JP2001322768A - Control method and device for false twister - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a control method and device for a false twister capable of identifying some of abnormality factors of an original yarn package using an analysis in a frequency region of untwisting tension. SOLUTION: This control method for the false twister is characterized in that a characteristic value found from a component of a specific frequency region related to the characteristics of the original yarn package of the untwist tension according to each original yarn package, a case where the characteristic value is changed from a preset control value by a prescribed value or more is detected as a control phenomenon to store the generation time point, and the time series generation distribution of the control phenomenon about the original yarn package of a designated control range is displayed according to the request. This control device for the false twister implements it.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and an apparatus for managing a false twisting machine, which provide information on the quality control of a yarn package processed by the false twisting machine from operation information of the false twisting machine. The present invention relates to a method and an apparatus for managing a false twisting machine that provide quality control information of a yarn package based on untwisting tension in the false twisting machine.

[0002]

2. Description of the Related Art As is well known, a false twisting machine receives a yarn package as a raw yarn package from a yarn forming process of forming a yarn having desired properties from a polymer, winding the yarn into a yarn package, and This is configured to be subjected to false twist processing to form a processed yarn and formed into a product package.

[0003] This false twisting machine specifically has a configuration in which necessary processing equipment such as various guides, rollers, heaters, and false twisting disks are arranged in a section having a total length of 8 to 10 m. In addition, a plurality of yarn packages are arranged for each weight, and the yarn packages are switched for continuous production.

In such a false twisting machine, quality control of a product package to be manufactured is important, and is disclosed in Japanese Patent Application Laid-Open Nos. Hei 6-264318 and Hei 7-138.
As disclosed in Japanese Patent Publication No. 828, etc., by monitoring the untwisting tension related to the product quality online in a time series, it is possible to judge the quality of the product package in the false twisting machine or to control the quality such as rating. It is carried out.

[0005]

However, in the above-described conventional quality control based on abnormal fluctuation of untwisting tension, it is difficult to find out the cause of the abnormality.

[0006] On the other hand, one of the inventors of the present invention collaborated with another researcher to examine the factor of variation of the untwisting tension as described in the specification of Japanese Patent Application No. 11-21695 filed earlier. In addition to the abnormalities of the false twisting machine itself, the untwisting tension changes relatively greatly depending on the yarn properties of the yarn of the yarn package, and the characteristics of the yarn package such as the winding state during spinning. And also found that these factors can be identified by analyzing the fluctuations in the frequency domain of untwisting tension. Then, a method was proposed in which the quality of the processed yarn, specifically, the quality of the false twisted yarn was managed by monitoring the time-sequential fluctuation in a specific frequency region by performing a fast Fourier transform of the untwisting tension.

According to this method, it is possible to identify whether the cause of the quality abnormality is on the false twisting machine side or on the raw yarn package side, and a quick solution is possible.

However, in the yarn-making process for producing the yarn package, the cause of the abnormality in the yarn package obtained by the above-mentioned method is determined in order to quickly solve the problem. Specifically, it is necessary to identify factors such as those related to the physical properties of the yarn or, in other words, those related to winding, and a management method capable of identifying such factors has been demanded.

An object of the present invention is to solve the current situation, and to manage a false twisting machine capable of identifying some of the causes of abnormalities in a yarn package using the above-described analysis of the untwisting tension in the frequency domain. It is to provide a method and a management device.

[0010]

The above object is achieved by the present invention described below. That is, according to one aspect of the present invention, in a method of managing a false twisting machine, a characteristic value obtained from a component in a specific frequency region related to the characteristic of the untwisted tension of the untwisted yarn for each untwisted yarn is processed. From a predetermined management value to a predetermined time interval until the end of machining, a case where this characteristic value fluctuates by a predetermined value or more from a predetermined management value is detected as a management event, and the occurrence time is stored. And displaying a time-series occurrence distribution of management events of the original yarn package.

Another aspect of the present invention is a management device for a false twisting machine which implements this managing method, wherein a tension detecting means for detecting a untwisting tension of false twisting, and a detected untwisting tension signal Fourier transform means for performing a Fourier transform at a predetermined time interval to convert to a spatial signal in the frequency domain, a characteristic value extracting means for determining a characteristic value from a signal component in a specific frequency domain in which the spatial signal is set, and the obtained characteristic value Event detection means for comparing a set value with a set control value, detecting a case where the variation is equal to or larger than a control limit value as a control event, and storing the occurrence point of time for each yarn package; Display means for displaying, in parallel, a time-series occurrence distribution of management events of the package, the management device for the false twisting machine.

In the above-described apparatus of the present invention, the Fourier transforming means includes a digitizing means for converting the untwisted tension signal into a digital signal and at least a predetermined time interval determined by a desired frequency band and resolution at the time of Fourier transform. Tension storage means for storing the digitized untwisting tension signal, and converting the untwisting tension signal between the predetermined time intervals stored at the predetermined time interval into a spatial signal in the frequency domain by the fast Fourier transform method. A fast Fourier transform means is preferable in terms of stabilization processing and online processing. However, when the frequency domain is fixed, a band-pass filter or the like can be applied in place of the fast Fourier transform means.

Further, in the present invention, when displaying the management events of the yarn package within the designated management range, it is preferable to display the management events in a graph on the same time basis in view of quality evaluation and early detection of an abnormal part. preferable.

