CN116981943A - Method for monitoring a process for producing synthetic yarns - Google Patents

Abstract

The application relates to a method for monitoring a manufacturing process of a synthetic yarn, wherein yarn tension of the yarn is continuously measured. The measured values of the yarn tension are evaluated and a plurality of defect maps are generated, and the defect maps of the yarn tension are assigned corresponding defect causes during the production process. In order to be able to weight the defect elimination in the case of a plurality of defect causes, it is provided according to the application that the defect causes are detected individually and their occurrence within a time window is recorded as the interference frequency.

Description

Method for monitoring a process for producing synthetic yarns

Technical Field

The present application relates to a method for monitoring the manufacturing process of a synthetic yarn according to the preamble of claim 1.

Background

One type of method for monitoring the manufacturing process of synthetic yarns is known from WO2019/137835 A1.

The synthetic yarns are typically continuously monitored during manufacture and handlingProduct parameters and/or process parameters in order to thus obtain as stable a process guidance as possible and in particular as stable a yarn product quality as possible. In particular in the manufacture of textured yarnsIn this case, monitoring of the yarn tension on the running yarn has been demonstrated in order to detect process disturbances and/or product fluctuations. In the known method for monitoring a manufacturing process, the yarn tension is measured continuously on the running yarn. The measurement signal generated in this case of the yarn tension is compared with a threshold value for the permissible yarn tension in order to identify a so-called defect map and can be used for further analysis. The individual defect causes in the production process are identified from the defect map by means of a machine learning program, for example, in order to be removed by the operator in order to finally maintain the yarn quality at a uniform level.

In practice it has now been determined that a plurality of defect maps and thus a plurality of defect causes occur during the production process, which cause influences the production process more or less strongly. Thus, the causes of defects in the manufacturing process are known, which require immediate intervention, for example, when the yarn breaks.

Disclosure of Invention

The object of the present application is to further develop a method of the generic type for monitoring a production process of synthetic yarns in such a way that effective operability of the production process for ensuring uniform yarn quality can be achieved.

It is a further object of the present application to minimize the presence of operator instructions for controlling multiple machining positions.

According to the application, this object is achieved in that the defect causes are individually detected and their occurrence within a time window is recorded as the interference frequency. Within the scope of the application, a defect map is preferably understood to be a measurement signal curve ("Snap-shot") of the yarn tension in which the measurement signal of the yarn tension falls below or exceeds a predefined limit value threshold and/or in which the measurement signal of the yarn tension leaves a range defined by a predefined limit value (limit value violation). It has proven advantageous if the defect map is limited in time, for example up to 20 seconds, preferably up to 10 seconds. Furthermore, it has proven to be advantageous if the point in time of the limit value violation is located approximately in the middle of the defect map. The point in time of the limit value violation may be located at the edge of the defect map.

Advantageous further developments of the application are defined by the features and feature combinations of the dependent claims.

The application is also not evident from the methods known from DE4329136A1 for adjusting the yarn tension on a false twisting device. In this case, the frequency of the abnormal yarn tension yarn state is detected. Thus, although the frequency of yarn tension defects can be recorded, the frequency does not contain a cue for the likely cause of the defect. However, the defect cause cannot be deduced from the defect frequency.

The application has the particular advantage that the cause of the defect can be distinguished in order to evaluate the influence of the corresponding interference source on the manufacturing process and on the yarn quality. Thus, the occurrence within a time window can be recorded as the disturbance frequency of the defect cause. Thus, the interference frequency of the corresponding defect cause can be used as a parameter in order to initiate a differentiated intervention for eliminating the defect.

Since a large number of different types of defect causes may occur during the manufacture of the synthetic yarn, which occur for example in the raw material, in the machine or in the settings etc., their influence on the process and the yarn quality is very different. In order to be able to take account of this difference in defect causes, a variant of the method is particularly advantageous in which the interference frequency of each of the defect causes is assigned a respective permissible frequency limit value. It is therefore possible to take into account the occurrence of critical defect causes and the occurrence of less critical defect causes with different weights when intervening in the manufacturing process. An immediate intervention in the manufacturing process is of course required for yarn breakage. However, this is not the case when the component is contaminated or worn.

