medium for cooling the tool of the injection molding machine used to produce the injection-molded parts, a flow velocity of the cooling medium, a value characteristic of an ambient condition, such as ambient temperature, air humidity or altitude, a value characteristic of a property of the starting materials, such as granule moisture content or granule temperature, a value characteristic of a property of the finished injection-molded parts (determined, for example, by an inspection and / or testing device), such as mass or physical properties (inspection and / or test) value,a plasticizing performance or a characteristic variable thereof, a thermal and / or material homogenizing performance or a characteristic variable thereof, a conveying behavior of the material used to produce the injection-molded parts in a feed area of ​​the plasticizing screw, a compression variable characteristic of a compression behavior of the material used to produce the injection-molded parts in the plasticizing screw, a mold cavity pressure variable characteristic of a mold cavity pressure, and the like, as well as combinations thereof.

[0057] Additionally or alternatively, the training data sets can comprise at least one and preferably a plurality of machine-internal parameters that are characteristic of a machine-internal control and / or measurement variable. These machine-internal parameters are, in particular, parameters that are (usually) not set and / or monitored by a user, but which the machine control system determines for controlling and / or regulating production.

[0058] For example, a machine-internal parameter can be an (internal) parameter of the temperature control such as dead time and / or a slope with respect to a controller edge (temperature / time) and / or an injection pressure and / or an effective pressure during the injection process (of the material used to produce the injection-molded part into the cavity of the injection molding machine).

[0059] Additionally or alternatively, the training data sets may comprise at least one machine state and preferably a plurality of machine states of the machine and / or a machine of the production cell and / or a machine part of the production cell.

[0060] A machine status can be selected from a group of machine statuses, which includes a fault status, aging, a cleaning status (e.g., time of last cleaning or the like, or deposits detected in a ventilation duct) and / or maintenance status, a status of a heating element, a production status, a susceptibility to errors, and the like, as well as combinations thereof. It is also conceivable for the training data sets to additionally include data characteristic of a respective use case. A use case is understood to mean, for example, the manufacture of parts for a specific application (PET preforms, medical articles, beverage closures, and / or thin-walled articles) or manufacture using a special process (foaming, injection-compression molding, or the like).This advantageously makes it possible for the operating setting model to be trained during its training to assign the parameters (or combinations) and / or machine states associated with a usage case to the prevailing or selected usage case.

[0061] Preferably, the data used to generate the training data sets (production parameters, injection molding part parameters, machine parameters, sensor data, machine-internal parameters, machine status, ...) are stored on a non-volatile memory device (and retrieved therefrom to generate a training data set), preferably repeatedly and particularly preferably in (in particular predeterminable) regular time intervals.

[0062] Preferably, the (collection and / or determination and) storage of the data to be used to generate the training data sets on the non-volatile storage device is triggered by a user input from an operator (as user). User input data characteristic of the user input (such as the type of user input) are also stored (on the non-volatile storage device) and associated with the data (collected at the same time or within the same period). A training data set (with data associated with a common timestamp) can then be created from this data.It is also conceivable that data collected in a (specified and / or specifiable) period before the time of the user input and / or after the time of the user input with regard to the parameters (for which data are to be collected in order to generate a training data set) are also stored and used for further data analysis and / or for generating a training data set, so that a temporal progression of the values ​​of the corresponding parameters can advantageously be depicted.

[0063] It is preferably adjustable and / or specifiable and / or predetermined which data (of the above-mentioned parameters, such as production parameters, injection-molded part parameters, machine parameters, sensor data, machine-internal parameters, machine status) are collected and / or used to generate the training data sets. This can be specified, for example, by a manufacturer of the injection molding machine or the production cell. However, it is also conceivable that this can be specified and / or adjusted by an operator (as user).

[0064] Additionally or alternatively, it is possible that the type and / or number and / or selection of various parameters (e.g., those mentioned above) (e.g., production parameters, injection molding part parameters, machine parameters, sensor data, machine-internal parameters, machine status) which are used or are to be used (in particular in addition to the user input data) to generate a training data set and, in particular, are therefore to be collected (to generate a training data set), is carried out automatically (e.g., by the setting device).