[0014] The yarn package in the above-mentioned predetermined control range is usually manufactured in a series of the same manufacturing conditions for a certain period in the spinning process for forming it, in other words, a plurality of yarn packages of the same lot, or A plurality of raw yarn packages manufactured under the same manufacturing conditions with the same weight in the spinning process for manufacturing the same are preferable in terms of quality control, particularly in terms of analysis of the cause of abnormality, but should be appropriately selected according to the purpose of other management. Can be.

[0015] From the viewpoint of analysis of the factor of the original yarn, the control event may include, in addition to the characteristic value of the untwisting tension described above, a time-series variation of the untwisting tension, occurrence of yarn breakage, generation of fluff, and the like. It is preferable to include them.

Further, as the management device, a plurality of yarn packages are arranged per weight, and yarn switching detection means for detecting the switching is provided. And the configuration that automatically determines the end point of machining is the reliability of data,
It is preferable from the viewpoint of productivity and the like.

The characteristic value from the frequency component of the frequency range of 0.01 to 0.3 Hz is an index of the thickness unevenness of the raw yarn.
%, And the characteristic values from the frequency range of 0.6 to 1.4 Hz have a good correlation with the OPU, which is an index of the amount of oil agent adhering to the yarn, and these frequency regions are specified for monitoring these characteristics of the yarn. It is preferable to set the frequency range, but since the frequency slightly varies depending on the spinning conditions of the original yarn package, it is more preferable to set the frequency range in reference to the test result of the abnormal yarn. The characteristic value is preferably used because the integrated value obtained by integrating the components in the characteristic frequency region has good measurement stability and good correlation in the above characteristics, but the peak value and average value of other components are also applicable. It is possible.

[0018]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The details of the present invention will be described below with reference to the drawings. FIG. 1 is an explanatory diagram of a weight configuration and a sensor mounting position of the false twisting machine of the embodiment. FIG. 2 is a block diagram of the configuration of the management device of the embodiment, FIGS. 3 and 4 are flowchart diagrams of the management unit of the machine management device of the embodiment, and FIG. 5 is a flowchart diagram of the central management unit of the central management device of the embodiment. FIG. 6 is an explanatory diagram of a display example of the embodiment.

As is well known from the above-mentioned publications and the like, the false twisting machine usually has a processing weight for processing one yarn of 200 to 200.
FIG. 1 shows the basic processing flow configuration of one weight of this false twisting machine.

In FIG. 1, reference numerals 1a and 1b denote POs set in a creel portion (not shown) and manufactured in a spinning process.
An original yarn package made of a synthetic yarn such as Y (partially oriented yarn). The yarn Y is drawn out of the original yarn package 1a by the supply roller 2 and supplied to the false twisting section. In this example, as shown in the figure, two yarn packages 1a and 1b are arranged per one spindle, and in this example one yarn package, the tail yarn derived from the innermost layer of 1a and the other yarn package are arranged. In the present example, when the yarn Y of the yarn package 1a currently being supplied is lost in the figure, the yarn supply package is automatically connected to the standby yarn drawn out from the outermost layer of the yarn package 1b. The package is switched to the package 1b so that the yarn is continuously supplied.

The yarn Y supplied to the false twisting section is twisted by a false twisting unit 4 arranged on the upstream side of the feed roller 3, and the false twist goes back to the twisting guide 5. Then, the false twist that has been traced back to the twist stopping guide 5 is heat-set by the first heating device 6 to form a false twist shape. The cooling devices 8a and 8b cool the heated yarn Y. At this time, the second heating device 7 is applied as necessary to adjust the physical properties of the processed yarn.

Then, the yarn Y finally formed into a false twist shape, in other words, the drawn false twisted yarn Y is applied to the guide roller 9.
And 10 to a winder 11 and wound as a product package 12. The winding machine 11 is usually configured to automatically perform the doffing, and can perform continuous processing from yarn supply to winding.

The above-described units such as the false twisting unit, the heating device, the cooling device, and the winder are the same as known devices.
The details are described in the above-mentioned publications and the like, and the detailed description is omitted.

The following management apparatus according to the present invention is provided in the false twisting machine configured as described above. This management device measures the untwisting tension of the yarn, the presence or absence of fluffing of the supplied yarn, and the switching timing of the yarn package online,
Provided as time-series information summarized for each yarn package, and analyzed in conjunction with the production history of the yarn to detect abnormality in the yarn, and in turn, the cause of the abnormality in the spinning machine that spins the yarn in the spinning process Provide information that can be used to determine
The following configuration is shown in FIGS.

Each detection means which is the basis of the management apparatus is provided in the false twisting machine as shown in FIG. A tension detecting means 31 is provided between the feed rollers 3 downstream of the false twisting unit 4 to detect the untwisting tension online.
Also, the trailing yarn of the yarn package 1a and the yarn package 1b
The knot detecting means 32 for detecting the passage of the knot (knot) is provided at the yarn knot tied to the yarn, and the switching of the yarn supply package is detected by the passage of the knot. Further, a fluff detecting means 33 for detecting the fluff of the yarn Y is provided between the yarn feeding roller 2 and the twist stopping guide 5, and the presence or absence of the fluff of the original yarn is monitored. The tension detecting means 31, the knot detecting means 32, the fluff detecting means 33
Are known, and commercial products and the like can be applied as they are.

