JPS5815430B2 - Automatic thread winding machine monitoring method and device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention is equipped with an automatic yarn splicing machine, and the statistical average value of the yarn splicing time of the pipe yarn to be wound in the 7Lh automatic yarn winding machine, and the statistical average of the net winding time of the normal pipe yarn. By calculating the values in advance through observation, and using these statistical quantities as standards, we can individually judge half-weld replacement, warning stoppage, abnormal stoppage of automatic yarn splicing machine, etc. from the observed values obtained through observation. The present invention relates to a new method for monitoring an i-moving yarn winding machine and a device for the same, which enables early and accurate detection of abnormal spindles.

Conventionally, in this type of self-winding machine, various monitoring devices have been adopted to monitor yarn splicing from the viewpoint of machine management and maintenance. Although insufficient information is available for effective management,
Ubui had the problem of requiring a faulty device in order to obtain sufficient information.

In view of such problems, the applicant of the present application filed a patent application in 1983-15.
According to Meikosaki No. 035, the total number of yarn splices, the number of tube yarn exchanges, and the number of uncut yarns due to an abnormal slab catcher are counted, and from these counts, each winding The number of yarn splicings due to normal yarn splicing causes and the number of yarn splicings due to abnormal yarn splicing causes of the weight are detected.

For example, if the number of yarn splices in the former is abnormally high, it is determined that there is a problem with the pipe yarn or the slab catcher, and if the number of yarn splices in the latter is large, it is determined that the take-up spindle is malfunctioning. We are proposing a monitoring method and device for determining that the thread exchange mechanism is malfunctioning when the number of thread exchanges is abnormally high.

By adopting this monitoring method, it has become possible to automatically detect nearly 90% of mechanical failures discovered during table maintenance and periodic inspections.

However, even in the above-mentioned monitoring method, various abnormalities are not judged based on the net winding time of the yarn, and this winding time is determined by the amount of yarn, quality, and weight of the yarn. Since it varies from spindle to spindle depending on the state of preservation, etc.
There are cases where apparent abnormalities are detected, and the cause of the abnormality is not discovered, so in the end, the specific judgment of the abnormality has to depend on the judgment of the maintenance engineer. There was room for improvement.

The present invention has been made based on this viewpoint, and calculates various statistical quantities for each pipe yarn during winding the pipe based on preliminary observations, and calculates various weight distributions obtained in actual monitoring. Analyze the data based on the above statistical quantities, and based on the analysis, calculate the number of half-weld replacements, warning stoppage time and number of times, abnormal stoppage time of automatic thread splicing machine, etc. for each spindle, and calculate these calculated values. The basic purpose of this invention is to provide a method for monitoring an automatic thread winding machine, which is characterized by identifying abnormal spindles through statistical processing of the weight.

The automatic thread winding machine monitoring method according to the present invention consists of a three-step process: preliminary observation, main observation, and statistical judgment by computer.

In the preliminary observation, we measured the winding time and yarn splicing time from the signals of each spindle related to the automatic yarn splicing machine, statistically processed the data distribution obtained from the measurement, and calculated the net winding time of normal pipe yarn. Statistical mean value <tvt> and its upper and lower limit tl
〉±Δ<t w〉 and the statistical average value of one yarn splicing time ku tk〉 are calculated, and the judgment constant K regarding half-weld exchange is calculated.
h. Statistically and empirically set 8 as the determination constant regarding the intermediate stop.

In order to detect abnormal spindles, judgments in this observation are made regarding the presence or absence of half-bead replacement and the presence or absence of dead spindles in the middle of each tube thread.

These determinations are made using the determination constant Kh regarding the half ball exchange and the determination constant Ks regarding the mid-stop spindle calculated in the preliminary observation.

Furthermore, if the dead weight is analyzed by cause and information about the dead weight is obtained for each cause, more specific information can be obtained when detecting an abnormal weight.

The data obtained for each spindle through this observation will be statistically processed by computer.

This statistical processing is performed by detecting the probability of an abnormality that occurs for each weight, and statistically determining whether or not the weight should be considered an abnormal weight based on the probability distribution of the abnormality.

The monitoring device according to the present invention is basically a device that detects information necessary to execute the above monitoring method and executes a program according to the above monitoring method based on the detected information.

The automatic yarn splicing machine detects the detection signal by using the open/close signal of a reed switch that is opened and closed by a magnet fixed to the automatic yarn splicing machine, and by identifying the waveform using a signal processing circuit and computer, the automatic yarn splicing machine can simply It is determined whether the pipe has simply passed through the pipe or whether the pipe has stopped for thread splicing, and if the pipe has stopped for thread splicing, it is further determined whether or not it is due to pipe yarn exchange.

