A double-sided or single-sided machine tool, and a method for controlling a double-sided or single-sided machine tool.

Abstract

To monitor a production process faster and safer while minimizing defective products.SOLUTION: A double-sided or single-sided machine tool, in which a first working disk and an opposing supporting member can be relatively rotated by rotary driving means, and a working gap for double-sided or single-sided processing of a flat wafer is formed between the first working disk and the opposing supporting member, is provided with a plurality of sensors for recording measurement data of tool parameters and / or processing parameters in operation, and a control device for acquiring measurement data recorded by the sensors. The control device is provided with an artificial neural network designed to generate a state vector of the double-sided or single-sided machine tool from the measurement data and to compare the state vector with at least one target state vector. A control method thereof is also provided.SELECTED DRAWING: Figure 1

Description

Technical Field

[0003] <00000之3> The present invention preferably comprises an annular first working disk and preferably an annular opposing support element, and relates to a double-sided or single-sided working machine in which the first working disk and the opposing support element are relatively rotationally drivable by rotational drive means. The double-sided or single-sided working machine is for machining both sides or one side of a flat workpiece, preferably a wafer, and preferably an annular working gap is formed between the first working disk and the opposing support element, and during operation of the double-sided or single-sided working machine, a plurality of sensors for recording measurement data relating to the tool parameters and / or machining parameters of the double-sided or single-sided working machine are provided. The present invention also relates to a method of controlling such a double-sided or single-sided working machine.

Background Art

[0002] For example, in a double-sided polishing machine, a flat workpiece such as a wafer is preferably polished between annular working disks. Preferably an annular working gap is formed between the working disks, and in this working gap a flat workpiece, for example a wafer, is held during the polishing operation. For this purpose, a so-called rotor disk is usually arranged in the working gap and has recesses in which the workpiece is mounted in a floating manner. For machining, the working disks are rotationally driven relative to each other by a rotational drive device, and the rotor disk is also usually rotated in the working gap by the outer teeth of the rotor disk which engage with the corresponding teeth of a pin ring. As a result, the workpiece is conveyed through the working gap along a cycloid path during machining. Furthermore, an abrasive known as slurry is introduced into the working gap during double-sided polishing to perform the polishing process. Furthermore, in a double-sided polishing machine, polishing cloths called polishing pads are regularly arranged on the surfaces of the working disks so as to delimit the working gap.

[0003] The goal of machining is to make the shape of the fully machined workpiece as flat as possible. For this to happen, the shape of the working gap is critically important. Patent Document 1 discloses a double-sided machine tool equipped with means for deforming one side of the working disc overall. In particular, the upper working disc can be deformed between an overall concave shape and an overall convex shape. In such an overall deformation, first, viewed radially, the concave or convex shape of the working disc occurs over the entire diameter of the working disc. The ring surface of a preferably annular working disc, which defines the boundary of the working gap, remains flat itself. However, the opposing ring portions of the ring surface are deformed relative to each other, resulting in an overall concave or convex shape.

[0004] A double-sided machine tool equipped with means for causing local deformation of one side of a work disc, particularly deformation between a local convex shape and a local concave shape, is known from Patent Document 2. In the case of such local deformation, a convex shape or a concave shape occurs radially, for example, between the inner and outer edges of an annular work disc. Unlike global deformation, in the case of local deformation, the ring portion itself deforms into a concave or convex shape.

[0005] The two embodiments described above can be combined in a double-sided machine tool. In this way, a wide range of working gap shapes can be generated. Therefore, it is possible to always ensure the machining of workpieces that are as parallel as possible, or, regardless of whether they are parallel or not, a working gap setting that is favorable for the quality of the workpiece, even when, for example, there is partial wear of the abrasive cloth or when the temperature of the parts that define the working gap changes.

[0006] The shape of the working gap has a decisive influence on the shape and uniformity of the machined workpiece. In addition to the working gap shape, the machining results are also influenced by numerous additional tool parameters and / or machining parameters, such as the temperature of various machine components, the thickness and wear potential of the machining lining, such as abrasive cloth, the working discs and / or opposing support elements rotating relative to each other, and the rotational speed of a rotor disc rotatably mounted in the working gap, or the load between, for example, the first working disc and the opposing support elements.

[0007] It is known that during the operation of double-sided or single-sided machine tools, these types of tool parameters and / or machining parameters are monitored by sensors. It is also known that corresponding sensors are used to detect the shape and thickness of a flat workpiece (e.g., a wafer) being machined within the working gap. The parameter window suitable for machining with a double-sided or single-sided machine tool must be found from a number of tool and machining parameters as part of the machine tool's setup. The double-sided or single-sided machine tool must be adjusted to suit the common conditions of its relevant use, such as the type of working lining (e.g., abrasive cloth), whether abrasives are applied, and other parameter specifications of the operator. In subsequent production processes using the double-sided or single-sided machine tool, the process needs to be monitored by sensors. This allows for early identification and, if necessary, correction of deviations from specified target values, such as GBIR or SFQR values ​​of the machined wafer, during the machining process.

