Method and device for monitoring a synthetic thread

Abstract

Method for monitoring a synthetic thread in a melt spinning process, in which the thread is formed by extruding a plurality of filament strands, in which the thread is wetted with a fluid to hold the filament strands together, and in which the wetted thread is guided by at least one driven roller, characterized in that at least one drive parameter of the driven roller is measured and that a wetting state of the thread is determined by evaluating the measured values ​​of the drive parameter.

Description

[0001] The invention relates to a method for monitoring a synthetic thread in a melt spinning process according to the preamble of claim 1 and to a device for carrying out the method according to the preamble of claim 11.

[0002] In the production of synthetic yarns, particularly for textile applications, it is common practice to wet the multifilament yarns after melt spinning and cooling. This serves two purposes: firstly, to bind the numerous fine filaments within the yarn together, and secondly, to create an antistatic state, enabling the yarn's filament structure to be reliably guided over guide elements and dies. Typically, the yarn is wetted with a fluid after the filament strands have cooled for the first time. An oil-water emulsion or pure oil is preferably used as the fluid. However, it is essential that the yarn receives continuous and uninterrupted, uniform wetting. Yarn sections with insufficient or no wetting lead directly to defects and disruptions in the subsequent processing of the yarn, ultimately resulting in the final product.For example, it was found that insufficient wetting of the thread during subsequent dyeing leads to color defects in the fabric. Therefore, the uniformity of thread wetting in the melt spinning process is crucial for the quality of the thread.

[0003] To monitor the wetting state of the yarn, a generic method and a generic device for monitoring a synthetic yarn are known from WO 2005 / 033 697 A1, in which the wetted yarn is passed through an electric field of a sensor device. A change in the capacitance of a capacitor in the sensor device is used to detect the wetting state of the yarn. However, for this purpose, it is necessary that the sensor device be kept in an encapsulated environment, if possible, to protect it from the environmental stresses typical of a melt spinning process. Such environmental influences can negatively affect the electric field and lead to measurement errors and misinterpretations.

[0004] From EP 0 918 217 B1, a further method and device for monitoring a thread is known, in which several electric fields are generated within a sensor device through which the thread is guided. While this allows for more accurate measurement results, environmental influences must still be taken into account. Furthermore, directly detecting the wetting state of the thread requires additional sensor devices that must be integrated within the thread path, thus requiring additional installation space.

[0005] The object of the invention is to provide a generic method for monitoring a synthetic thread in a melt spinning process and a device for carrying out the method, with which continuous monitoring of the wetting state of the thread is possible directly within the melt spinning process.

[0006] Another objective of the invention is to provide a method and a device for online monitoring of the wetting state of a thread, enabling fast and direct process changes.

[0007] This problem is solved according to the invention by a method with the features of claim 1 and by a device with the features of claim 11.

[0008] Advantageous embodiments of the invention are defined by the features and combinations of features of the respective dependent claims.

[0009] The invention is based on the finding that the surface properties of the thread have a direct influence on its guidance against a moving surface. The inventor thus recognized that the relationship between the thread and the driven roller shell must change depending on the thread's surface properties. Therefore, at least one drive parameter of the driven roller is measured in order to determine the thread's wetting state by evaluating the measured values ​​of this parameter. A sensor for detecting the drive parameter of the driven roller is thus assigned to it and connected to an evaluation module for determining the thread's wetting state.

[0010] Since the wetting state of the yarn is essentially determined by the interaction between the yarn surface and the roller shell, a further development of the method is preferably implemented in which the motor current and / or motor torque of the roller motor are measured as drive parameters for the driven roller. For example, the motor torques of the roller motor for guiding a dry yarn differ significantly from those for guiding a wet yarn. This allows for continuous monitoring of the wetting state of the yarn through direct measurement of the motor torque or indirect measurement via the motor current.

[0011] Since the absence of wetting of the thread and insufficient wetting of the thread can have many causes, the method variant in which the evaluation of the measured values ​​of the drive parameter is carried out by an evaluation algorithm of a machine learning unit is particularly advantageous. This machine learning unit is trained on a large number of values ​​of the drive parameter in relation to a degree of thread wetting. In this way, artificial intelligence can be used to quickly identify and highlight possible causes of insufficient thread wetting.

[0012] However, to determine the cause of insufficient thread wetting, it is particularly advantageous to measure an additional drive parameter of a driven metering pump in the wetting device that delivers the fluid for wetting the thread. It is known that air bubbles in the lines or contamination of the preparation nozzles during fluid delivery to the thread can be significant sources of insufficient wetting. This considerably improves thread monitoring, especially for identifying the cause.