In this case, it has proven to be particularly advantageous if the frequency limit value of the disturbance frequency of one of the defect causes contains a plurality of defect causes which are considered to be permissible during the production process. For each defect cause in the process, the frequency limit value can therefore be determined only from the number of defect causes present, in order to be able to carry out a targeted intervention in the production process.

For further differentiation, a method variant is provided in which the frequency limit value of the disturbance frequency of one of the defect causes relates to the time period and/or the processing position. Thus, for example, in the production of crimped yarns, the yarn is usually drawn off from a yarn supply bobbin and crimped. Thus, for example, the operating time of the yarn feed bobbin can be selected as a time period in order to evaluate the cause of defects occurring therein with respect to the frequency of disturbances thereof. Furthermore, it is common in this case for the individual yarns to be guided in parallel next to one another in a plurality of processing stations, so that in the case of the defect reasons in the respective processing stations a separation is likewise advantageous.

In order to eliminate the cause of defects occurring during the production process, a variant of the method is provided in which a report is transmitted to the person when one of the frequency limit values of the interference frequency is exceeded. Thus, a targeted intervention of the manufacturing process by the responsible personnel is ensured.

A further development of the application is particularly advantageous since the occurrence of different forms of defect causes can occur, which makes the assignment of frequency limit values difficult, wherein the disturbance frequencies of the defect causes are evaluated by means of algorithms based on statistical methods and machine learning methods in order to identify frequent sequences of disturbance frequencies and/or anomalies in the disturbance frequencies. The method variant has the particular advantage that the frequency of the defect cause that is prominent is autonomously recognized by the system without the necessity of defining a predefined frequency limit value.

Such machine learning systems are based on the occurrence of continuous learning. For this purpose, a variant of the method is particularly advantageous in which a sequence of interference frequencies is reported to a person in the population and the sequence of interference frequencies is evaluated and evaluated by the person. Thus, not only is a quick access ensured for stopping the cause of the defect during manufacturing, but also feedback to the machine learning system is ensured.

In order to be able to find possible quality differences when producing a plurality of yarns in a plurality of processing positions, a further development of the application is provided in which the plurality of yarns are guided and monitored in parallel in the plurality of processing positions during the production process and the frequency of disturbances of the defect cause in the processing positions is analyzed in relation. In addition to the mass of the yarn, the mass of the individual processing locations can also be deduced therefrom.

Depending on the defect map and the defect cause derived therefrom, different processing actions may be required to be focused on by different populations. Thus, maintenance personnel are responsible for servicing the machine. In contrast, process personnel are responsible for setting up the machine. In contrast, the raw materials for the manufacturing process are provided by the operator. In order to be able to intervene as quickly and with greatest possible specificity when eliminating a defect cause in a production process, a variant of the method is therefore provided in which one of a plurality of groups of people is assigned to the defect cause in each case and the group is reported if the frequency of interference of the defect cause concerned exceeds a limit value or if the sequence/abnormality of the frequency of interference is notified. Time delays in eliminating the cause of the defect can thus advantageously be avoided.

In order to automate the production process, a method variant is provided in which a defect cause is assigned to one of a plurality of control commands, which are activated if a limit value for the disturbance frequency of the defect cause concerned is exceeded or if a sequence/abnormality of the disturbance frequency is signaled. Thus, automated interventions in eliminating defects during manufacturing may be advantageously integrated.

The application has the particular advantage that it is possible to weight a plurality of defect causes in a single manufacturing process, the weighting of the defect causes having an effect on the quality of the manufacturing process and the quality of the yarn, for example. In particular in the case of defect maps with large amounts of data, targeted and rapid intervention to eliminate the cause of the defect can thus advantageously be achieved.

Drawings

The method according to the application for monitoring the manufacturing process of a synthetic yarn is explained in more detail below with reference to the drawings by means of an embodiment of the manufacturing method.

In the figure:

figure 1 schematically shows an embodiment of the processing position of a textile machine for manufacturing synthetic crimped yarns,

figure 2 schematically shows a flow chart for monitoring a manufacturing method according to the embodiment according to figure 1,

fig. 3 shows schematically a plurality of defect maps, each having a measurement signal profile of yarn tension for different defect causes,

fig. 4 schematically shows a flow chart for monitoring a manufacturing method according to the embodiment according to fig. 1.