[0008] In particular, because there are many different machining processes, expert staff need to interpret the sensor measurement results in order to draw the correct conclusions for application to the production process. This type of expert staff is not available wherever double-sided or single-sided machine tools are used. This can negatively impact the production process. Furthermore, corrective operations are often applied only after a considerable time has passed since the harmful parameter deviation occurred. One reason for this is that the large number of tool and machining parameters that affect the production process makes it difficult for experts to identify deviations in measured parameters early. Often, this situation occurs after the machined workpiece has been measured. If an undesirable deviation is discovered during the production process, a considerable number of defective products will be produced in the meantime. [Prior art documents] [Patent Documents]

[0009] [Patent Document 1] German Patent No. 10 2006 037 490 [Patent Document 2] German Published Patent No. 10 2016 102 223 [Overview of the Initiative] [Problems that the invention aims to solve]

[0010] The object of the present invention is to provide a double-sided or single-sided machine tool, and a control method for a double-sided or single-sided machine tool, thereby enabling faster and more reliable monitoring of production processes using double-sided or single-sided machine tools while minimizing defective products. [Means for solving the problem]

[0011] The present invention solves this objective by independent claims 1 and 9. Advantageous embodiments can be found in the dependent claims, description and drawings.

[0012] With regard to the double-sided or single-sided machine tool mentioned at the beginning, the present invention provides a control device that acquires measurement data recorded by a sensor during the operation of the double-sided or single-sided machine tool, and the control device comprises an artificial neural network designed to create a state vector of the double-sided or single-sided machine tool from the measurement data and compare the state vector with at least one target state vector.

[0013] Regarding the control method described at the beginning, the present invention achieves its objective by training an artificial neural network by inputting a number of target state vectors that yield acceptable machining results for a flat workpiece.

[0014] The double-sided or single-sided machine tool according to the present invention can be a double-sided or single-sided polishing machine in particular. However, the double-sided or single-sided machine tool can also be a double-sided or single-sided lapping machine or a double-sided or single-sided grinding machine. The double-sided or single-sided machine tool preferably has an annular first working disk and preferably an annular opposing support element. In a single-sided machine tool, the opposing support element can be designed, for example, as a simple weight or pressure cylinder. The opposing support element can preferably be an annular second working disk. The first working disk and the opposing support element can be rotated relative to each other, and a preferably annular working gap is formed between the first working disk and the opposing support element for processing a flat workpiece such as a wafer. In particular, in the case of a double-sided or single-sided polishing machine, at least the first working disk, preferably the opposing support element, or each of the second working disks can have polishing linings (polishing pads) on their surface(s) that define the working gap. Furthermore, during machining, a polishing medium, such as an abrasive, especially a polishing fluid (slurry), can be introduced into the working gap in a manner known to itself. The work disc may also be provided with a tempering channel, through which a tempering fluid, such as cooling water, is tempered during the operation of the work disc.

[0015] Double-sided or single-sided machine tools are used, in particular, for planar parallel machining of flat workpieces. For machining, the workpiece can be housed in a floating manner in a recess of a rotor disc positioned in the working gap, in a manner known to itself. The first working disc and the opposing support element are rotationally driven relative to each other during operation, for example, by a corresponding drive shaft and at least one drive motor. It is also possible that only one of the first working disc or the opposing support element is rotationally driven. However, it is also possible that both the first working disc and the opposing support element are rotationally driven, in which case they are usually driven in opposite directions. For example, in the case of a double-sided machine tool, the rotor disc can also be rotated through the working gap by a suitable kinematic system during the relative rotation between the first working disc and the opposing support element, so that the workpiece positioned in the recess of the rotor disc traces a cycloidal path within the working gap. For example, the rotor disc may have teeth on its outer edge that mesh with the corresponding teeth of a pin ring. Such a machine forms a so-called planetary motion system.

[0016] The first working disk and / or opposing support element can each be held by a support disk. Similar to the first working disk and opposing support element, the support disk may also be annular or have at least an annular support portion.

[0017] By known methods, sensors, particularly appropriate measuring devices, record measurement data regarding the tool parameters and / or machining parameters of a double-sided or single-sided machine tool during its operation. These are, in particular, the tool parameters and / or machining parameters mentioned at the beginning. Specifically, the sensors record measurement data at predetermined intervals or continuously. The measurement data characterizes the operating parameters and tool parameters of the double-sided or single-sided machine tool, and the associated manufacturing process. The measurement data recorded by the sensors is also supplied, in particular, at predetermined intervals or continuously to the control device of the double-sided or single-sided machine tool. Tool and / or machining parameters can be recorded in real time. This also applies to the transfer of measurement data to the control device and the processing of measurement data, as described later. The recorded measurement data can be stored in data memory and transferred from there to the control device, for example, in real time or with a delay.

[0018] For processing measurement data, the control device according to the present invention comprises an artificial neural network that creates a state vector for a double-sided or single-sided machine tool from the received measurement data. The state vector consists of or is formed from the current measurement data of the sensor. The state vector characterizes the double-sided or single-sided machine tool, in particular the current production process. The artificial neural network compares this state vector to at least one, preferably more, in particular a set of target state vectors. Within the scope of training, target state vectors are specified to the artificial neural network as state vectors representing appropriate machining results during machining of a workpiece on the double-sided or single-sided machine tool. Target state vectors can be defined for different purposes, for example, with a specific quality parameter (e.g., GBIR and / or SFQR) and / or production throughput or other parameters in mind. The same can be determined by comparing the state vector created from the current measurement data with the available target state vectors, thereby determining whether the current state vector matches one of the target state vectors that have been trained as appropriate. If it is determined that the recorded state vector does not match any of the appropriate target state vectors, corrective actions can be taken, for example, if a deviation from the appropriate target state vector has occurred. For example, intervention in the production process can be performed by adjusting the tool parameters and / or machining parameters. Of course, a defined tolerance range in which identified minor deviations from the target state vector are classified as acceptable can be specified for comparison. By adjusting the tool parameters and / or machining parameters based on the comparison, the production process can be influenced so that the recreated state vector matches at least one target state vector within a sufficient range.