For this reason, it is preferable that the magnet provided in the automatic yarn splicing machine be attached to the tip of a known switch lever.

Further, the thread cutting signal for the slab catcher is obtained by using a known reed relay and opening and closing the reed relay in synchronization with the operation of the cutter.

The signal processing circuit that outputs input information for each spindle to the computer is composed of a yarn splicing signal processing circuit and a yarn cutting signal processing circuit, and is used to determine abnormal stalling of the automatic yarn splicing machine or detect malfunction of the slab catcher. Entered into a computer as information.

Hereinafter, the present invention will be specifically described with reference to the accompanying drawings.

As shown diagrammatically in Fig. 1, each winding weight is 1-1°...
. . , 1-10 is set with 10 spindles as one section, and the automatic yarn splicing machine 2 has a traveling track 3 provided so that each section can sequentially pass through each winding spindle 1 of that section. When a winding spindle that requires yarn splicing is detected, it stops and splics the yarn.

As shown in FIG. 2, each take-up weight 1 actually has a thread 5 that is replaced when each bobbin shutter (4-1..., 4-10 in FIG. 1) is opened. from the drive drum 9 via an unwinding controller 6, a tension regulator 7, and a slab catcher 8 that automatically cuts the yarn when a yarn defect is detected.
The cheese 10 is rolled up and the fully rolled cheese is placed on a cheese tray (not specifically shown).

) and collected. The automatic yarn splicing machine 2 that reciprocates on the running track 3 has a total of four switch levers 2-1, . . . , 2-4 that protrude downward. 2-3 includes a non-magnetic cover (not specifically shown).

), while a duct 12 made of a non-magnetic material and having a rectangular cross section is provided parallel to the running track 3, and within the duct 12 are each winding weight 1-1. . . . , a reed switch 13-1 corresponding one-to-one to 1-10. ......, 1
3-10, each reed switch 13-1.・・・・・・
, 13-10 detects whether the automatic yarn splicing machine 2 is simply passing through or is stopped for yarn splicing.

Switch lever 2-1゜ provided in the automatic yarn splicing machine 2 above...
..., 2-4, the first switch lever 2-1 has a bobbin pocket (not specifically shown).

) is a lever that operates the feeler that detects the presence or absence of the amount of yarn in the pipe.
The switch levers 2-2, 2-3, and 2-4 move up or down to form a thread exchange mechanism (not shown).

) and replace it with a new tube.

The third switch lever 2-3, to which the magnet 11 is attached, is once raised during this pipe exchange, and is lowered again when the pipe thread exchange is completed.

As described above, when the structure is such that the magnet 11 is supported by the third switch lever 2-3 that operates when exchanging the yarn, four types of operations can be performed regarding the automatic yarn splicing machine 2 from the open/close signals of each reed switch 13. can be identified.

That is, when the automatic yarn splicing machine travels back and forth passing through each take-up spindle in order to perform yarn splicing, the open/close signal of each reed switch 13 has a small pulse width of about 100 m5ec, as shown in FIG. The passing pulse shown in Fig. 3 1 has a pulse width of 5 to 5 seconds, or the splicing pulse shown in Fig. 3 has a large pulse width of usually 5 to 5 seconds. This is one of the yarn exchange pulses in Figure 3 (c) that rises again when the yarn splicing is completed and then falls again when the yarn splicing is completed during this period, or when the yarn splicing machine is stopped for a long time longer than the yarn splicing stop time for maintenance etc. The open/close signal of the reed switch does not change for a long time when the yarn splicing machine stops, and is the third cause of an abnormal stop of the yarn splicing machine.By identifying these pulses, it is possible to determine whether the automatic yarn splicing machine 2 is simply passing through or the pipe Stopping force for yarn splicing due to causes other than yarn exchange It is possible to determine whether the stop is due to yarn splicing accompanied by incoming yarn exchange or an abnormal stop of the yarn splicing machine.

In other words, according to this detection method, only one type of simple detection device can be used to detect the passing of a yarn splicing machine, yarn splicing detection, pipe yarn exchange detection, and abnormal stoppage of a yarn splicing machine without installing any other detector. Can be detected.

Note that the detection of passage of the yarn splicing machine is used to determine a warning stoppage, which will be described later, but it is also used for fault diagnosis of the detector, so the reliability of the detection device described above is greatly improved.

In the present invention, in addition to the signals related to the automatic yarn splicing machine 2, the yarn cutting signal of the slab catcher 8 is used as input information.

For this reason, the slab catcher 8 includes a reed relay (not specifically shown).

), and its output signal is sent to the slab catcher 8 controller (not shown).