[0065] For example, a large amount of data on the above-mentioned parameters can be collected and stored (together with a timestamp). The data sets can then be evaluated using pattern recognition and / or detection of deviations and / or anomalies and / or cluster analysis (e.g., using a k-means algorithm). This makes it possible to identify clusters / patterns of parameters in which these parameters assume values ​​from specific value ranges, and in which error states and / or malfunctions and / or (accelerated) aging and / or contamination states occur with increased / specified probabilities. In this way, critical parameters (constellations) to be monitored can be identified.

[0066] Likewise, parameter constellations can be determined in a corresponding manner (cluster / pattern determination) which result in particularly favorable or desirable production (e.g. particularly energy- and / or resource-saving production and / or particularly high production quality).

[0067] These determined parameters can then be used or (automatically) specified to generate the training data sets and then to generate the training data.

[0068] Preferably, the respective training data sets include, as a classification feature, the user input associated with the respective data set or characteristic data thereof and / or information as to whether the state of the production cell or machine corresponds to (fault-free) normal operation and / or whether a malfunction is present and / or whether the machine and / or production cell is in a (progressive) state of aging. Preferably, the training data sets and / or the data used to generate the training data sets are / are anonymized and / or pseudonymized so that the data used can no longer be (directly) traced back to a specific user.

[0069] In a preferred method, the training process for the operating setting model uses training data that includes data collected from the machine and / or the production cell and / or consists of these (own) collected data. It is also conceivable to use training data that was / are generated based on data collected in at least one and preferably several different (preferably identical) production cells.

[0070] For example, it could be a different production cell operated by the same operator. For example, training data generated from a production cell used by the same operator in previous years could be used. This offers the advantage that a large number of training data sets are available in a short period of time, allowing the production process to be adapted to the specific operator's production operations right from the start. This allows for the beneficial use of experience and knowledge gained in previous production cells immediately after the new production cell begins production.

[0071] However, it is also conceivable that training data sets (in particular exclusively and / or as part of a pre-training step) can be used which are / were generated on the basis of data collected from at least one production cell and preferably a plurality of (preferably identical) production cells by different operators than the operator of the (material) production cell. This offers the advantage that, for example, rarely occurring malfunctions and / or fault conditions of the production cell can be mapped in the training data, and the occurrence of such undesirable conditions can advantageously be prevented by training the operating setting model based on this training data.

[0072] The training data sets and / or the data used to generate the training data sets are preferably available in anonymized and / or pseudonymized form, so that the data used cannot be traced back to a specific operator and / or a specific production cell. In a preferred method, training data is used in the training process of the operating setting model, which includes historically recorded data, in particular for limit values ​​and / or states, limit values ​​for repeated achievement of limit values ​​and / or states, limit values ​​for a temporal change and / or a rate of change of values ​​characterizing the current state, statistical limit values ​​and / or states of the machine and / or the production cell.

[0073] The machine learning operating setting model is preferably based on an (artificial) neural network. The neural network is preferably a deep neural network (DNN), in which the parameterizable processing chain has a plurality of processing layers, and / or a so-called convolutional neural network (CNN) and / or a recurrent neural network (RNN).

[0074] Preferably, the data (to be processed), in particular the data / values ​​collected for the specified parameters (or selection of the above parameters) or variables derived therefrom, are fed to the operating setting model or the (artificial) neural network as input variables.

[0075] Preferably, the operating setting model or the artificial neural network maps the input variables to output variables depending on a parameterizable processing chain.

[0076] Preferably, characteristic variable(s) are selected as output variable(s) for (at least) one (above-mentioned) selection with respect to a function block and / or with respect to a signal source and / or with respect to (at least) one parameterization of a transformation function and / or with respect to (at least) one further control input of a further machine.

[0077] This allows for the replacement of operation by a (human) user, or advantageously, an operating setting can be predefined that an operator would perform with a given probability. This advantageously results in more user-friendly operation of the machine or production cell and, in particular, more precise and productive operation of the machine, since the operation / control and / or regulation is already performed by the setting device and / or suggested to the operator.

[0078] Additionally or alternatively, at least one variable is selected as the output variable which is characteristic of a manipulated variable for controlling and / or regulating a production step, for example the heating process of the material to be used to produce the injection-molded parts by means of controlling at least one heating element and / or the compression process and / or conveying process of the material to be used to produce the injection-molded parts and / or an actuation and / or a closing process of an opening (for example of the tool for producing the injection-molded parts), in particular via a servo and / or stepper motor, and / or the cooling process for cooling the tool to be used to produce the injection-molded parts, and / or a production parameter (preferably one of the above-mentioned).