Detailed Description

Fig. 1 schematically shows an embodiment of a processing position of a textile machine 1, in this case a texturing machine for producing crimped synthetic yarns. Such a texturing machine has a plurality of such identically configured machining positions. In this connection, only one of the processing positions of the textile machine is described in detail here.

Fig. 1 schematically shows a processing position 1 and a winding position 2 of a textile machine. The processing station 1 has a creel 4 in which a feed bobbin 5 and a standby bobbin 6 are held. The yarn feed bobbin 5 provides a yarn 3 which is transported into the processing location 1 for drawing and texturing. The yarn feed bobbin 5 is also called a so-called POY bobbin, because it was previously produced in a melt spinning process and was wound with freshly spun POY yarn. The yarn end of the yarn supply bobbin 5 and the yarn start of the standby bobbin 6 are connected to each other by a yarn knot. This achieves a continuous withdrawal of the yarn 3 after the end of the yarn feed bobbin 5. The yarn end of the standby bobbin 6 is then connected to the yarn start of the new yarn feed bobbin 5.

The yarn 3 is drawn off from the yarn supply bobbin 5 by the first feed mechanism 7.1. The feed means 7.1 is driven via a drive means 8.1. In this embodiment, the feed mechanism 7.1 is formed by a driven yarn guide roller and freely rotatable rollers, which are multiply wound with yarn. In a further process, a heating device 9, a cooling device 10 and a deformation assembly 11 are arranged downstream of the feed mechanism 7.1. Texturing unit 11 is preferably designed as a friction false twister in order to produce a false twist on the multifilament yarn, which false twist leads to a crimping of the individual filaments of the yarn. For drawing the yarn, a second feed mechanism 7.2 is arranged downstream of the texturing module 11, which is driven by a drive 8.2. Structurally, the feed mechanism 7.2 is identical to the first feed mechanism 7.1, wherein the second feed mechanism 7.2 operates at a higher peripheral speed for the drawing of the yarn. Within the processing station 1, the synthetic yarn 3 is thus deformed and simultaneously drawn.

After crimping the yarn 3, the yarn is guided to the winding position 2 by a third feed mechanism 7.3. The feed mechanism 7.3 is driven by a drive means 8.3. The feed mechanism 7.3 is designed as a so-called pinch feed mechanism, which has a driven shaft and pinch rollers. The yarn 3 is guided in a clamping gap on the circumference of the shaft. The winding station 2 has a creel 13 which supports a bobbin 14. The creel 13 is pivotably constructed and can be operated manually or automatically for changing the bobbins 14. The creel 13 is provided with a drive roller 15 which is driven by a roller drive 15.1. In order to lay the yarn on the circumference of the bobbin 15, the winding position 2 is provided with a switching device 12 which has a switching yarn guide that can be driven. For this purpose, the change-over yarn guide is driven in a reciprocating manner by a change-over drive 12.1. The switching drive 12.1 and the roller drive 15.1 of the winding position 2 are embodied as separate drives and are connected to the machine control unit 16. Likewise, the drives 8.1, 8.2 and 8.3 of the feed mechanisms 7.1, 7.2 and 7.3 and the deformation drive 11.1 of the deformation assembly 11 of the machining position 1 are designed as separate drives and are coupled to the machine control unit 16.

For monitoring the production process in the processing station 1, the yarn tension on the yarn 3 is continuously measured in a measuring point between the feed mechanisms 7.2 and 7.3. The measuring points are shown in the illustrated position, for example, between the supply means 7.2 and 7.3. Preferably, the yarn tension is also measured during the yarn before the feed mechanism 7.2. For this purpose, a sensor device 17 is provided, which has a yarn tension sensor 17.1 and a measuring signal receiver 17. The sensor device 17 is connected to a data analysis unit 18. The data analysis unit 18 is coupled to a transmitter 19 which is connected to a suitable receiving system for transmitting information and signals using wired or wireless transmission techniques. The data analysis unit 18 comprises a plurality of software modules in order to analyze the measurement signals of the sensor device 17 and the data obtained therefrom for monitoring the manufacturing process.

For this purpose, fig. 2 shows a first schematic diagram of the data evaluation unit 18 in the form of a flow chart.