) is exceeded, the reed relay is activated by a cutter activation signal issued from the amplifier section, and this reed relay activation signal is used as a thread cutting signal.

Next, the signal processing circuit will be explained with reference to FIG.

Figure 4 shows a total of 5 sections S1.1 with 10 spindles. ..., S5 is one unit, and generally there are a total of N units U1. . . . Of the entire processing circuit in which the output of the UN is inputted to the central control unit 21 via the data bus 20, especially the section S1 of the unit U1.
This figure shows the processing circuit for .

The processing circuit of each section above is basically the automatic yarn splicing machine 2.
A total of 10 reed switches 13-1.・・・・・・
A yarn splicing signal processing circuit 30 for processing open/close signals of 13-10, and a total of 10 reed relays 18-1, which are installed in each slab catcher 8. .......

Yarn cutting signal processing circuit 4 that processes the opening/closing signal of 18-10
Consists of 0.

The yarn splicing signal processing circuit 30 includes each reed switch 13-
1. . . . , an encoder 31 to which opening/closing signals 13-10 are individually input and converting these opening/closing signals into binary code signals corresponding to the winding spindle number;
A D-type latch circuit 32 to which a binary code signal of 1 is input, and signals encoded and grouped by the encoder 31 are sent to each reed switch 13-1.・
・・・・・・．

It also has an appropriate time constant (Rk-Ck) to avoid the influence of chatter caused by opening and closing of the binary code 32, and when the binary code signal is stabilized, the binary code is latched to the
In addition to outputting it as a set signal to the central control unit 21
a control circuit 33 that outputs an interrupt signal IRPk to each grouped reed switch 13-1, . . .
, 13-10 input read command signal 110-
The gate 36 opens the gate with Rg and applies the voltage to the central control unit 21 via the data bus 20.

Note that each input terminal of the encoder 31 is supplied with a power supply V.

Pull-up resistors r1゜ are each connected to 0...
, rlo are provided, and each reed switch 13-1°...
..., the opening/closing signal of 13-10 is stabilized.

As is clear from the above configuration, the yarn splicing signal processing circuit 30 transmits information regarding the four types of behavior of the automatic yarn splicing machine 2 to each reed switch 13-1 provided corresponding to each winding spindle.
．． . . . is obtained as an opening/closing signal of 13-10, is converted into a binary code signal corresponding to the weight number by the encoder 31, and is stored in the latch 32 by a set signal from the control circuit 33, and at the same time is assigned. The central control unit 21 is interrupted by an input signal IRP, and the binary code signal is read from the latch 32 to the central control unit 21 via the data bus 20 in response to an input read command signal 110-Rg from the central control unit 21. The weight number is known by entering the number.

It is possible to determine whether the automatic yarn splicing machine 2 is simply passing through, has stopped for yarn splicing, has stopped for pipe yarn exchange, or has stopped for an abnormally long time. , input reading finger from central control unit 21 +11O-R
This is done by opening the gate 36 and reading the open/close state signal of the reed switch into the central control unit 21 via the data bus 20.

As described above, the thread cutting signal processing circuit 40 includes a total of ten reed relays 18-1. . . ., a chatter-free circuit 41 that removes and stabilizes the chatter of each switching signal of 18-10, and a zero circuit to which the switching signal individually stabilized by the chatter-free circuit 41 is applied. - Catcher-type latches 43 and opening/closing signals applied individually to the latches 43 are grouped, and as described above, one or more slab catchers 8
The non-zero detection circuit 44 applies an interrupt IRP to the central control section 21 when the thread cutting signal is outputted by the operation of the non-zero detection circuit 44.

Note that the latch 43 is cleared by the clear command signal CL.
Do it with R.

As is clear from the above configuration, the thread cutting signal processing circuit 4
0 means that when any one or more slab catchers 8 are activated, the non-zero detection circuit 44 generates an interrupt command signal IRP, interrupts the central control unit 21, and the number of slab catchers 8 is activated. data bus 2
0 to the central control unit 21.

In addition, in the central control unit 21, since the actuation of the slab catcher 8 always accompanies yarn splicing, the corresponding yarn splicing signal is processed as a yarn splicing signal associated with uncutting by the slab catcher 8.

Next, a method of processing various signals obtained as described above will be explained.

σ) Calculation of statistical quantities 6 through preliminary observation In the preliminary observation, statistical quantities necessary for detecting abnormal weights are calculated based on the operation of the automatic thread winding machine.

Now, regarding one winding weight, the operating state of the winding weight each time the tube yarn is replaced is illustrated as a model in FIG.