As shown in FIG. 2, the tension detecting means 31 of each weight is switched at a predetermined sampling cycle by a scanning circuit 35 via each filter circuit 34 in order to remove the noise, and a common analog / digital conversion circuit is used. (A / D conversion circuit) 36, is converted into a digital signal, and is input via an interface circuit 37 to a machine management device 38 provided for each machine and comprising a computer for managing the machine. . As is well known, the sampling period is selected based on the sampling theorem so that significant information is not lost from the tension signal. The output signals of the knot detecting means 32 and the fluff detecting means 33 of each weight are pulse signals indicating the passage of the knots and the presence or absence of fluff, in other words, digital signals, which are directly input to the machine management device 38 via the interface circuit 37. Is done.

In addition, a bar code reader 40 is connected to the machine management device 38, whereby the management information necessary for the later-described processing of the yarn forming process of the yarn packages 1a and 1b set on the creels of the respective weights is performed. More specifically, yarn production management information such as a production machine stand number, its weight number, a doff number, or a production time is input. In the present example, this input is performed by reading the management card attached to each of the yarn packages 1a and 1b into the creel with the barcode reader 40. Alternatively, a scanner or the like may be used.

The machine management device 38 is connected to a higher-level central management device 39 which is common to the plurality of machine management devices 38, and as described later, it takes a relatively long time for processing, and the need for immediate processing is low. The processing is performed by the central processing unit 39. With such a hierarchical configuration, high-speed processing of a part that requires online processing is realized. The machine management device 38 is configured by a microcomputer, stores machine management means including the flowcharts shown in FIGS. 3 and 4, and performs the following processing. The machine management means is configured to perform two tasks simultaneously in the background and the foreground as shown in FIGS.

In the background, a data collection task is performed. The data collection task has a fixed period of 10 in this example.
An interrupt command is input every millisecond, and data is collected as follows when the interrupt command is input. First, a knot signal / fluff signal scanning step is performed, and a knot passage signal of the knot passage detection means 32 of all the weights of the machine and a fuzz generation signal of the fuzz detection means 33 are scanned.
The presence or absence of fluff occurrence is read, and if yes, the date and time of occurrence and the weight number are stored.

Next, tension data collection is performed as follows.
First, in order to collect tension data from the tension detecting means 31 of each weight, the first weight is set to the weight number of the scanning circuit 35 in the A / D conversion step, and the A / D conversion circuit 36 is set.
To start the A / D conversion of the tension signal. Thereby, A / D conversion of the tension signal from the tension detecting means 31 of the first weight is performed. When the A / D conversion is completed, the process proceeds to a tension data storage step, and the obtained tension data is stored in the tension data storage area together with the collection date and time and the weight number.

Next, in this embodiment, a moving average is obtained in order to obtain a reference value for fluctuation determination based on the tension data. The moving average is a time-series average of a predetermined number of data as is well known. In this example, the moving average is obtained by taking a moving average of 120 pieces of data. Therefore, it is necessary to collect 120 pieces of data until a steady state in which a normal moving average value is obtained, and it takes 1 time in time. It takes 2 seconds.

Therefore, an average data number determination step is provided as shown in the figure to determine whether or not the number of data collected for the moving average has reached 120 required for the moving average.
In the case of "No" of less than 20 pieces, the process proceeds to a knot / fluff discrimination step to be described later during this time, that is, until the steady state.

When the number reaches 120, the result is "Yes". In this example, a moving average calculation step is performed, and a moving average is obtained to obtain a reference value for fluctuation determination. As described above, in this example, the moving average is obtained by taking a moving average of 120 data, and therefore, the latest 120 data is stored in the tension data storage area for each weight. Then, in this step, the latest 120
The average value of the pieces of data is calculated, the result is stored as the reference value of the weight, and the process proceeds to the next tension fluctuation detection process.

In the tension fluctuation detecting process of this embodiment, the detection of the tension fluctuation is performed for a predetermined period of time, specifically, based on a predetermined number of tension data.
First, in a variation flag ON determination step, it is determined whether the variation flag of the weight is ON. Then, ON “Y
In the case of "es", the latest data stored in the weight is stored in the variable data storage area and the number of detected data is set to 1
The number is incremented, and the process proceeds to the next data number determination step. If it is not “ON”, the process proceeds to a variation candidate determination step.

In this variation candidate determination step, a candidate event of tension variation is detected as follows. First, the moving average value obtained for the tension, that is, the reference value, and A /
The current tension values collected in the D conversion step are compared, and the difference is equal to or greater than a preset candidate detection value for detecting a candidate event of a tension variation, and in this example, a difference of 5 g or more is defined as a tension variation described later. It is determined that there is a candidate event.

In the case of "Yes" indicating that there is a variation candidate, a variation flag ON step is entered, the variation flag of the weight is turned on, the latest data is stored in the variation data storage area, and the number of detected data is set to one. Then, the process proceeds to the next data number determining step. In the case of "No" with no change candidate,
The process proceeds to a knot / fluff determination step described later.

In the next step of determining the number of detected data, the variation flag is turned on, that is, the number of stored data after the detection of the variation candidate is set to a predetermined number necessary to obtain the entire image of the variation (500 in this example, which corresponds to 5 seconds). ) Has been determined. And "No" in which the number of data is less than 500
In the case of, as in the case where there is no change candidate, the process proceeds to a knot / fluff determination step described later.

In the case of "Yes" in which the number of data has reached 500, the event flag ON step is entered as the collection of detection data of the variation candidate is completed, the event flag is turned ON, and the tension data and the generation of tension data during the detection time are generated. The date, occurrence time, and occurrence weight are stored and saved in the event candidate storage area, and the process proceeds to the next knot / fluff determination step.