[0013] The drive parameter of the metering pump is preferably measured as a pump speed and / or a motor current of a pump motor.

[0014] In the event that the yarn is thermally treated, drawn, and drawn off by a driven galette in the melt spinning process, an additional drive parameter of the driven galette can be used to determine the wetting state of the yarn. This further increases the reliability of diagnosing the wetting state of the yarn.

[0015] The drive parameters of the driven galette are measured as motor current and / or motor torque and / or motor speed of a galette motor.

[0016] To ensure continuous monitoring of the yarn in the melt spinning process, the preferred method variant is implemented in which the measured values ​​of all drive parameters are combined into a data stream that is continuously fed to the machine learning unit. In particular, this allows for the generation of early predictions about potential wetting issues, which can be used to improve production.

[0017] To ensure that the machine learning unit continuously improves in the event of product and process changes, a variant of the procedure is provided in which the data stream is fed into a database of historical values ​​of the drive parameters. This database contains a multitude of drive parameter values ​​in relation to the degree of thread wetting and is connected to the machine learning unit. This allows for the continuous training and improvement of the evaluation algorithm for analyzing the data stream.

[0018] To enable an operator to take the fastest possible action to improve the process based on the results of the continuous evaluation of the data stream, the method variant in which the machine learning unit is connected to a user interface unit is particularly advantageous. This interface displays the wetting status of the thread and / or process instructions. This allows for rapid and direct implementation of process changes.

[0019] The device according to the invention has at least one sensor means for detecting a drive parameter of the driven roller in order to carry out the method, which sensor means is connected to an evaluation module for determining a wetting state of the thread.

[0020] Due to the multitude of possible causes and symptoms of insufficient thread wetting, the evaluation module for determining the thread's wetting status incorporates a machine learning unit with an evaluation algorithm. This allows for the use of large datasets to obtain fast and precise results when evaluating sensor signals.

[0021] To incorporate a significant source of interference during thread wetting into the monitoring process, the wetting device features a metering pump driven by a pump motor. The drive parameters of this pump are measured by a sensor connected to the evaluation module. This ensures the integration of further data for monitoring the thread's wetting status.

[0022] A further improvement in monitoring can be achieved by further developing the device according to the invention, in which at least one galette device is provided with a galette driven by a galette motor for drawing off the thread, and in which a further sensor means for detecting a drive parameter of the driven galette is connected to the evaluation module.

[0023] Due to the complexity of a melt spinning process, the further development of the device according to the invention has proven particularly effective, in which the sensor means are connected to a control unit that generates a continuous data stream of the sensor signals and is connected to the evaluation module. This allows all sensor signals of the drive parameters, acquired in real time, to be directly subjected to evaluation and analysis.

[0024] To ensure the continuous training of the machine learning unit, a database for historical drive parameter values ​​is also provided, which is connected to the evaluation module and the control unit. This allows the database to contain both offline and online data, enabling access to an improved, trained machine learning unit in the event of process changes.

[0025] For practical implementation in the process, the further development of the device according to the invention is particularly advantageous, in which a user interface is provided for displaying the wetting status of the thread and / or process instructions, which is connected to the evaluation module. This allows an operator to make continuous process improvements to ensure high thread quality at the end of the process.

[0026] The inventive method for monitoring a synthetic thread is explained in more detail below with reference to some exemplary embodiments of the inventive device for carrying out the method, with reference to the accompanying figures:

[0027] They represent: Fig. 1 schematically a first embodiment of the device according to the invention for carrying out the method according to the invention for monitoring a synthetic thread Fig. 2.1 to Fig. 2.3 Schematic representation of several time profiles of a drive parameter of the driven roller of the exemplary embodiment. Fig. 1 Fig. 3 schematically a further embodiment of the device according to the invention for carrying out the method according to the invention for monitoring a synthetic thread Fig. 4 schematically a time course of a drive parameter of a metering pump of the exemplary embodiment. Fig. 3 Fig. 5 schematically a time course of the drive parameter of a galette of the exemplary embodiment from Fig. 3

[0028] In the Fig. Figure 1 is a first embodiment of the device according to the invention for carrying out the inventive method for monitoring a synthetic filament in a melt spinning process. The embodiment shows a melt spinning device 1 comprising an extruder 1.1 and at least one spinning head 1.2, which is connected to the extruder 1.1 via a melt line 1.6. The spinning head 1.2 contains a spinning pump (not shown here) and a spinneret 1.3 arranged on the underside of the spinning head 1.2. The spinneret 1.3 has a plurality of fine nozzle openings for extruding a polymer melt, melted by the extruder 1.1, into fine filaments. The filament strands 2 exiting the spinneret 1.3 pass through a cooling duct 1.4, which is arranged within an air chamber 1.5 and has at least a partially permeable wall for the inlet of cooling air.