[0079] It is also conceivable that (additionally) at least one machine state and / or a variable characteristic of a maintenance and / or a fault state of the machine and / or a machine part and / or the production cell is selected as the output variable.

[0080] Preferably, the machine learning operating setting model is / was trained using predefined training data (in particular as described above), whereby the training parameterizes the parameterizable processing chain. Preferably, a neural network trained in this way (as an operating setting model) is used by the setting device to determine a (above-mentioned) selection with respect to a function block and / or a signal source and / or parameterization of a transformation function and / or another control input of another machine.

[0081] Training is preferably carried out using supervised learning. However, it would also be possible to train the operating setting model or the artificial neural network using unsupervised learning or stochastic learning.

[0082] Preferably, the operating setting model is trained using reinforcement learning. In other words, a so-called reinforcement learning agent can be used to train the setting device (or the underlying operating setting model), especially continuously. The training data sets specified as training data can, for example, be or include the training data sets mentioned above. For reinforcement learning, a reward function is specified, which, for example, rewards maximizing the performance of the setting device.

[0083] Preferably, the setting device uses a reinforcement learning agent to determine a control variable for a control method and / or regulation method of a production step (e.g., as mentioned above), wherein the agent is suitable and intended for exploring an environment of the agent, i.e., for exploring the environment by actions deviating from the best actions known to the agent, wherein the agent determines an exploration rate and / or decides whether further exploration is to be carried out.

[0084] The agent generally prefers to explore its environment in order to better evaluate the state space. Exploration and / or an exploration amount or reward can be determined depending on the (further) user inputs with which an operator, as the user, overrides the proposed control and / or regulation.

[0085] Furthermore, a use of the above-mentioned method is proposed, namely for automatically creating or tracking inputs to the machine control by the operating device instead of the user, in particular temporally between manual inputs by the user, in order to reduce (a frequency of) inputs by the user.

[0086] In addition, a corresponding machine is proposed, in particular an injection molding machine, for a production cell, wherein the machine has a machine control for automatically controlling the machine and an operating device with a human-machine interface for inputting settings (and control commands) for the machine control by a user, the machine control has a plurality of control inputs for transferring the settings (and control commands) from the operating device to the machine control, and the human-machine interface has a display and input means or (an) operating element(s), and the operating device comprises a processor with a program and associated data which are designed to carry out the method listed above.

[0087] The display can be a screen, and the input device or control element can be an on-screen keyboard / keypad, mouse, trackpad, or touchscreen buttons. A tablet could also serve as at least an input device for the control device, as could a smartphone or other handheld device.

[0088] In one embodiment of the machine, the processor with the program and the associated data is designed to carry out the process described above.

[0089] Furthermore, a computer program is proposed for a processor of an operating device of a machine, in particular an injection molding machine, of a production cell, comprising instructions which, when the computer program is executed by the computer / processor, cause the computer / processor to carry out the method (or the steps of this method) listed above.

[0090] Furthermore, a computer-readable data carrier on which the above-mentioned computer program is stored and / or a data carrier signal which transmits the computer program to the machine or to the processor is proposed.

[0091] It should be noted that the computer program (firmware / software / code) for the operating device can run on a processor within the operating device or on a processor belonging to the machine control system. It is also conceivable for the computer program to consist of multiple parts executed by different processors. For example, one part of the computer program can be assigned to the human-machine interface (e.g., a graphical user interface), which is executed by a processor located in the operating device, while another part of the computer program, which implements the function blocks, is executed on a processor belonging to the machine control system.

[0092] BRIEF DESCRIPTION OF THE DRAWINGS

[0093] Non-limiting embodiments of the present invention are explained in more detail below with reference to the figures. They show:

[0094] Fig. 1 is an exemplary block diagram of a production cell;

[0095] Fig. 2 shows another exemplary block diagram of a production cell;

[0096] Fig. 3 is a schematic representation of a control circuit according to the invention;

[0097] Fig. 4 shows a first exemplary representation on a graphical user interface of an operating device according to the invention; Fig. 5 shows a second exemplary representation on a graphical user interface of an operating device according to the invention; and

[0098] Fig. 6 shows a third exemplary representation on a graphical user interface of an operating device according to the invention.