In the knot / fuzz discriminating step, the previously collected data of knot passing and fluff occurrence are scanned to check whether the weight has passed the knot and fluff occurrence.
In the case of "s", the process proceeds to an event flag ON step indicating the occurrence of an event to be a management event. In this step, the event flag is turned ON, and the contents of the event, that is, the occurrence of fluff, the passage of the knot, the date and time of occurrence, the time of occurrence, and the weight number are stored and saved in the event candidate storage area, and the process proceeds to the next all weight end determination step. In the case of "No" in the knot / fluff discrimination step, the flow immediately proceeds to the all-spindle end determination step as shown.

In the step of determining the end of the entire weight, it is determined whether or not the end of the entire weight has been reached based on whether or not the weight number has reached the final number. The number is incremented by 1 and the process proceeds to the next spindle. If the weight number is the last number and “Yes” indicating the end of all weights, the process proceeds to the FFT sampling step for collecting the next frequency conversion data.

In this way, in this example, a tension change of a predetermined value or more, which is a control event from 10 milliseconds to a change detection time of 5 seconds, can be accurately detected, and the control event is disconnected as described later. The events can be classified into occurrence, threading, and occurrence of a tension fluctuation exceeding a predetermined value.

As described above, if any control event is detected, including the case where the knot is passed and the fluff is generated in addition to the fluctuation of the tension, the event flag is set as the data of the event of interest in the event flag ON step in FIG. When turned on, necessary data, specifically, the weight number and the content of the event, that is, tension fluctuation, knot presence, fluff presence, and the like are stored in the event candidate storage area.

When all the weights have been completed, the process enters an FFT sampling step for collecting the next data for frequency conversion. In this frequency conversion data collection routine, data collection of tension signals of all weights required for fast Fourier transform (FFT) is performed. First, in the FFT sampling step, the latest data stored in the tension data storage area is sequentially scanned for all weights and stored in the FFT storage area of each weight. In this example, the frequency range and the frequency resolution of the fast Fourier transform can be changed, and the data can be collected by setting the number of sampling data determined from the frequency range and the frequency resolution according to the purpose.

Therefore, in the next FFT sampling completion determination step, completion is determined based on whether or not the number of data collected for each spindle has reached the number of sampling data required for the set fast Fourier transform. The weight that reaches the number of sampling data required for the fast Fourier transform is "Ye
s ", and enters a sampling completion flag ON step, in which the sampling completion flag of the spindle is turned on as completion of sampling of data necessary for fast Fourier transform (FFT). When the determination of all the weights is completed, the background interrupt processing ends. Note that the weight for which the number of data has not reached is “No”, and the completion flag is not turned ON only for data collection.

As described above, in the background, 10
The above processing is repeated every millisecond to collect fluff generation, knot passage, tension fluctuation, and FFT data.

On the other hand, in the foreground, the following management event collection task is constantly repeated while the machine is operating. First, in the in-operation determining step, whether or not the machine is operating is confirmed by a flag or the like linked with the operation switch. If it is not running, no processing is performed.

During operation, the following processing is constantly repeated. First, an event flag ON determination step checks whether the event flag is ON. If “Yes” is ON, the process proceeds to the event determination process, and if “No” is not ON, the process proceeds to a fast Fourier transform process described later.

In the event discrimination process, first, a knot / fluff discrimination step is entered, the relation data of the event candidate storage area is read, and it is checked whether the event content is a knot passing or fluff occurrence. And if either is "Yes",
Proceed to the data storage step, and the event details, specifically knot passing or fluff occurrence, date and time of occurrence,
The generated weight is extracted and stored in the event file set in the storage device.

In the case of "No" where no knot has passed and no fluff has occurred, the event is defined as a tension variation, and the content of the event is determined to be thread breakage, threading is performed, and a tension higher than a predetermined value based on 500 pieces of tension data collected in the background. The following tension fluctuation classification processing of classifying any of the fluctuations is performed. The tension fluctuation classification process of this example uses the moving average value of the tension data. First, in the thread breakage determination step, when the moving average value is continuously lower than the predetermined thread breakage determination value for a predetermined time, it is determined that a thread break has occurred. . Then, in the case of "Yes" indicating that thread breakage has occurred, the process proceeds to the above-described data storage step with the content of the event being broken, and the above-described related data is similarly stored.

On the other hand, when the moving average value exceeds the yarn breakage determination value and the yarn breakage does not occur (No), the process proceeds to the yarn hooking determination step. Judgment is performed. Then, in the case of "Yes" for the thread hooking execution, the content of the event is set to the thread hooking execution, and the process proceeds to the data saving step in the same manner as in the case of the thread break.
Save related data.

If the state does not continue for a predetermined time or more, "N
In the case of "o", the process proceeds to the next data storage step with the event content as the tension variation, and the related data is stored and stored in the event file in the same manner as in the above-described events. Therefore, the date, time, and weight of the event extracted together with the contents of the event (whether the knot has passed, the fluff has occurred, the thread has occurred, the thread has been threaded, or the tension has exceeded a predetermined value) have been extracted in the event file. Will be saved. In the operation example described below,
Good results were obtained by setting the thread break determination value to 20 g and the predetermined time to 3 seconds.