[0029] Below the melt spinning unit 1, a wetting unit 4 is provided, which includes a wetting thread guide 4.1. The wetting thread guide 4.1 is connected to a metering pump 4.2 to apply a continuously supplied fluid to the filament strands 2. The metering pump 4.2 is driven by a pump motor 4.3 such that a minute quantity of fluid can be continuously supplied to the wetting thread guide 4.1. The filament strands 2 are thereby joined together to form a thread 5.

[0030] Below the wetting device 4, a driven roller 6 is arranged in the thread path. The thread 5 is guided by a partial wrap around the circumference of the roller 6. The roller 6 is driven by a roller motor 6.1. A sensor 6.2 for detecting a drive parameter is assigned to the roller motor 6.1. The sensor 6.2 is connected to an evaluation module 7.

[0031] Evaluation module 7 includes a machine learning unit 7.1, which analyzes the sensor signals to determine the wetting state of the thread 5. The result of the analysis by machine learning unit 7.1 is fed to a user interface 8. The user interface 8 can be operated by an operator, so that the results of the sensor signal analysis can be immediately visualized and displayed to the operator.

[0032] During operation, the extrusion of the filament strands 2 is carried out continuously, so that they are continuously wetted with a fluid, preferably an oil-water emulsion or pure oil, by the wetting device 4. A continuous, uniform application of the fluid is necessary for the quality of the thread 5. However, during the process, disturbances can occur in the form of air bubbles in the supply lines of the wetting thread guide 4.1, or due to contamination of the wetting thread guide 4.1, or irregularities in the drive of the metering pump 4.2, leading to undesirable insufficient wetting of the thread 5. Such deficiencies in the wetting of the thread 5 have a very negative impact on the quality of the thread, especially in downstream processing. To monitor the wetting state of the thread 5, the thread 5 is guided on the circumference of the driven roller 6.The surface properties of the thread 5 in relation to the roller shell of the roller 6 can be determined by measuring at least one drive parameter of the roller 6, in particular of the roller motor 6.1, using the sensor 6.2. The motor current of the roller motor 6.1 is particularly suitable as a drive parameter and is continuously recorded by the sensor 6.2.

[0033] In the Fig. Sections 2.1 to 2.3 show some time-dependent curves of a drive parameter of the roller motor 6.1, in this case the motor current, under different operating conditions. These current curves of the roller motor 6.1 are based on a large number of measurement points within a predefined measurement period. A mathematical procedure was then used to smooth the large number of measurement points in order to eliminate the effects of the current in the Fig. 2.1 to 2.3 to obtain significant curve profiles. Here, the motor current of the roller motor 6.1 was measured by the sensor 6.2 under different operating conditions. The first operating condition represents a normal time course of the motor current. In contrast, an operating condition was chosen in which insufficient wetting was deliberately created. The time course of the motor current with insufficient wetting is shown in each case as a dashed curve.

[0034] Thus, in the Fig. 2.1 The time course of the motor current of the roller motor 6.1 under normal conditions and under conditions of insufficient wetting caused by air bubbles in the feed line of the wetting thread guide 4.1 is compared. Clearly different curves in the motor current of the roller motor 6.1 can be observed. The wetted thread surfaces and the dry thread surfaces of the thread 5 directly influence the drive torque of the roller 6 and thus the motor current.

[0035] In Fig. 2.2 The insufficient wetting is caused by contamination of the wetting thread guide 4.1. Here too, the curves of the motor current of a normal process and a process with insufficient wetting are compared and show clear differences.

[0036] In the Fig. Figure 2.3 illustrates the situation where insufficient wetting is caused by a temporary interruption of the wetting process. Such interruptions can result, for example, from a faulty metering pump 4.2. Here, too, the curves of the motor current of the roller motor 6.1 vary considerably.

[0037] In the Fig. Sections 2.1 to 2.3 illustrate only a few examples of insufficient wetting. In principle, there are multiple causes for insufficient thread wetting, which can be detected by continuously measuring a drive parameter. The motor current, motor speed, or motor torque of the roller motor 6.1 can be individually measured and monitored by the sensor 6.2. Preferably, however, all drive parameters available to the roller motor 6.1 when driving the roller 6 are recorded and analyzed.