[0099] DETAILED DESCRIPTION OF THE INVENTION

[0100] Fig. 1 shows an example block diagram of a production cell 1. In the simplest case, the production cell 1 consists of a (production / manufacturing / main) machine 2, such as an injection molding machine, which produces injection-molded parts (also called injection moldings). The machine 2 has a machine controller 4 for the automatic control of the machine 2 and an operating device 3 with a human-machine interface (or a graphical user interface, also called a "graphical user interface" or GUI) for the user to enter settings and / or control commands for the machine controller 4. The machine controller 4 usually has a plurality of control inputs 8 for transferring the settings and / or control commands from the operating device 3 (via signal outputs 7) to the machine controller 4. In addition, various signal sources orSignal outputs 9 must be present, each of which can be connected to the operating device 3 (via signal inputs 6). In the simplest case, these signal outputs 9 are located on the machine 2 or on the machine control 4. This allows signals, such as sensor signals or actual states, to be fed back from the machine 2 or from the machine control 4 to the operating device 3. There are also signals which are generated by the machine control 4 for the control of individual components of the machine 2 and those which are generated within the machine 2 and fed to the machine control 4. Not all of these machine-internal signals are made available for the operating direction 3 at the signal outputs 9, i.e. only a portion of them.

[0101] In the proposed method, various function blocks 10 are presented to the user of the machine 2 or the operating device 3 for selection on a display, e.g., via a graphical user interface. These function blocks 10 each comprise at least one function block input, at least one function block output, and a transformation function with which at least one input signal or input value at at least one function block input is transformed into at least one output signal or output value at at least one function block output. The user can select a desired function block 10 and then configure it. Selection and configuration are carried out using suitable input means or one or more operating elements.To configure the selected function block 10, the (previously mentioned) various signal outputs 9 (from the machine 2 or the machine control 4) or various default settings are made available for selection by the operating device 3 on the display for connection to the at least one function block input. The desired signal output 9 or the desired default setting is selected by the user and connected to the function block input. Accordingly, the control inputs 8 of the machine control 4 are made available for selection by the operating device 3 on the display for the user, and the user selects the desired control input 8 that they wish to connect to the function block output. This enables the user to define control sequences or control loops themselves, without additional programming of the machine control 4 by the manufacturer of the machine 2, i.e.in particular without changing the program code of the machine control 4. This allows the user to optimise the operation of the machine 2 himself, in particular without assistance from the manufacturer, thus avoiding the transfer of internal company information from the user to the manufacturer.

[0102] The control device does not necessarily have to be located on the machine 2 itself. The control device 3' can also be located (far) away from the machine 2, for example, in a remote control room. This is indicated in Fig. 1 by the remote control device 3' shown in dashed lines.