In the event determination step described above, if the event is the occurrence of a yarn breakage, a conventional yarn breaker that cuts and grips the yarn Y with a yarn cutter provided upstream of the existing yarn feed roller 2. (Not shown), a yarn break signal is output to perform the yarn processing. By the way, regarding the detection of the yarn breakage, a commercially available yarn breakage detector may be used in addition to the method based on the tension data, and the processing may be performed in the same manner as with the fluff.

After the above processing is completed, or when the event flag is OFF (“No”), the next fast Fourier transform (FFT) processing is performed. In this FFT processing unit, first,
In the sampling completion determination step, it is confirmed whether or not the sampling of data necessary for the fast Fourier transform (FFT) has been completed based on the sampling completion flag. If the sampling-uncompleted flag is OFF (No), the process returns to the beginning.

In the case of "Yes" of ON, the process proceeds to the FFT execution step, in which the fast Fourier transform (FFT) of the sampling completed weight is executed. The fast Fourier transform (FF
T) used a well-known fast Fourier transform method. A commercially available program or the like can be used for this.

When the FFT is completed, the process proceeds to a characteristic value extracting step, in which characteristic values are extracted by the characteristic value extracting means from the frequency distribution data obtained by the fast Fourier transform, and the characteristic values set in the storage device including the related data. Save to file sequentially. The characteristic value extracting means of the present embodiment integrates a frequency component of a specific frequency region set in advance and stores the integrated value as a characteristic value.

The characteristic values of this example are the first specific frequency range of 0.01 Hz to 0.3 Hz, which has been confirmed to correlate with U% related to the thickness unevenness of the yarn of the original yarn package, and the oil agent adhesion. The U% characteristic value and the OPU characteristic value obtained by integrating the components in each of the second specific frequency regions of 0.6 Hz to 1.4 Hz, which are correlated with the OPU which is an index of the amount. Then, the obtained characteristic values are stored in a characteristic value file including relevant data such as the weight number, the date and time when the characteristic value is extracted, and the like.

Note that specific numerical examples of this example are that the processing speed of the false twisting machine is about 1000 m / min, the stretching ratio is about 1.8 times,
The yarn Y of the yarn package is 12 m at approximately 3000 m / min.
It is a study result under the condition of spun partially oriented yarn (POY) of 5 de. This frequency range may slightly vary depending on the spinning conditions of the original yarn package, depending on the desired characteristics, and is preferably determined by actual measurement. Regarding the above-mentioned U% and OPU, good results were obtained irrespective of slight changes in the processing conditions in the above-mentioned frequency region due to the characteristic values using the above-mentioned integrated values.

When the characteristic value extraction is completed, the process returns to the beginning, and the above process is repeated. Thus, the machine management device 3
8 is a collection of management events such as the time of occurrence of fluff of the original yarn package, the time of passing through the knot of the false twisting machine, the time of occurrence of thread breakage, the time of thread hooking, the time of occurrence of tension fluctuations exceeding a predetermined value, and the characteristics by high speed Fourier transform. Value extraction is performed, and these are stored in an event file and a characteristic value file.

On the other hand, the central management device 39 fetches data from the individual machine management devices 38 at predetermined time intervals, and also performs data recording processing of the weight that has detected the passage of the knot. When a distribution display request command is received from the operator console, the time series distribution status of the management events for each yarn package is displayed. The details will be described below based on the flowchart of FIG.

First, in the raw yarn information processing step, it is checked whether or not there is an input from the bar code reader 40, and if there is, the raw yarn is stored in the yarn storage management area of the raw yarn package based on the input yarn production management information. An original yarn package file including a necessary management item column of the package is created, and the machine number and the weight number of the set false twisting machine are stored in the column along with each item of the yarn production management information.

Next, a display discriminating step is entered to check for a distribution display command from the operator console. In the case of “Yes” with the distribution display command, the process proceeds to the distribution display processing. This processing will be described later.

In the case of "No" with no distribution display command,
Proceed to the next time determination step. This is for determining the time because the data of each machine management device 38 is read every predetermined time, that is, at a predetermined cycle. In this example, the predetermined time is 2 minutes. If "No" has not been reached for the predetermined time, the process returns to the beginning of the process. In the case of “Yes” that has reached the predetermined time, the process enters a data fetching step, in which each machine management device 38 stores all data stored in the event candidate area and the characteristic value storage area in the processing of FIGS. Is taken out and stored in the central management device 39. At this time, the corresponding machine number is also stored according to the number of the machine management device 38.

When the data fetch is completed, the next event detection step is started. In the event detection step, the yarn characteristic fluctuation event is detected as a management event by the event detection means as a management event based on the characteristic value obtained in the characteristic value extraction step of the machine management means described above. That is, in the event detection means of this example, the average value of the past characteristic values during normal operation is compared with the characteristic value obtained in the characteristic value extraction step as the control value, and the difference is determined as the control limit value, specifically, the control value. Is detected as a yarn characteristic change event, and the time of occurrence is stored in a file of the yarn package of the spindle with the characteristic value as a management event.

Next, it is checked whether or not there is a knot passage corresponding to the switching of the yarn package in the data of the management event extracted from the machine management device 38 in the above-described data extraction step. In the case of “No” with no knot passing, the processing returns to the top. In the case of "Yes" with knot passing, the following data recording processing is performed.

In the data recording process, first, in the data recording step, the current knot passage time is stored as the processing end time in the storage file of the current yarn package being processed by the spindle, and the processing of this yarn package is completed. At the same time
The current knot passing time is entered as the processing start time in this file as the storage file of the new yarn package that started feeding after passing through the knot through the knot storage file.