[0019] Unlike operators, artificial neural networks can generate state vectors very quickly from large amounts of measurement data, and therefore tool parameters and / or machining parameters, and can compare these state vectors very quickly with at least one, preferably many, target state vectors. As a result, unacceptable deviations of the production process from an acceptable process can be quickly and reliably identified, especially when there are not sufficiently trained or experienced personnel at the production site of a double-sided or single-sided machine tool. The present invention takes advantage of the fact that in an optimal production process, measurable tool parameters and / or machining parameters have a certain relationship with each other. An artificial neural network trained with the tool parameters and / or machining parameters of the optimal process can thus quickly and reliably identify deviations of the current process from the optimal process. The artificial neural network constitutes an anomaly detector that identifies unacceptable deviations (anomalies) in the production process. Therefore, process optimization can be achieved more quickly and with fewer test production runs, even when the production process starts after an initial setup process. In the best-case scenario, only one production test is required, and external measurements downstream of the machined workpiece are unnecessary. The production process can be monitored in a simpler and faster manner, minimizing the number of defective products even at the start of production. In particular, the production of workpieces that fall outside the desired tolerances can be reduced, and in the best-case scenario, completely prevented, especially in double-sided or single-sided machine tools.

[0020] According to one embodiment, the control device can be designed to issue a warning message if the generated state vector deviates from at least one target state vector. The warning message can be provided to the operator of the double-sided or single-sided machine tool, for example, through the user interface of the double-sided or single-sided machine tool. In the simplest case, the operator receives a warning message that the tool parameters and / or machining parameters have deviated to an unacceptable extent from values ​​acceptable for an optimal production process. Based on this, the operator can manually intervene in the process and, in particular, adjust the tool and / or machining parameters in the target manner so that the state vector re-formed from the current measurement data corresponds to at least one target state vector.

[0021] In another variation, the warning message may already include adjustment suggestions for adjusting specific tools and / or machining parameters. These adjustment suggestions can be issued by the control unit based on adjustment rules stored in the control unit. These types of adjustment rules may have been created in advance by the operator of a double-sided or single-sided machine tool. The operator can then evaluate the adjustment suggestions and implement them if necessary. In this way, the control unit can automatically generate suggestions for changing the tools and / or machining parameters by combining the determined deviation between the state vector value and at least one target state vector with the formalized causal relationships of the tools and / or machining parameters of the double-sided or single-sided machine tool.

[0022] According to another embodiment, the control device may further include an adjuster designed to control the tool and / or operating parameters of a double-sided or single-sided machine tool, in particular a double-sided or single-sided machine tool, so that if the created state vector deviates from at least one target state vector determined by the comparison means, the created state vector conforms to at least one target state vector. The adjuster can, in particular, control actuators to affect tool parameters and / or machining parameters. Further automation is achieved by the adjuster in that it autonomously controls the double-sided or single-sided machine tool based on the comparison performed so that the state vector created from the current measurement data conforms to at least one target state vector. The adjuster may be incorporated into the control device.

[0023] The adjustment device may be designed to control the tool parameters and / or operating parameters of a double-sided or single-sided machine tool based on adjustment rules stored in the device. In particular, the adjustment rules can specify to the device specific control rules regarding a particular determined deviation of a state vector. In this case, the adjustment rules may be pre-created by an operator, for example. Based on this, automatic control based on control specifications pre-stored in the form of adjustment rules becomes possible, especially without operator intervention.

[0024] According to another embodiment, an additional artificial neural network can be provided that is designed to evaluate measurement data regarding tool parameters and / or machining parameters by machine learning, control the tool parameters and / or operating parameters of a double-sided or single-sided machining machine, particularly a double-sided or single-sided machining machine, based on this evaluation, and / or create and / or modify adjustment rules stored in an adjustment device. This additional artificial neural network may be an additional artificial neural network in addition to the above-described first artificial neural network that forms an anomaly detector. However, it is also conceivable that the additional artificial neural network is designed to be integrated with the above-described artificial neural network that forms an anomaly detector. Further, the adjustment device can also be integrated into the additional artificial neural network.

[0025] The additional artificial neural networks provided can, in particular, constitute a so-called Learning Classifier System (LCS), i.e., an artificial intelligence system. This type of system is based on established if-then relationships and can modify the tool and / or machining parameters of a double-sided or single-sided machine tool as a function of outliers, i.e., the deviation between the current state vector detected by the (first) artificial neural network and at least one target state vector. The LCS creates output data from input data and rules. The control device may also have a memory that stores previously acquired tool and / or machining parameters, including data about the workpiece being machined. The stored data can be made available to the artificial neural network, in particular the LCS, which takes this data into consideration when evaluating the measurement data and the resulting control data of the double-sided or single-sided machine tool. Preferably, an artificial neural network designed as an LCS can recognize the probability that the machining results of the workpiece, such as characteristic values ​​like GBIR or SFQR, will deviate from a specified target value as early as possible in the production process. Based on this, the artificial neural network can intervene as early as possible in the production process, or at the latest in a subsequent production process, to prevent defective products, for example, by controlling actuators for specific tools and / or machining parameters. The artificial neural network can also be used, through machine learning, to improve adjustment rules initially created by operators based on further experience from the production process. For this purpose, the artificial neural network can modify adjustment rules stored in the adjustment device. It is also conceivable that the adjustment rules be created by the artificial neural network and then optimized as needed based on additional process data. The above embodiments enable maximum automation of the production process without requiring operator intervention.