[0103] In many cases, the production cell 1 comprises, in addition to the main machine 2, such as an injection molding machine, additional (auxiliary) machines 5 for the production of the parts. Such additional (auxiliary) machines 5 can be, for example, conveyor machines for introducing raw materials for production into the main machine and / or for transporting the finished parts away from the main machine. Furthermore, these can also be processing machines which, for example, prepare the raw materials and testing machines which check the finished parts for defects. Additional measuring devices or sensors, for example, are also used. These can, for example, record the ambient conditions or determine properties of the raw materials or the finished parts. Furthermore, they can be simple actuators, such as servo / stepper motors, for operating or adjusting openings. As an example, such an additional (auxiliary / additional) machine (or a sensor / actuator) 5 is shown in Fig. 1.This also has a signal output 9' and / or a control input 8'. These can also be connected to the selected function block 10 via an input signal (between 9' and 6) or output signal (between 7 and 8'), as shown in Fig. 1. Fig. 2 shows another example block diagram of a production cell 1 consisting of the (main) machine 2 and another (auxiliary / additional) machine 5 (e.g. with a sensor 13 and an actuator 11 as well as an intermediate control system 12). Fig. 2 illustrates various possible control loops ad. For example, there can be machine-internal control loops a (see dotted line), in which signal outputs from the machine 2 or the machine control 4 are fed directly back to signal inputs of the machine 2 or the machine control 4. Furthermore, there can be machine-internal control loops b (seedashed line), in which signal outputs 9 from the machine control system 4 are passed to the operating device 3 for processing (via signal inputs 6) and then (an) output signal(s) of the operating device 3 are fed back to signal inputs 8 of the machine control system 4. Furthermore, there can be machine-external control loops c (see dotted-dashed line), in which signal outputs 9' from a supplied auxiliary machine 5 (or sensor 13) are passed to the operating device 3 for processing (via signal inputs 6) and then (an) output signal(s) of the operating device 3 are fed back to signal inputs 8' of the supplied auxiliary machine 5 (or actuator 11). In addition, combined or coupled (internal + external) control loops b + c are also conceivable, in which signal outputs 9 from the machine control system 4 are also used, for example, for the indirect control of external actuators 11 via the operating device 3.External sensors 13 then determine the effect of the external actuators 11 on the control system 12 of the supplied (auxiliary / additional) machine 5, and the sensor signals are fed to the operating device 3, where they are processed and corresponding control signals or inputs are generated for the machine control system 4. Finally, external control loops d are also conceivable (see double-dotted-long-dashed line), in which signal outputs 9 from the machine control system 4 are used, for example, directly to control external actuators 11. External sensors 13 then determine the effect of the external actuators 11 on the control system 12 of the supplied auxiliary machine 5, and the sensor signals are in turn fed directly to the machine control system 4. In this case, the operating device 3 is used by the user to generate configurations for the machine control system 4 and transmit them to it (via the configuration output 7 of the operating device 3).In addition, in the production cell 1 according to Fig. 2, as well as in the one according to Fig. 1, selected signal outputs 9 from the (main) machine 2 or from the machine control 4 as well as from the further (auxiliary / additional) machine 5 can be fed via input signals to the operating device 3, where they are converted by means of a selected transformation function into one or more output signals, which are then connected to selected signal inputs 8 of the machine control 4 or the further (auxiliary / additional) machine 5. Fig. 3 shows a schematic representation of a control loop according to the invention which regulates the production cycle. One or more selected actual values ​​of the production cycle are fed to the inputs of a selected function block. The actual values ​​are scaled at the inputs and / or outputs of the function block. This scaling can also include the addition of a negative or positive offset.The function block, which has a transformation function with one or more transformation parameters, such as a mathematical function or a logic block, was previously parameterized by the user according to the process requirements, e.g. by assigning values ​​to the transformation parameters. The function block generates setpoints, which are then validated, for example, to ensure that the setpoint does not lead to a malfunction or damage to the machine or other machines. During validation, it is determined, for example, whether the setpoints exceed specified limits (or thresholds). To prevent a malfunction or damage, the setpoints are restricted, for example, so that they do not exceed the specified limits. In addition, the output setpoints can be logged, for example to be able to determine the cause after a malfunction or damage.

[0104] In one example of such a control loop in an injection molding machine, the dosage of the starting material, i.e., the thermoplastic resin in the form of granules, is adjusted according to a melt pressure measurement (e.g., at the inlet of an injection nozzle). In another example of such a control loop in an injection molding machine, the viscosity of the injected plastic is adjusted based on the plasticizing temperatures. The flow index is used as the actual value, and the temperatures of the plasticizing zones are specified as the target value.

[0105] In another example of such a control loop in an injection molding machine, the diameter of the beverage closures produced by injection molding should lie within a certain tolerance. For this purpose, the diameter of the beverage closures produced by the injection molding machine is determined by a downstream measuring system. Depending on the mean value of the measured diameter of the produced beverage closures, the pressure and / or the cooling time and / or the temperature of the injection molding machine are changed. In another example of such a control loop in an injection molding machine, the molded parts should have a certain color. For this purpose, the color of the molded parts produced by the injection molding machine is determined by a downstream measuring system. Depending on the mean value of the measured color of the produced molded parts, the amount of colorants added to the granulate is changed.