As described above, the data recording process is performed at the time of detecting the passage of the knot. In other words, this data recording process is performed every time the yarn package is replaced, and the processing start time and the processing end time of the relevant yarn package of the relevant weight, each control event, the spinning machine in the yarn manufacturing process, the number of the spindle, the manufacturing Management information such as a lot number is extracted from the stored data for the weight.

These data are stored in the storage device of the central management unit, the file of the original thread package of the weight is fetched by the machine, and the respective items of the management information are stored in the respective management information item columns. Therefore, in the central management device, management information necessary for managing the yarn package is stored in one file for each yarn package.

By the way, the recorded processing start time and the processing end time are the knot passing time by the knot detecting means 32 as described above, and the actual processing start time, that is, the passing time of the twist stop guide 5, the actual processing end time. The time point is different from the time point when the sheet passes through the feed roller 3 or the guide roller 9.
Therefore, in the next data correction step, the processing start time and the processing end time are corrected as follows. That is, the time at which the correction time obtained by dividing the distance to the twisting guide 5 and the feed roller 3 by the processing speed is added to the current recorded data, that is, the time at which the knot passes, is the processing start time,
Rewrite as the end of processing.

At the same time, a control event whose occurrence time is before the corrected processing start time of the yarn package is extracted from the stored file of the yarn package, and the management event is extracted from the file of the yarn package. It is moved to the file of the old yarn package that was processed just before. Whether the control event at the time of switching should be assigned to any of the new and old yarn packages should be accurately determined in consideration of the processing end time for each control event. There is no practical problem with the judgment.

Next, on the basis of the file of the old original yarn package determined by this correction in the data alignment step, for each management event from the processing start time to the processing end time, each data is processed for each management event. Using the start time as a reference time, each occurrence time is sorted in chronological order according to the elapsed time from that time, and stored in the file again. As a result, for each management event, each event is stored in each yarn package file in the order of occurrence with the processing start time as a reference time, specifically, as a zero point, and the distribution display processing described later is simplified. Become.

Thus, the data recording process ends. By this processing, the central management device 39 stores the necessary yarn management information up to the latest yarn package that has been processed in the original yarn package file including the original yarn package file in a predetermined format for each original yarn package. Databases are built sequentially.

By the way, when the distribution display request command is input from the keyboard of the operator console or the like, that is, when the display determination step is "Yes" in FIG. 5, the distribution display processing is as follows.

First, in the range designation step, the range designation means displays a range designation table in a predetermined format on the display means of the central management device 39, specifically, a liquid crystal display or the like. Then, input the lot designation items such as the range of the original yarn package to be displayed, for example, the number of the spinning machine, the range of the spinning spindle, the manufacturing doff number or the time which can specify the same lot, and the completion of the input. To instruct.

Then, the process proceeds to a package extracting step.
A yarn package belonging to the specified range is extracted from the yarn management database formed in the data recording process. Specifically, in the extraction process of this example, the data of the file for each yarn package in the yarn management database obtained in the data recording process is searched, and the yarn management information within the specified range, for example, the designation of the same lot In the case (1), the original yarn package is extracted based on whether or not the designated items of the spinning machine number, the weight range, and the forming time are within the specified range.

When the extraction is completed, the process proceeds to the next distribution calculation step, in which the control event including the fluctuation event of U% and OPU in the present example is performed on all the extracted yarns. The package can be displayed as a time-series occurrence distribution arranged by generation time along a time axis with the processing start time as a reference point, in other words, the origin. In addition, if necessary, the number of all management events targeted for all the yarn packages extracted is added up. It is also displayed as the time series occurrence frequency distribution shown.
The latter display has the advantage that, for example, quality abnormalities of the same lot can be detected with high sensitivity.

Further, in this example, all the control events of all the yarn packages extracted in the same manner as the above-mentioned fluctuation phenomenon of the yarn characteristics for all the control events, specifically, in addition to the above-mentioned fluctuation event of the yarn characteristics, Then, the occurrence of thread breakage and fluff is integrated to obtain the time-series occurrence frequency distribution of all management events. Then, the distribution calculation means causes the obtained time series occurrence distribution to be generated in the lot file and stores it as management data.

Then, in the next display step, the information is displayed on a display means such as a liquid crystal display of the central management device 39. FIG. 6 shows a typical example of this display. In the example of FIG. 6, the time axis of the obtained time series occurrence distribution of the management
In a graph in which each yarn package is arranged in parallel on the vertical axis, the time-series occurrence distribution of each yarn package in a specified range is displayed in parallel on the time axis with the origin at the same position on the screen.

FIG. 6 shows that in the false twisting machine of this example, a raw yarn package made of partially oriented polyester yarn (POY) spun at a spinning speed of about 3000 m / min was processed at a processing speed of about 1000 m / min and a draw ratio of about 1. 8 is a display example of a time-series occurrence distribution of all management events of the yarn package in the designated range when false twisting is performed, and is a typical example in which the management event has occurred.

In the figure, the vertical axis 20 is a typical example of the original yarn package in the applicable period range of the applicable spindle of the designated spinning machine. Although the numbers in the display are indicated by numbers 1 to 9 for mere distinction, actually, the lot number of the raw yarn package and the like are used. The horizontal axis 21 represents the time from the start of processing of each yarn package. The starting point of machining is 00: 0 of the origin 22 on the horizontal axis.
It is set to 0, and the knot passing time point indicated by the gray square mark 23 represents the processing end time point. Therefore, a period between the processing start time 22 and the processing end time 23 represents the processing time of the original yarn package. In addition, at the time of thread breakage, since the execution time of re-threading is known, time during which no processing is performed can be omitted.