[0026] According to another embodiment, the sensor is for measuring the working gap, particularly the shape and / or width of the working gap, more particularly the distance between the first working disk and the opposing support element, and / or for measuring the temperature of the first working disk and / or the opposing support element and / or other mechanical parts of the double-sided or single-sided working machine, and / or for measuring the temperature and / or flow rate of the processing agent supplied to the working gap for machining the workpiece, and / or for measuring the rotational speed of the rotor disk rotatably mounted in the working gap of the first working disk and / or the opposing support element and / or the working gap, and / or for measuring the load between the first working disk and the opposing support element, and / or for measuring the rotational speed and / or torque and / or temperature of the rotational drive device, and / or for measuring the pressure and / or force of the means for generating deformation of the first working disk and / or the opposing support element, and / or for measuring the pressure and / or force of the means for generating deformation of the first working disk and / or the opposing support element (16), and / or for measuring the thickness of the machining lining of the first working disk and / or the opposing support element, and / or for measuring the thickness and / or shape of the workpiece machined on the double-sided or single-sided working machine. The above-mentioned measuring devices may exist together or may exist in any desired combination. The processing agent may be, for example, an abrasive, particularly a polishing liquid such as a slurry. The above-mentioned measuring devices record the tools and processing parameters of the double-sided or single-sided working machine related to the production process, including, for example, environmental data.

[0027] If a double-sided or single-sided machine tool, in particular the tool and / or machining parameters of a double-sided or single-sided machine tool, is controlled based on the deviation between a generated state vector and at least one target state vector determined by a comparison means, this may, in particular, consist of controlling actuators to affect the working gap, in particular the shape and / or width of the working gap. More specifically, the actuator is controlled to affect the distance between the first working disc and the opposing support element, and / or the temperature of the first working disc and / or the opposing support element and / or other machine parts of the double-sided or single-sided machine tool, and / or the temperature and / or flow rate of the work fluid supplied to the working gap for machining the workpiece, to affect the rotational speed of the first working disc and / or the opposing support element and / or a rotor disc rotatably mounted in the working gap, and / or the load between the first working disc and the opposing support element, and / or the rotational speed and / or torque and / or temperature of the rotary drive, and / or to measure the pressure and / or force that causes deformation of the first working disc and / or the opposing support element, and / or to measure the pressure and / or force that causes deformation of the first working disc and / or the opposing support element, and / or to affect the thickness of the machining lining of the first working disc and / or the opposing support element, and / or to affect the thickness and / or shape of the workpiece machined on the double-sided or single-sided machine tool. The effects of the actuators described above, or their individual controls, can be performed together or in any combination. The processing agent may be, for example, an abrasive, especially an abrasive fluid such as a slurry. Thus, the controlled actuators affect the tool and machining parameters of the double-sided or single-sided machine tool, which are related to the production process, such as environmental data.

[0028] According to another embodiment, the opposing support element is preferably formed by an annular second working disk, the first and second working disks are arranged coaxially with respect to each other and are rotatably driven relative to each other, and a working gap is formed between the working disks for double-sided or single-sided machining of a flat workpiece.

[0029] The present invention also relates to a system comprising at least two double-sided or single-sided machine tools according to the present invention, wherein a higher-level artificial neural network is provided, which is connected to the artificial neural networks of the at least two double-sided or single-sided machine tools, and the higher-level artificial neural network is designed to train at least one of the artificial neural networks of the at least two double-sided or single-sided machine tools based on data obtained by the artificial neural networks of the at least two double-sided or single-sided machine tools by inputting state vectors that result in acceptable machining outcomes for a flat workpiece.

[0030] In this embodiment, a system comprising at least two, in particular two or more, double-sided or single-sided machine tools according to the present invention is provided. Furthermore, a higher-level artificial neural network is provided, connected to the artificial neural networks of at least two double-sided or single-sided machine tools. The higher-level artificial neural network is designed to train the artificial neural networks of at least two double-sided or single-sided machine tools based on data obtained by the artificial neural networks of at least two double-sided or single-sided machine tools. In this way, the higher-level neural network forms a higher-level structure that can integrate similar double-sided or single-sided machine tools. Furthermore, a cross-plant memory may be provided that acquires data from similar double-sided or single-sided machine tools in the system and provides such data to the higher-level artificial neural network. In this way, individual double-sided or single-sided machine tools in the system can be optimized by taking into account the data stored in memory and, where applicable, by using the individual data of the double-sided or single-sided machine tools in the system together. The above-described embodiment provides advantageous effects, for example, with respect to production planning, fleet management, or predictive maintenance.

[0031] A double-sided or single-sided machine tool according to the present invention can be designed to carry out the method according to the present invention. Therefore, the method according to the present invention can be carried out using a double-sided or single-sided machine tool according to the present invention.

[0032] As already described, in the method according to the present invention, the artificial neural network is trained by inputting a number of state vectors that lead to a suitable machining result for a flat workpiece. Training can be performed by having an operator carry out a production process using a double-sided or single-sided machine tool with different tool parameters and / or machining parameters, and depending on the machining result, whether the production process yielded a suitable machining result is specified to the artificial neural network with respect to each tool parameter and / or machining parameter. In this case, the relevant tool parameters and / or machining parameters are stored in the artificial neural network as target state vectors. This startup learning is generally performed before starting normal machining of a flat workpiece using a double-sided or single-sided machine tool.