[0106] To create a new control, the user performs the following steps (for a "Single Input" / "Single Output", SISO controller):

[0107] 1 . Create a new rule system by pressing the "Plus" button

[0108] 2. Select the controller block

[0109] 3. Select controller type (transformation function)

[0110] 4. Adjust the basic parameters of the selected controller

[0111] 5. Select input port

[0112] 6. Assign input signal to the input port

[0113] 7. Set the scaling of the input signal at the input port

[0114] 8. Select output port

[0115] 9. Assign output signal to the output port

[0116] 10. Set the scaling of the output signal at the output port

[0117] 11 . Limit the value range of the output signal

[0118] 12. Activate control

[0119] The user can then test the effect of the new control by activating it and then deactivating it again and observing the different results.

[0120] Fig. 4 shows an example of a graphical user interface for an operating device. Using this user interface, a new controller unit is created and configured. In the view shown, the output is mapped to a desired range of the machine-relevant control variable by scaling (gain), adding an offset, and setting an upper and lower limit.

[0121] Fig. 5 shows another example of the graphical user interface of the control unit. In this view, a specific controller is selected and parameterized. A proportional / P controller is selected and its gain factor is set. Furthermore, the corresponding input variable (i.e., a specific actual signal) and a corresponding output variable (i.e., a specific setpoint signal or control signal) are connected to the input and output of the controller.

[0122] Fig. 6 shows yet another example representation of the graphical user interface of the operating device. In this view, the input and output signals of several controllers are displayed together. The individual controllers can be activated and deactivated in this view. In addition, warnings are displayed which indicate possible problems or errors so that the user can correct or improve settings. The invention enables the user to further control machines with a machine control and operating device, particularly in the field of injection molding machines, and in particular machines that are part of a production cell, and to operate them according to specific needs and special requirements without the user having to contact the manufacturer so that the manufacturer can make the adjustments desired by the user.

[0123] In addition, the invention makes it possible to reduce the number of user interventions in the production process through automatic adjustment.

[0124] LIST OF REFERENCE SYMBOLS

[0125] 1 Local production cell

[0126] 2 (main) machine, injection molding machine (production / manufacturing machine)

[0127] 3 Control device

[0128] 3' Remote control device

[0129] 4 Machine control

[0130] 5 additional (auxiliary / additional) machines / sensors / actuators

[0131] 6 Signal input of the control device

[0132] 6' Signal input of the remote control device

[0133] 7 Signal output of the control device

[0134] 7' Signal output of the remote control device

[0135] 7" configuration output of the control unit

[0136] 8 Control / signal input of the machine control

[0137] 8' Control / signal input of the additional (auxiliary / additional) machine / sensor / actuator

[0138] 9 Signal output of the machine control

[0139] 9' Signal output of the additional (auxiliary / additional) machine / sensor / actuator

[0140] 10 Function block in the operating device

[0141] 10' Function block in the remote control unit

[0142] 11 External actuator (of the additional machine)

[0143] 12 Control system of the additional (auxiliary / additional) machine

[0144] 13 External sensor (of the additional (auxiliary / additional) machine) a Direct internal control loop of the machine control (dotted line ■■■) b Internal control loop via operating device (dashed line — ) c External control loop including additional (additional) machine via operating device (dotted-dashed line — — ) d External control loop including additional (additional) machine with configuration via operating device (double-dotted-long-dashed line ■—■■——)

Claims

Patent claims 1. Method for operating and controlling a machine (2), in particular an injection molding machine, of a production cell (1), wherein the machine (2) has a machine control (4) for automatically controlling the machine (2) and an operating device (3; 3') with a human-machine interface for inputting settings and / or control commands for the machine control (4) by a user, the machine control (4) has a plurality of control inputs (8) for transferring the settings and / or control commands from the operating device (3; 3') to the machine control (4), and the method comprises the following steps: for selecting various function blocks (10; 10') by the operating device (3; 3') on a display of the human-machine interface, wherein the function blocks (10;10') each comprise at least one function block input, at least one function block output and a transformation function with which at least one input signal or input value at at least one function block input is transformed into at least one output signal or output value at at least one function block output; Selection of one of the available function blocks by the user; and Configuring the selected function block by the user, where the configuration includes the following steps: • to select different signal sources or different default settings by the operating device (3; 3') on the display for connection to the at least one function block input; • Selection of at least one of the signal sources offered for selection by the user; • to select the plurality of control inputs (8) of the machine control (4) by the operating device (3; 3') on the display for connection to the at least one function block output; and • Selection of at least one of the control inputs (8) provided for selection by the user.