[0038] The in the Fig. The exemplary motor current curves shown in sections 2.1 to 2.3 serve to train an evaluation algorithm of the machine learning unit 7.1. The machine learning unit 7.1 is thus provided with a large number of drive parameter values ​​in relation to the degree of thread wetting, in order to perform effective process monitoring with the aid of the evaluation algorithm. This allows for the detection of insufficient thread wetting conditions with a high degree of probability.

[0039] At the in Fig. Figure 1 of the embodiment of the device according to the invention for carrying out the inventive method for monitoring a synthetic yarn shows only the components of a melt spinning process that are essential for carrying out the invention. Usually, after wetting, the yarn is treated by stretching, twisting, or even crimping, so that the driven roller 6 is preferably arranged at the end of a treatment sequence. Since the yarn is wound onto a spool at the end of a melt spinning process, the arrangement of the driven roller 6 is therefore preferably provided directly in front of a winding station of a winding machine.Furthermore, it was found that monitoring the synthetic thread to determine the wetting state can be significantly improved by measuring and evaluating as many drive parameters as possible from other driven units involved in the process. This is described in [reference to relevant document]. Fig. 3 Another embodiment of the device according to the invention for carrying out the method for monitoring a synthetic thread is shown schematically.

[0040] The in Fig. The embodiment shown in section 3 is essentially identical to the embodiment shown in section 3. Fig. 1, so that only the differences will be explained here.

[0041] At the in Fig. In the embodiment shown in Figure 3, a galette assembly 9 with several galettes 9.1 and 9.2 is arranged between the wetting device 4 and the driven roller 6. The yarn 5 is guided around the circumference of the galettes 9.1 and 9.2. The galettes 9.1 and 9.2 are each driven by a galette motor 9.3 and 9.4 at a predetermined peripheral speed. Thus, the galette 9.1 essentially serves to draw the yarn 5 from the melt spinning device 1. The galette 9.2 can have a different peripheral speed than the galette 9.1 in order to stretch the yarn 5.

[0042] To monitor the synthetic thread 5, sensor devices 9.5 and 9.6 are assigned to the galette motors 9.3 and 9.4, respectively. The sensor devices 9.5 and 9.6 of the galette unit 9, as well as the sensor device 6.2 of the roller 6, are connected to a control unit 10. The wetting unit 4 is also assigned a sensor device 4.4, which, for example, acts as a speed sensor to detect the motor speed of the pump motor 4.3. The sensor device 4.4 is also connected to the control unit 10.

[0043] The control unit 10 is coupled in parallel with the drives and actuators (not shown here) to control the melt spinning process. Within the control unit 10, the measurement signals generated by the sensors 4.4, 6.2, 9.5, and 9.6 are combined with the measured values ​​of the respective drive parameters to form a data stream. The data stream of the sensor signals is fed by the control unit 10 to the evaluation module 7 with the machine learning unit 7.1. Simultaneously, the data stream is routed from the control unit 10 to a database 11, which contains a large number of historical values ​​of the drive parameters.

[0044] Within evaluation module 7, the sensor signal data stream is processed and fed to machine learning unit 7.1 for analysis. Within machine learning unit 7.1, the trained evaluation algorithm performs an analysis and evaluation of the sensor signals to detect the wetting state of the thread and any changes in its wetting state. The results are then fed to the user interface unit 8 to display the current wetting state of the thread or directly issue a process instruction to an operator.

[0045] Evaluation module 7 is connected to database 11 to provide further training to machine learning unit 7.1, particularly in the case of process changes or product changeovers. This allows the historical values ​​of the drive parameters from error-free or faulty processes to be supplemented by the sensor signal data stream and used for further training of the machine learning unit.

[0046] At the in Fig. In the embodiment shown in Figure 3, the machine learning unit is further trained with the evaluation algorithm beforehand in order to be able to use the drive parameters of the wetting device 4 and the galette device 9 for analysis. In the Fig. Figure 4 schematically compares the motor current of pump motor 4.3 and the wetting device 4 between a normal, fault-free process and a faulty process with insufficient wetting. The faulty process is caused by air bubbles in the wetting device 4.3. The motor current of pump motor 4.3 in the faulty process is shown as a dashed line. Significant differences between a normal process and one with insufficient wetting of the thread are also evident here. The motor current of pump motor 4.3 in the process with insufficient wetting of the thread is shown as a dashed line.

[0047] In the Fig. 5 and Fig. Figure 6 compares the motor current curves of the galette motors 9.3 and 9.4 with insufficient thread wetting and a normal process. Here too, the curve representing the faulty process is shown with a dashed line. The comparison of the motor current curves of galette 9.1 according to Fig. 5 and the galette 9.2 of the Fig. 6 reveal differences between the process disturbed by air bubbles and a normal process.