As can be seen from the illustration in fig. 2, the sensor device is provided with a yarn tension analysis module 20. The yarn tension analysis module 20 continuously receives the measurement signals of the sensor device and generates a so-called defect map. The defect map defines an abnormal measurement signal curve of yarn tension. Such defect maps are always produced in a data-technology manner when the yarn tension measured on-line leaves the range defined by the predefined limit values during the production process. If the yarn tension of the yarn is significantly moved outside a defined range, it has an effect on the quality of the yarn and thus on the quality of the product produced. Accordingly, a plurality of defect maps are generated by the yarn tension analysis module 20. However, the defect map is based on a defined cause, which influences the yarn manufacturing process as a disturbance variable. In order to be able to assign the basic defect cause to the defect map, a machine learning module 21 is provided. The machine learning module 21 has an algorithm to identify a cause of a defect of the defective image based on a statistical method and a machine learning method. Such a machine learning module is suitable for determining the cause of a defect that exists behind the defect map due to a reinforcement learning phase that has passed in advance. The analysis is performed continuously, so that corresponding defect causes are generated from the provided defect map. In this way a plurality of different defect causes can be determined, which are recorded in the defect cause module 22. The defect cause is indicated by the characteristic letter FC in fig. 2. The cause of the defect is thus also continuously determined by continuously monitoring the yarn tension and the continuously generated defect map. Since not every defect cause requires direct intervention in the manufacturing process, the frequency of its occurrence is detected within a time window or within a processing location. For this purpose, an interference frequency module 23 is provided, in which the interference frequency of the defect cause is continuously determined. Here, the disturbance frequency is the sum of defect causes and is referred to as Σfc.

In order to determine which defect cause and which disturbance frequency of the defect cause requires intervention for defect elimination, a frequency limit value is assigned to each defect cause. The frequency limit value determines the maximum permissible occurrence of the defect cause within a time period or within a processing position. These limit values, denoted L, are stored in the comparison module 24 for each defect cause. The frequency limit value of the interference frequency is then compared with the interference frequency detected per time period or per processing position.

For the case where the disturbance frequency has not been reached and the limit value has not been exceeded, the occurrence of the cause of the defect concerned is further recorded. When the limit frequency is exceeded, a message, alarm or control signal is generated by the output module 25 and transmitted via the transmitter 19 to a person or control unit in the population.

In this case, it is also possible to require a direct action by a specific person, depending on the cause of the defect concerned and the frequency of its disturbance. Thus, maintenance personnel are typically responsible for maintenance on the machine components. In contrast, the setting of the machine is performed by a process person. Thus, the information may be transmitted by the data analysis unit 18 to the group responsible for the information.

In the production process according to the embodiment of fig. 1, a plurality of defect maps occur, the defect causes of which require different interventions in the production process. Fig. 3 shows a typical defect map and several embodiments of the defect cause thereof, the disturbance frequency of which is characterized by a predefined frequency limit value.

In fig. 3, the individual defect maps are indicated by capital letters A, B, C, D, E and are shown by dedicated measurement signal curves for yarn tension. Defect map a shows a typical yarn break that may occur in the processing location. Such a defect cause leads to an interruption of the manufacturing process in the machining position and in this embodiment a frequency limit value of the disturbance frequency l=1 is obtained. The first occurrence of such a defect cause has led to a defect report to the process personnel in order to eliminate yarn breakage.

Defect map B shows a typical measured signal curve at junction overrun (knotenu berlauf). The connection between the yarn supply bobbin and the reserve bobbin is thus established by the yarn knots which are visible during the yarn tension measurement. This defect cause allows one to occur during the running time of the yarn feed bobbin. Therefore, here, the interference frequency l=2 is selected as the frequency limit value, wherein the frequency limit value L relates to a time period of 60 minutes. Thus, for example, the run time of the yarn feed bobbin may be 60 minutes.

The defect map C signals the cause of the defect, and the yarn is guided in the pinch feed mechanism outside the pinch gap between the pinch roller and the drive shaft. Such a defect can be very adversely affected, in particular, when winding the yarn and changing the bobbin. In this regard, a frequency limit value l=4 is given as an interference frequency in two minutes. Thus, a very short period of time is used in order to compare the interference frequency with the frequency limit value.