[0033] Furthermore, the artificial neural network trained in this manner can be further trained by inputting additional target state vectors that lead to appropriate machining results for flat workpieces during the operation of a double-sided or single-sided machine tool. This additional learning during the production process with a double-sided or single-sided machine tool further optimizes the tool parameters and / or machining parameters.

[0034] According to another embodiment, an additional artificial neural network can be trained using a trained artificial neural network by inputting a number of target state vectors that result in a suitable machining outcome for a flat workpiece during the operation of a double-sided or single-sided machine tool. The additional artificial neural network may be untrained or already (pre-)trained. For example, the additional artificial neural network may be a copy of the trained artificial neural network, which is further trained based on it. This is useful, for example, when the trained artificial neural network is a general-purpose neural network trained for a specific type of double-sided or single-sided machine tool, but has not yet been specialized for a specific double-sided or single-sided machine tool, particularly with respect to the individual machining parameters of the site. As a result, a specialized version of the trained artificial neural network is generated, which can ultimately replace the trained artificial neural network. In this case, training is performed based on test data or laboratory data from the manufacturer of the double-sided or single-sided machine tool, and then further specialized for the client's individual manufacturing process using the additional, modern neural network. Therefore, there is little need to understand the production process at the installation site of the double-sided or single-sided machine tool. [Brief explanation of the drawing]

[0035] Hereinafter, exemplary embodiments of the present invention will be described in more detail with reference to schematic drawings. [Figure 1] This is a cross-sectional view showing a part of a double-sided or single-sided machine tool according to the present invention in a first operating state. [Figure 2] Figure 1 shows the second operating state. [Figure 3] Figure 1 shows the third operating state. [Figure 4] This is a schematic diagram of the function of a double-sided machine tool according to the present invention, according to a first exemplary embodiment. [Figure 5] This is a schematic diagram illustrating the function of a double-sided machine tool according to the present invention, according to another exemplary embodiment. [Figure 6] This is a schematic diagram illustrating the function of a double-sided machine tool according to the present invention, according to another exemplary embodiment. [Figure 7] This is a schematic diagram illustrating the function of a double-sided machine tool according to the present invention, according to another exemplary embodiment. [Figure 8] This is a schematic diagram illustrating the function of a double-sided machine tool according to the present invention, according to another exemplary embodiment. [Figure 9] This is a schematic diagram illustrating the function of a double-sided machine tool according to the present invention, according to another exemplary embodiment. [Figure 10] The system according to the present invention is shown in general terms. [Modes for carrying out the invention]

[0036] Unless otherwise specified, the same reference number refers to the same object in the figure.

[0037] The double-sided machine tool illustrated in Figures 1 to 3 comprises an annular upper support disc 10 and an annular lower support disc 12. A first annular upper working disc 14 is fixed to the upper support disc 10, and a second, similarly annular lower working disc 16 is fixed to the lower support disc 12. An annular working gap 18 is formed between the annular working discs 14 and 16, and a flat workpiece such as a wafer is processed from both sides in this working gap 18 during operation. The double-sided machine tool can be, for example, a polishing machine, a lapping machine, or a grinding machine.

[0038] The upper support disc 10 and together with it the upper working disc 14 and / or the lower support disc 12 and together with it the lower working disc 16 can be rotationally driven relative to each other by a suitable drive system consisting, for example, an upper drive shaft and / or a lower drive shaft and at least one drive motor. The drive system is known in itself and for reasons of clarity, no further description is given. In embodiments that are also known in themselves, the workpiece to be machined can be held in the working gap 18 in a floating manner on the rotor disc. Using appropriate kinematics, such as planetary kinematics, it is ensured that the rotor disc also rotates through the working gap 18 during the relative rotation of the support discs 10, 12 or the working discs 14, 16, respectively. Temperature control channels can be designed in the upper working disc 14 or the upper support disc 10, and optionally in the lower working disc 16 or the lower support disc 12, through which a temperature control fluid, such as cooling water, can be transported during operation. This is also known in itself and is not shown in more detail.

[0039] The double-sided machine tools shown in Figures 1 to 3 are further equipped with distance measuring devices, also known by themselves as sensors. The sensors may be optically operated, for example, or electromagnetically operated (e.g., eddy current sensors). In the illustrated example, for example, three distance measuring devices 20, 22, and 24 are provided, which measure the distance between the upper work disk 14 and the lower work disk 16 at three radially spaced positions on the work disk 18, as indicated by the arrows in Figure 1. As can be seen, distance measuring device 20 measures the distance between the upper work disk 14 and the lower work disk 16 in the radially outer region of the work gap 18. Distance measuring device 24 measures the distance between the upper work disk 14 and the lower work disk 16 in the radially inner region of the work gap 18. Distance measuring device 22 measures the distance between the upper work disk 14 and the lower work disk 16 at the center of the work gap 18.

[0040] The distance measuring devices 20, 22, 24 are not shown in FIGS. 2 and 3 for the sake of clarity. The measurement data of the distance measuring devices 20, 22, 24 are applied to the control device 34.