[0048] The in Fig. 4, Fig. 5 to Fig. The curves of the motor current of pump motor 4.3 and the galette motors 9.3 and 9.4 shown in Figure 6 are exemplary. In principle, such differences between a faulty and a fault-free process can also be identified by curves of the motor torques or motor speeds of the respective drives. The time intervals for recording the measurement points of the drive parameters are in the range of <100 ms. The measured values ​​of the drive parameters generated in this way are used to train the machine learning unit 7.1 and the evaluation algorithm in order to derive the following from the data stream of the sensor signals of the exemplary embodiment: Fig.3. To obtain an analysis of the respective wetting state of the thread. Due to the multitude of drive parameters resulting from the wetting unit 4, the galette unit 9, and the roller, the respective wetting state of the thread can be determined with a high degree of probability directly in the online process. This allows prolonged periods in which poor thread quality is produced to be largely minimized and eliminated. Thus, direct information exchange and direct process intervention by an operator are possible via the user interface unit 8.

Claims

[1] Method for monitoring a synthetic thread in a melt spinning process in which the thread is formed by extrusion of a plurality of filament strands, in which the thread is wetted with a fluid to hold the filament strands together and in which the wetted thread is guided through at least one driven roller, characterized by , that at least one drive parameter of the driven roller is measured and that a wetting state of the thread is determined by an evaluation of the measured values ​​of the drive parameter. [2] Method according to claim 1, characterized by , that the drive parameter of the driven roller is a motor current and / or a motor torque of a roller motor of the roller. [3] Method according to claim 1 or 2, characterized by, that the evaluation of the measured values ​​of the drive parameter is carried out by an evaluation algorithm of a machine learning unit, wherein the machine learning unit is trained by a multitude of values ​​of the drive parameter in relation to a degree of wetting of the thread. [4] Method according to any one of claims 1 to 3, characterized by , that a further drive parameter of a driven metering pump is measured to determine the wetting state of the thread, which delivers the fluid to wet the thread. [5] Method according to claim 4, characterized by , that the drive parameter of the dosing pump is a pump speed and / or a motor current of a pump motor. [6] Method according to any one of claims 1 to 5, characterized by , that a further drive parameter of a driven galette is measured to determine the wetting state of the thread, which galette guides the thread. [7] Method according to claim 6, characterized by , that the drive parameters of the driven galette are measured as a motor current and / or a motor torque and / or a motor speed of a galette motor of the galette. [8] Method according to any one of claims 1 to 7, characterized by that the measured values ​​of the drive parameters are continuously supplied to the machine control unit as a data stream. [9] Method according to claim 8, characterized by , that the data stream is fed into a database for historical values ​​of the drive parameters, which contains a large number of values ​​of the drive parameters in relation to a degree of wetting of the thread and which is connected to the machine learning unit. [10] Method according to any one of claims 1 to 9, characterized by that the machine learning unit is connected to a user interface unit through which the usage status of the thread and / or process instructions are displayed. [11] Device for carrying out the method according to one of claims 1 to 10, comprising a melt spinning device (1), a wetting device (4) and at least one roller (6) driven by a roller motor (6.1) for guiding the thread (5), characterized by a sensor means (6.2) for detecting a drive parameter of the driven roller (6), which sensor means (6.2) is connected to an evaluation module (7) for determining a wetting state of the thread (5). [12] Device according to claim 11, characterized by , that the evaluation module (7) for determining a wetting state of the thread (5) includes a machine learning unit (7.1) with an evaluation algorithm. [13] Device according to claim 11 or 12, characterized by, that the wetting device (4) has a metering pump (4.2) driven by a pump motor (4.3) and that a drive parameter is detected by a sensor means (4.4) which is connected to the evaluation module (7). [14] Device according to any one of claims 11 to 13, characterized by , that at least one galette device (9) is provided with a galette (9.1) driven by a galette motor (9.3) for drawing off the thread (5) and that a further sensor means (9.5) for detecting a drive parameter of the driven galette (9.1) is connected to the evaluation module (7). [15] Device according to any one of claims 11 to 14, characterized by , that the sensor means (4.4, 6.2, 9.4, 9.5) are connected to a control unit (10) through which a continuous data stream of the sensor signals can be generated and which is connected to the evaluation module (7). [16] Device according to any one of claims 11 to 15, characterized by , that a database (11) is provided for historical values ​​of the drive parameters, which is connected to the evaluation module (7) and the control unit (10). [17] Device according to any one of claims 11 to 16, characterized by , that a user interface device (8) is provided for displaying usage states of the thread (5) and / or process instructions, which is connected to the evaluation module (7).

Images