Accordingly, the defect map D signals defects in the raw material of the yarn feeding bobbin. Thus, for example, fineness fluctuations in the feed yarn can be disadvantageously detected in the measurement signal of the yarn tension. The cause of such defects can be evaluated with a smaller weight. For this purpose, the frequency limit value of the control frequency is defined as l=10 in a period of 5 minutes.

The defect map E shows a measured signal curve of the yarn tension, which indicates a soiled guide surface of the cooling rail of the cooling device. If the frequency limit value of the disturbance frequency l=3 is exceeded within a period of 30 minutes, this defect cause can, for example, cause an intervention in the process. Most of the reasons for such defects occur very accidentally first, so that accumulation in a short period of time requires intervention.

In the exemplary embodiment shown in fig. 3, the frequency limit values each relate to a processing position of the textile machine. The time period on which the frequency limit value is based is a simple example and is not decisive for the process. It is important here that a defect cause is identified in the case of a plurality of defect maps and defect causes, which decisively influence the quality of the product production.

However, the method according to the application can also be advantageously implemented such that the occurrence of the disturbance frequency, which is responsible for the defect, is analyzed directly, irrespective of the frequency limit value. For this purpose, a flow chart of an alternative embodiment of the data evaluation unit 18 is schematically shown in fig. 4. This embodiment is essentially the same as the above-described embodiment according to fig. 2, so that only the differences are explained here and otherwise reference is made to the above description.

In the flow chart shown in fig. 4, the defect cause module 22 is provided with a further machine learning module 26. The machine learning module 26 has an algorithm that performs an analysis of the interference frequencies based on statistical methods as well as machine learning methods to identify interference frequencies or sequences of interference frequencies that are prominent in anomalies. Such a sequence or anomaly of the disturbance frequency of the defect cause concerned is thus used to display to a person, preferably to an operator, via the output module 25, so that the operator can analyze and evaluate the defect cause. On the one hand, therefore, defect causes can be eliminated in order to ensure an optimal quality in the production process. On the other hand, experience by the operator is helpful in providing corresponding feedback to the machine learning module 26. Such a machine learning module thus obtains a phase in which targeted control instructions of the in-process intervention can be implemented with a high probability. Thus, automated manufacturing monitoring may be achieved.

Claims (10)

1. Method for monitoring a manufacturing process of synthetic yarns, wherein yarn tension of the yarns is measured continuously, measured values of the yarn tension are analyzed and a plurality of defect maps are generated, and the defect maps of the yarn tension are assigned corresponding defect causes during the manufacturing process, characterized in that the defect causes are detected individually and the occurrence of the defect causes within a time window is recorded as a disturbance frequency accordingly.

2. The method according to claim 1, characterized in that the interference frequency of each of the defect causes is assigned a respective allowable frequency limit value.

3. The method according to claim 2, characterized in that the frequency limit value of the disturbance frequency of one of the defect causes is the number of occurrences of the defect cause considered to be allowed during the manufacturing process.

4. A method according to claim 2 or 3, characterized in that the frequency limit value of the disturbance frequency of one of the defect causes relates to a time period and/or a machining position.

5. The method according to any of claims 2 to 4, characterized in that a report is transmitted to one person in the population in case one of the frequency limit values of the interference frequency is exceeded.

6. Method according to claim 1, characterized in that the disturbance frequency of the cause of the defect is analyzed by means of algorithms based on statistical methods and machine learning methods in order to identify frequent sequences of disturbance frequencies and/or anomalies of disturbance frequencies.

7. The method of claim 5, wherein the sequence of interference frequencies is reported for one of the people and is assessed and evaluated by that person.

8. Method according to any one of claims 1 to 7, characterized in that the defect cause is assigned to one of a plurality of people, respectively, which are notified in case the limit value of the disturbance frequency of the defect cause concerned is exceeded or in case the sequence/abnormality of the disturbance frequency is notified.

9. Method according to any one of claims 1 to 8, characterized in that a plurality of yarns are guided and monitored in parallel in a plurality of processing locations during the manufacturing process and the frequency of interference of the cause of defects in the processing locations is analyzed in relation.

10. Method according to any one of claims 1 to 9, characterized in that the defect cause is assigned to one of a plurality of control commands, which control commands are activated in the event of exceeding a limit value for the disturbance frequency of the defect cause concerned or in the event of a sequence/abnormality informing the disturbance frequency.