[0041] In the present embodiment, the lower working disk 16 is fixed to the lower support disk 12 only in the regions of its outer edge and inner edge, as shown by reference numerals 26 and 28 in FIG. 1, and is, for example, screwed along a partial circle in each case. In contrast, the lower working disk 16 is not fastened to the lower support disk 12 between these fixing positions 26 and 28. Instead, an annular pressure volume 30 is arranged between the lower support disk 12 and the lower working disk 16 between these fixing positions 26, 28. This pressure volume 30 is connected via a dynamic pressure line 32 to a pressure fluid reservoir, not shown in detail in the figure, such as a liquid reservoir, particularly a water reservoir. A pump and a control valve can be arranged in the dynamic pressure line 32, and these can be controlled by the control device 34. In this way, the desired pressure acting on the lower working disk 16 can be generated in the pressure volume 30 by the fluid introduced into the pressure volume 30. The pressure in the pressure volume 30 can be measured by a pressure measuring device not shown in more detail. The measurement data of the pressure measuring device can also be applied to the control device 34, and the control device 34 can set a predetermined pressure in the pressure volume 30.

[0042] Due to the free movement between the fixing positions 26, 28, the lower working disk 16 can be made locally convex, as shown by the dotted line with reference numeral 36 in FIG. 2, by setting a sufficiently high pressure in the pressure volume 30. When a pressure p0 is assumed in the pressure volume 30 in the operating state of FIG. 1 where the lower working disk 16 has a planar shape, the convex deformation of the lower working disk 16 shown by 36 in FIG. 2 can be achieved by setting a pressure p1 > p0. On the other hand, the local concave deformation of the lower working disk 16 can be achieved by setting a pressure p2 < p0 in the pressure volume 30, as shown by the dotted line with reference numeral 38 in FIG. 3.

[0043] In this case, the lower working disk 16 can be seen to have a locally convex shape (Figure 2) or a locally concave shape (Figure 3) between its inner edge in the area of ​​fixed position 26 and its outer edge in the area of ​​fixed position 28 when viewed from the radial direction.

[0044] In addition to this local radial deformation of the lower working disc 16, means for the overall deformation of the upper working disc 14 can be provided. These means can be designed as described above or as described in DE 10 2006 037 490 B4, respectively. The upper support disc 10 and the upper working disc 14 fixed thereto are deformed overall such that the overall concave or convex shape of the working surface of the upper working disc 14 occurs throughout the entire cross-section of the upper working disc 14. In contrast, the upper working disc 14 may remain flat between its radial inner and radial outer edges, or it may be locally deformed in the manner described above by the pressure volume 30. Means for adjusting the shape of the upper working disc 14 can also be controlled by a control device 34.

[0045] The distance measuring devices 20, 22, and 24 form sensors that record measurement data related to the tool parameters and / or machining parameters of the double-sided machine tool, in this example, the thickness and shape of the working gap 18, particularly during the operation of the double-sided machine tool. Preferably, the double-sided machine tool is equipped with a plurality of additional sensors, each having a corresponding additional measuring device. The measuring device may be, in particular, of the type described above. The measuring device records additional tool parameters and / or machining parameters during the operation of the double-sided machine tool.

[0046] Measurement data recorded by the sensor is supplied to the control device 34. From the measurement data, the control device 34 creates a state vector for the double-sided machine tool using an artificial neural network 34 integrated into the device, and compares the state vector with at least one target state vector, preferably a set of target state vectors assigned to production processes acceptable within the training range.

[0047] The training of the artificial neural network 34 will be described in more detail with reference to Figure 4. In Figure 4, a double-sided machine tool according to the present invention is shown by reference numeral 40. An unprocessed workpiece 42, for example, an unprocessed wafer, is supplied to the machine tool for processing, and a completed processed workpiece 44, in particular a processed wafer 44, is output by the double-sided machine tool 40. A data memory 46 is provided, and measurement data relating to tool parameters and processing parameters recorded by sensors, for example, is supplied, and the data is made available to the operator 48, as shown by reference numeral 50 in Figure 4. Furthermore, measurement data relating to the shape of the processed workpiece, for example, is supplied to the data memory 46 as additional tool and / or operating parameters, and the data is also supplied to the operator 48, as shown by reference numeral 52 in Figure 4. Finally, external environment data is also supplied to the data memory. The external environment data may also be supplied to the operator 48. Based on this, the operator 48 performs an evaluation of the production process underlying each piece of data to determine whether the processing result is acceptable. Operator 48 makes this evaluation available to the artificial neural network 34 of the control device 34, as shown in Figure 4, at 56. The corresponding state vector is stored by the artificial neural network 34 as the target state vector.

[0048] Figure 5 illustrates how a double-sided machine tool is controlled. In this case, process data relating to the tool and / or machining parameters can be directly supplied to the artificial neural network 34 of the control device 34, as shown in 58 in Figure 5. The artificial neural network 34 creates a state vector from the obtained measurement data relating to the tool and / or machining parameters and compares the state vector with a stored target state vector. If an unacceptable deviation or mismatch is detected, the control device 34 issues a corresponding warning message to the operator 48, as shown in 60 in Figure 5. Based on this, and, if applicable, taking into account the measurement data of the workpiece 44 made available via 52, the operator 48 can control the double-sided machine tool 40, in particular the actuators that affect the tool and / or operating parameters, as shown in 62 in Figure 5, to continuously monitor and match the created state vector to at least one target state vector. In this case, the operator 48 determines the result from an evaluation of the received data. The operator 48 is assisted in this by the control device 34 as an anomaly detector.