The diamonds at 24 in FIG. 11 indicate the point in time at which a tension fluctuation exceeding a predetermined value occurs, the crosses at 26a to 26e indicate the point at which thread breakage occurs, and the triangle at 25 indicates the point at which characteristic value fluctuations occur. . The 27 circle indicates the point at which fluff occurs. Displaying each type of management event in this manner is effective for factor analysis. In FIG. 6, the horizontal axis represents time, but it may be represented by the warp or winding weight of the yarn package. These are also equivalent to time, and the time-series distribution of the present invention includes such a display.

In the drawing, the period from time 22 to time 23 corresponds to the winding of the original yarn package from the complete winding to the start of winding. Contrary to the drawing, the progress from time 23 to time 22 is defined with time 23 as the origin of the time axis. If it is displayed, it will be displayed corresponding to the start and end of winding of the yarn package in the spinning machine, and it is easy to grasp the correspondence between the occurrence of management events and the spinning history, but it is disadvantageous in terms of processing. Become.

As shown in the figure, by displaying the control events of each yarn package as a time-series occurrence distribution on the same time basis, the production history of the yarn package can be effectively compared with that of the yarn package. It is possible to find out which weight has a problem with what was produced at which time, and it is possible to easily narrow down a portion to be investigated for the cause of the defect. Then, investigation and countermeasures can be immediately performed on the weight.

For example, in the yarn package of the supply yarn number 3, the characteristic value fluctuates frequently over almost the entire period 28, and it is estimated that the U% of the yarn package or the OPU abnormality has occurred. . In this example, the inspection result indicates that the abnormality is U%. However, if the U% abnormality and the OPU abnormality are distinguished and displayed, this point can be more easily understood. Then, since the U% abnormality occurs during the entire period of the package, the U condition may vary depending on the production conditions, equipment conditions, and the like when the yarn package is produced.
The direction to pursue the cause of the abnormality, such as investigating the part related to the% abnormality, is also determined.

Further, in the yarn package of the supply yarn number 8, the tension fluctuation of a predetermined value or more frequently occurs during the entire period 29 as shown in the figure. This abnormal tension causes abnormal dyeing of the processed yarn and can be eliminated as a defective product before shipment. On the other hand, by checking the production history of the yarn package in each period, checking the production conditions on the spinning machine side that causes the tension abnormality, the state of occurrence of the abnormality, etc., it is possible to pursue the cause and further take measures against the abnormality. it can.

Further, in FIG. 6, the fluff occurs only twice in the yarn package of supply yarn number 5, which is presumed to be sudden. It is possible to estimate which part is located, and useful information can be obtained in terms of quality control. Further, depending on the occurrence situation, the cause can be pursued. For example, when this occurs continuously, it is presumed that a problem has occurred in a device having a nozzle of the spindle of the spinning machine, particularly a device such as an oil application device and an interlace application device.

As described above, it is possible to classify the cause of the product abnormality related to the inner yarn package, obtain information on the investigation of the cause of the abnormality on the side of the spinning machine, and implement the measures promptly, thereby improving the productivity and the production cost. It has a great effect on downing.

Further, the pursuit of the cause of the yarn breakage is facilitated as follows. For example, when looking at the occurrences of thread breakage 26a to 26e, the cause of the thread breakage occurrence can be known from the time of occurrence in the present method as follows. That is, from the timing of the occurrence, the yarn breaks 26a and 26d are the yarn breaks (transfer breakage) when passing through the knot of the false twisting machine, 26e is the yarn breakage at the end of the original yarn package supply (no knot portion), and Thread breakage occurring at the time of the doff of twisted package (switching mistake)
It is estimated to be.

By omitting these specific yarns from FIG. 6 and displaying them, the occurrence distribution of other yarns becomes more clear. As a result, for example, the specific yarn break position in the winding shape of the original yarn package is known, and in the case where the specific yarn breakage occurs frequently, it is possible to estimate a cause such as a problem in control at the time of winding the original yarn package during spinning, Furthermore, we can pursue the measures. As described above, according to the present invention, even the problem of winding of the supply yarn is found out, and the present invention is also effective in reducing costs by reducing the yarn breakage rate of the false twisting machine.

As described above, in the present embodiment, the processing is performed by the management device including the detection means and the microcomputer, but the processing of the central management device can be performed offline. Further, it is also possible to display the waveform of the tension fluctuation and the waveform of the result of the fast Fourier transform on a graph for further detailed analysis. Further, the present embodiment is not limited to the false twisting machine having the configuration shown in FIG. 1, and can be widely applied to false twisting machines having other configurations for performing false twisting.

[0090]

As described above, according to the present invention, the fluctuation of the characteristic value obtained from the specific frequency component of the untwisting tension at the time of false twisting, which is equal to or more than the control limit value, is regarded as a control event, and the variation is determined for each raw yarn package in accordance with the time of occurrence. It is recorded in chronological order and displayed as a chronological occurrence distribution, and has a great effect on early detection of physical property abnormality of the raw yarn and investigation of its cause.