2. The method according to claim 1, further comprising parameterizing the transformation function of the selected function block (10; 10') with one or more transformation parameters by the user, wherein the parameterizing comprises assigning a value to one or more of the one or more transformation parameters.

3. Method according to claim 1 or 2, wherein the signal sources are one or more signal outputs (9) of the machine (2) and / or one or more signal outputs (9') of another machine (5) of the production cell (1) and / or a signal output (9') of one or more sensors (13') of the production cell (1) not belonging to the machine (2).

4. Method according to one of claims 1 to 3, wherein in addition to making the plurality of control inputs (8) of the machine control (4) available for selection, at least one further control input (8') of one or the further machine (5) of the production cell (1) is made available for selection by the operating device (3; 3') on the display for connection to the at least one function block output in order to influence a machine control of the further machine (5) by selection of the at least one further control input (8') by the user.

5. The method according to claim 3 or 4, further comprising physically connecting the selected signal output(s) (9') of a further machine (5) and / or the selected sensor(s) (13') not belonging to the machine (2) and / or the selected further control input(s) (8') of the further machine (5) to the operating device (3; 3').

6. The method according to one of claims 1 to 5, wherein a function block (10; 10') is the following: a controller, in particular a P, a PI, a PD or a PID controller, further in particular a linear or non-linear controller, even further in particular a fuzzy logic controller; a feedforward control; a mathematical function; a logic block, in particular a stateless or a stateful logic block; a state machine; a lookup table; a neural network, especially a trained or self-learning neural network.

7. Method according to one of claims 1 to 6, further comprising executing a program code assigned to the selected function block (10; 10') and / or activating a processing unit assigned to the selected function block (10; 10'), in particular in the operating device (3; 3'), in particular without changing a program code of the machine control (4).

8. The method according to any one of claims 1 to 7, wherein an update period of the selected functional block (10; 10') or a time constant of at least one of the transformation parameters is in a range from one minute to several hours, in particular up to one hour, in particular up to 30 minutes, in particular up to 10 minutes.

9. Method according to one of claims 1 to 8, wherein an actual value or actual state is present at the function block input and / or a desired or target value or desired or target state is present at the function block output.

10. The method according to one of claims 1 to 9, wherein the transformation function comprises at least one scaling and / or an offset, in particular an input scaling of the at least one input signal or input value and / or an addition of an input offset and / or an output scaling of the at least one output signal or output value and / or an addition of an output offset.

11. The method according to any one of claims 1 to 10, wherein the at least one output signal or the at least one output value is validated to ensure that the at least one output signal or the at least one output value does not lead to any malfunction or damage to the machine (2) or the further machine (5).

12. Method according to one of claims 1 to 11, wherein the at least one output signal or the at least one output value is limited, in particular depending on the result of the validation according to claim 11.

13. Method according to one of claims 1 to 12, wherein the at least one output signal or the at least one output value is logged, in particular by periodically, in particular once per update period, storing a sample value of the at least one output signal or the at least one output value, in particular together with a time indication.

14. Use of the method according to one of claims 1 to 13 for automatically creating or tracking inputs to the machine control (4) by the operating device (3; 3') instead of the user, in particular temporally between manual inputs by the user.

15. Machine (2), in particular an injection molding machine, for a production cell (1), wherein the machine (2) has a machine control (4) for automatically controlling the machine (2) and an operating device (3; 3') with a human-machine interface for the input of settings and control commands for the machine control (4) by a user, the machine control (4) has a plurality of control inputs for transferring the settings and control commands from the operating device (3; 3') to the machine control (4), and the human-machine interface has a display and input means or (an) operating element(s), and the operating device (3; 3') comprises a processor with a program and associated data, which are designed to carry out the method according to claim 1.

16. Machine (2) according to claim 15, wherein the processor with the program and the associated data are further designed to carry out the method according to one of claims 2 to 13.

17. Program for a processor of an operating device (3; 3') of a machine (2), in particular an injection molding machine, of a production cell (1), comprising instructions which, when the program is executed by the processor, cause the processor to carry out the method according to one of claims 1 to 13.

18. A computer-readable data carrier on which the program according to claim 17 is stored and / or a data carrier signal which transmits the program according to claim 17 to the machine or processor.

Images