[0049] Figure 6 shows another automated modification of the procedure shown in Figure 5. In this embodiment, the control device 34, in particular its artificial neural network 34, further comprises a control device 64 connected thereto, as shown in 66 in Figure 6. If a deviation occurs between the created state vector and at least one target state vector found by comparison, the control device 64 intervenes in the actuator of the double-sided machine tool 40, particularly without intervention by the operator 48. As a result, the recorded tool parameters and / or machining parameters of the double-sided machine tool 40 are adjusted, as shown in 68 in Figure 6. All relevant data can be stored in the data memory 46. For example, the integrated control and adjustment devices 34, 36 can control the tool and / or machining parameters of the double-sided machine tool 40 based on adjustment rules stored in, for example, the adjustment device 64. These adjustment rules may include, for example, specific control commands for specific established deviations of tool and / or machining parameters created by the operator 48, and the control and adjustment devices 34, 64 control the actuators according to these rules.

[0050] Figure 7 shows another embodiment of the procedure described with reference to Figure 6. In this embodiment as well, an operator 48 is involved. The operator also acquires process data regarding recorded tool parameters and machining parameters, as shown in 70 in Figure 7, and process data regarding the machined workpiece, as shown in 52. Finally, the operator 48 also acquires control commands executed by the control device 34, as shown in 72 in Figure 7. Based on this, the operator 48 can monitor the adjustments that have been made and, if necessary, adjust the adjustments of the adjustment device 64 in an appropriate manner, as shown in 74 in Figure 7.

[0051] Figure 8 shows another embodiment of possible training of an artificial neural network as an anomaly detector. Here, training proceeds from the control unit 34 using an already pre-trained artificial neural network 34, as described above with reference to Figure 4, for example. The control unit 34 issues any deviation or anomaly data to the operator 48, as shown in 60 in Figure 8 and as described above with reference to Figure 5. Based on this, the operator 48 can train an additional artificial neural network 76 in such a way that, during the operation of the double-sided machine tool 40, the operator 48 provides the additional artificial neural network 76 with target state vectors for appropriate tool parameters and / or machining parameters of the double-sided machine tool 40, as shown in 78 in Figure 8. The additional artificial neural network 76 may be an untrained artificial neural network 76, or it may be an already pre-trained artificial neural network 76, such as a copy of the neural network 34 of the control unit 34. Based on this, the additional artificial neural network 76 can be professionally trained at the start of production operation for the parameters of each individual process in the application scenario of the double-sided machine tool 40, based on the artificial neural network 34 of the control device 34. After the training, the additional artificial neural network 76 can replace the previously trained artificial neural network 34 of the control device 34.

[0052] Another embodiment of the present invention will be described with reference to Figures 9 and 10, which in particular comprises an additional artificial neural network 86 designed for machine learning. This may be a Learning Classifier System (LCS), i.e., an artificial intelligence system. In the embodiment shown in Figure 9, sensor measurement data relating to tool parameters and / or machining parameters is supplied to the data memory 46 on the one hand and to the control device 34 on the other hand, as shown in Figure 9 80. Measurement data relating to the workpiece, in particular to the shape of the machined workpiece, is also supplied to both the data memory 46 and the control device 34, as shown in Figure 9 82. The control device 34 also exchanges with the data memory 46, as shown in Figure 9 84. Figure 9 shows an additional artificial neural network 86 associated with the control device 34. The additional artificial neural network 86 can also be combined with the artificial neural network 34 of the control device 34. The additional artificial neural network 86 is designed for machine learning and, in particular, forms a Learning Classifier System (LCS).

[0053] Measurement data regarding the shape of the machined workpiece 44 is also supplied to the LCS 86 via 82. If, during operation of the double-sided machine tool 40, an unacceptable deviation is detected by the control device 34, particularly its artificial neural network 34, between the currently recorded state vector and the tolerance values ​​of the tool parameters and / or machining parameters stored as the target state vector, a corresponding abnormality signal is output to the LCS 86, as shown by 88 in Figure 9. The LCS 86 can also utilize past measurement data from the data memory 46. Based on this, the LCS 86 can autonomously decide, as shown by 90 in Figure 9, to change specific process parameters, in particular to control actuators to affect the recorded tool parameters and / or machining parameters, and control the actuators accordingly. In this way, maximum automation and autonomy are achieved.

[0054] Figure 10 shows another embodiment of the modification shown in Figure 9. In particular, Figure 10 shows a system according to the present invention comprising at least two double-sided machine tools 40. Of course, the system can also consist of two or more double-sided machine tools 40, as shown in three points in Figure 10. In Figure 10, two plants 92i, each corresponding in terms of design and function to the embodiment according to Figure 9, are shown as dashed blocks for illustrative purposes. The plants 92i can also be designed differently to achieve different objectives, such as optimal wafer quality, maximum output, etc. The plants are connected to a common data memory 46 via 80, 84. Furthermore, the system shown in Figure 10 includes a higher-order artificial neural network 94, which may also consist of, for example, an LCS and may be linked to an operator 48. The higher-order artificial neural network 94 is also connected to the data memory 46, as shown by 96. Furthermore, the higher-order artificial neural network 96 acquires control commands executed by the LCS 86 as needed, as shown by 98 in Figure 10. Based on this, the higher-level LCS94 can further optimize or specialize the LCS86 of plant 92i by specifying, for example, collective or individual control rules and / or target state vectors for each plant 92i. [Explanation of symbols]

[0055] 10 Upper support disc 12 Lower support disc 14 Upper working disc 16 Lower working disc 16 Opposing support elements 18 Working gap 20, 22, 24 Distance measuring device (sensor) 26 Fixed position 28 Fixed position 30 Pressure volume 32 Dynamic pressure lines 34. Control devices, artificial neural networks 40 Double-sided machine tools 42 Unprocessed workpieces 44 processed workpieces 46 Data Memory 48 Operators 64 Adjustment device 76,86,94 Artificial Neural Networks 92i Plant