Further, in addition to the characteristic value fluctuation, untwisting tension fluctuation, yarn breakage generation, fluffing generation, etc., are added to the control event, and a time series generation distribution obtained by integrating these fluctuations is displayed. There is an effect that the factor analysis of the raw yarn, which is an important factor, can be effectively performed from many aspects.

As described above, the present invention provides raw yarn information that enables analysis of the yarn forming factor of the raw yarn, which was difficult to analyze among the causes of defective yarns in the past. It greatly contributes to the improvement of quality and productivity.

[Brief description of the drawings]

FIG. 1 is an explanatory diagram of a weight configuration and a sensor mounting position of a false twisting machine according to an embodiment.

FIG. 2 is a block diagram illustrating a configuration of a management device according to an embodiment;

FIG. 3 is a flowchart of a background data collection task of the management unit of the machine management device of the embodiment.

FIG. 4 is a flowchart of a foreground management event collection task of the management unit.

FIG. 5 is a flowchart of a central management unit of the central management device of the embodiment.

FIG. 6 is an explanatory diagram of a display example according to the embodiment.

[Explanation of symbols]

 1a, 1b Original yarn package 2 Supply roller 3 Feed roller 4 False twist unit 5 Twist stop guide 6, 7 Heating device 8a, 8b Cooling device 9, 10 Guide roller 11 Winding machine 12 Stretched false twisted yarn package 31 Tension detection means 32 Knot passage detection means 33 Fuzz detection means 34 Filter circuit 35 Scanning circuit 36 Analog / digital conversion circuit 37 Interface circuit 38 Machine management device 39 Central management device 40 Barcode reader

Continued on the front page (72) Inventor Bunji Hamasu 77-77 Kitayoshida-cho, Matsuyama-shi, Ehime F-term in Teijin Limited Matsuyama Office (reference) 3F115 CA18 CA21 CA22 CA42 CA53 CB21

Claims (11)

[Claims] In a method of managing a false twisting machine, a characteristic value obtained from a component of a specific frequency region related to the characteristic of the untwisted tension of each yarn package is calculated from the processing start to the processing end for each yarn package. Determined at predetermined time intervals, the case where this characteristic value fluctuates from a preset management value by a predetermined management limit value or more is detected as a management event, and the time of occurrence is sequentially stored, and a specified management range is specified according to a request. And displaying a time-series occurrence distribution of management events for the original yarn package. 2. A management device for a false twisting machine, comprising: a tension detecting means for detecting untwisting tension in false twisting; and a Fourier transform of the detected untwisting tension signal at predetermined time intervals to obtain a frequency domain space. Fourier transforming means for converting into a signal, characteristic value extracting means for obtaining a characteristic value from a signal component in a specific frequency region in which a spatial signal is set, and comparing the obtained characteristic value with a set control value to manage fluctuations thereof Event detecting means for detecting a portion equal to or greater than the limit value as a control event and storing the time of occurrence for each yarn package, and displaying a time-series occurrence distribution of the management event for the yarn package within a specified predetermined management range. A management device for a false twisting machine, comprising a display unit. 3. The Fourier transform means includes: digitizing means for converting the untwisting tension signal into a digital signal; tension storing means for storing the digitized untwisting tension signal for at least a predetermined time interval; 3. The management of the false twisting machine according to claim 2, further comprising fast Fourier transform means for transforming the untwisted tension signal during a predetermined time interval stored at the time interval into a spatial signal in the frequency domain by the fast Fourier transform method. apparatus. 4. The method or apparatus for managing a false twisting machine according to claim 1, wherein a time-series occurrence distribution of management events of the yarn package in a predetermined management range is displayed in parallel on the same time basis. . 5. The yarn package according to claim 1, wherein the yarn package having a predetermined control range is a plurality of yarn packages manufactured in a series of the same manufacturing conditions for a certain period in a yarn forming process for forming the yarn package. Management method or management device for false twisting machine. 6. The provisional yarn package according to claim 1, wherein the yarn package in a predetermined control range is a plurality of yarn packages manufactured under the same manufacturing conditions with the same weight in a yarn manufacturing process for manufacturing the yarn package. Management method or management device for twisting machine. 7. The control event includes at least one of a change in the untwisting tension of a predetermined value or more, a breakage, and a fluff, in addition to the change in the characteristic value. 7. The method for managing any of the false twisting machines according to claim 6. 8. A tension fluctuation detecting means for detecting a fluctuation of the untwisting tension equal to or more than a set value as a control fluctuation, a yarn break detecting means for detecting a broken yarn of the yarn being false-twisted, and a fluff of the original yarn package. 7. The provisional device according to claim 2, wherein at least one of the fluff detecting means for detecting the fuzz is provided, and the event detecting means detects a detection signal from the provided detecting means as a management event. Control device for twisting machine. 9. A yarn yarn switching detecting means for arranging a plurality of yarn packages per weight and detecting the switching of the yarn packages, and using the switching detection signal to determine the processing start time and the processing end time of the original yarn package. The management method or management apparatus for a false twisting machine according to any one of claims 1 to 8, wherein the management method or the management apparatus is automatically determined. 10. The frequency range where the specific frequency range is related to U% of the yarn: 0.01-0.3 Hz or / and the O of the yarn.
The management method or management apparatus for a false twisting machine according to any one of claims 1 to 9, wherein the frequency range related to the PU is 0.6 to 1.4 Hz. 11. The method or apparatus according to claim 1, wherein the characteristic value is an integrated value of a frequency component in a specific frequency region.