Claims

1. A double-sided or single-sided machine tool comprising a first working disk (14) and opposing support elements (16), The first working disk (14) and the opposing support element (16) are rotated relative to each other by a rotational drive means. For double-sided or single-sided machining of flat workpieces (42, 44), a working gap (18) is formed between the first working disc (14) and the opposing support element (16). A double-sided or single-sided machine tool is equipped with multiple sensors (20, 22, 24) that record measurement data relating to the tool parameters and / or machining parameters of the machine tool in operation. The system includes a control device (34) that acquires the measurement data recorded by the sensors (20, 22, 24), The sensors (20, 22, 24) include measuring devices (20, 22, 24) for measuring the thickness and / or shape of the workpiece (44) processed with a double-sided or single-sided machine tool. The control device (34) includes an artificial neural network (34) designed to create a state vector for a double-sided or single-sided machine tool from measurement data and to compare the created state vector with at least one target state vector. The control device (34) determines by comparison that the created state vector deviates from at least one of the target state vectors, A double-sided or single-sided machine tool, further comprising an adjustment device (64) designed to control the double-sided or single-sided machine tool such that the generated state vector matches at least one of the target state vectors.

2. The double-sided or single-sided machine tool according to claim 1, characterized in that the control device (34) is designed to issue a warning message if the created state vector deviates from at least one of the target state vectors.

3. The double-sided or single-sided machine tool according to claim 1 or 2, characterized in that the adjustment device (64) is incorporated into the control device (34).

4. The adjustment device (64) controls the tool parameters and / or operating parameters of the double-sided or single-sided machine tool based on adjustment rules stored in the adjustment device (64), as described in claim 1 or 2.

5. The machine learning system evaluates the measurement data relating to the tool parameters and / or machining parameters, and controls the double-sided or single-sided machine tool based on that evaluation. The device is characterized by providing an additional artificial neural network (86) designed to create and / or modify adjustment rules stored in the adjustment device (64), A double-sided or single-sided machine tool as described in claim 4.

6. The double-sided or single-sided machine tool according to claim 5, characterized in that the adjustment device (64) is incorporated into the additional artificial neural network (86).

7. The aforementioned sensors (20, 22, 24) The distance between the first working disk (14) and the opposing support element (16), The temperature of the first working disk (14), and / or the opposing support element (16), and / or other machine parts of the machine tool, whether double-sided or single-sided, and / or the temperature and / or flow rate of the processing agent supplied to the working gap (18) for processing the workpiece (42, 44), and / or the rotational speed of the first working disk (14), the rotational speed of the opposing support element (16), and / or the rotational speed of the rotor disk rotatably mounted in the working gap (18), and / or the load between the first working disc (14) and the opposing support element (16), and / or the rotational speed, torque, and / or temperature of the rotational drive means, and / or the pressure and / or force of the means that generates deformation of the first working disk (14) and / or the opposing support element (16), and / or the thickness of the working lining of the first working disc (14) and / or the opposing support element (16), A double-sided or single-sided machine tool according to claim 1 or 2, characterized in that it includes measuring devices (20, 22, 24) for measuring.

8. The double-sided or single-sided machine tool according to claim 1 or 2, characterized in that the opposing support element (16) is formed by a second working disk (16), the first working disk (14) and the second working disk (16) are arranged coaxially with respect to each other and are rotatably driven relative to each other, and a working gap (18) is formed between the working disks (14, 16) for double-sided or single-sided machining of a flat workpiece (42, 44).

9. A system comprising at least two double-sided or single-sided machine tools according to claim 1 or claim 2, A higher-order artificial neural network (94) is provided, which is connected to the artificial neural networks (34, 86) of at least two double-sided or single-sided machine tools. The higher-order artificial neural network (94) specifies a state vector that yields an appropriate machining result for a flat workpiece (42, 44) based on data obtained by artificial neural networks (34, 86) of at least two double-sided or single-sided machine tools, A system characterized by being designed to train and optimize at least one artificial neural network (34, 86) of at least two double-sided or single-sided machine tools.

10. The method for operating a double-sided or single-sided machine tool according to claim 1, characterized in that the artificial neural network (34) is trained by inputting a number of target state vectors that result in a suitable machining outcome for a flat workpiece (42, 44).

11. The method according to 10, characterized in that the trained artificial neural network (34) is further trained during the operation of a double-sided or single-sided machine tool by inputting additional target state vectors that result in a suitable machining outcome for a flat workpiece (42, 44).

12. The method according to claim 10, characterized in that, during the operation of a double-sided or single-sided machine tool, an additional artificial neural network (76) is trained to be specialized for individual manufacturing processes by inputting a number of target state vectors that result in appropriate machining of flat workpieces (42, 44) based on the parameters of the individual manufacturing processes of the clients of the double-sided or single-sided machine tool.

13. A device designed to carry out the method described in any one of claims 10 to 12, A double-sided or single-sided machine tool as described in claim 1.

14. The double-sided or single-sided machine tool according to claim 1, wherein the adjustment device (64) is designed to control the tool parameters and / or operating parameters of the double-sided or single-sided machine tool.

15. Based on the evaluation of measurement data relating to the tool parameters and / or machining parameters by machine learning, the tool parameters and / or operation parameters of the double-sided or single-sided machine tool are controlled. A double-sided or single-sided machine tool as described in claim 5.

Images