The figure shows the yarn splicing time t when changing the tube yarn from the 0 point to the left.
BO, and yarn splicing due to activation of the yarn splicing unloading slab catcher during winding. Yarn splicing due to machine trouble.
If there is, add the yarn splicing time required for that yarn splicing, and on the other hand, on the right side based on the 0 point, the time required for winding the pipe yarn t and the warning stop time t0 or full winding stop time td. etc., if any, are added and shown.

As shown in the figure, the observation of the tube B1 started midway through the winding, and the observation of the tube B was stopped midway through the winding, as shown in the figure.

Each of the above-mentioned pipe threads B1.・・・・・・B6tB of Bn
7°B22 represents a so-called half-ball exchanged yarn in which the yarn is not completely wound and is automatically replaced and discharged leaving a considerable yarn length, which is one of the causes of the abnormality.

Also, the tube yarn B4 t B9 t B14 t B20
tB23 tBn-2 shows that the time required for pipe thread exchange was long as a whole, and these abnormalities were due to warning stoppage, full winding stoppage, abnormal stop of automatic thread splicing machine 2, etc. It is.

Note that the warning stop is a device that lights up a red lamp and automatically stops winding when yarn splicing is performed a predetermined number of times in succession in the forward or backward process of the automatic yarn splicing machine.

The main cause of the warning stoppage is a so-called knot error, where the thread is not knotted from the beginning, or the thread is knotted but is tied poorly and the knot monitor attached to the automatic splicing machine detects it and cuts it. It has become.

A full roll stopper is one that automatically stops winding when a predetermined amount of cheese is reached, and remains in place until the table holder processes it.

Further, an abnormal stop of the automatic yarn splicing machine means that the automatic yarn splicing machine itself stops due to a failure or maintenance.

As is clear from FIG. 5, the pipe yarns B4 t B6 t B7 and B9 t B14yB belonging to the above-mentioned abnormal winding
17°B20 p B221 B231 BH-2 and the first yarn B1 The time twi required for winding the yarn except for the last yarn Bn is the statistical average value <
It is distributed in the vicinity of tw >.

Therefore, in a preliminary observation, the population M(1,,i.

N) and statistically process the population M to obtain a statistical average value of the net winding time of normal tube threads <tw>±Δ<1,
, > and the statistical average value tk of the yarn splicing time iki required for one yarn splicing.

The number N of data for the population M is set to such an extent that the distribution of the winding time of the pipe yarn for only one splicing and the distribution of the winding time of the pipe yarn for two splicings both form a normal distribution. Statistical processing is performed when the number reaches the set number N.

The preliminary observation is carried out by using various signals inputted by the signal processing circuit shown in FIG. conduct.

Now, a method of preliminary observation regarding the exchange of the pipe thread for the i-th winding weight Si of a certain section in an arbitrary unit will be explained using the flowchart shown in FIG. 6. First, initialization is performed and observation is started.

Next, it is determined whether the yarn is spliced or not.

Now, in the yarn splicing signal processing circuit 30 of FIG. 4, when the interrupt signal IRPk is applied from the control circuit 33 to the central control section 21, the take-up spindle number related to the automatic yarn splicing machine 2 is read from the latch 32.

Then, the state of the open/close signal of the closed switch from the gate 36 is determined to determine whether or not to splice the yarn.

That is, if it is a simple passing signal, the answer is "NO", and if it is either a yarn splicing signal or a pipe exchange signal, it is determined that the yarn splicing is present, and is determined as "YES".

In the case of “YES”, the number of yarn connections r of the pipe yarn Bj of the weight S.
Add 1” to Jk.

Next, it is determined whether the above-mentioned yarn splicing is a yarn splicing that does not involve exchanging the tube or yarn, or a yarn splicing that involves exchanging the tube or yarn.

This determination is output from the gate 36.

Distinguish by the form of the waveform. If it is determined to be a yarn exchange signal, it indicates that the winding of the yarn Bj has been completed, so the measurement of the winding time of the yarn Bj up to that point is stopped and the winding time twj at that time is memorized. , the next pipe Bj-h
Start measuring the winding time twj+1.

At the same time, it is determined whether the cumulative exchange number N of pipe threads in the current target section has reached the preset number N, and if "NO", the preliminary observation is continued as is.
If the answer is ``Continue'', the preliminary observation ends.

Number of splices nk of each pipe yarn B obtained from the above preliminary observation
j and the winding time tWJ per yarn (j=1...
A population M consisting of a total of N pieces of data, where .

Those in which the number of yarn splices per one pipe yarn is nk. twl...
..., twn).

For this child group C, as shown in Fig. 7, if the horizontal axis is the winding time ``w'' and the vertical axis is the number of yarns wound during that winding time, a histogram will generally show a normal distribution. .

By statistically processing this normal distribution A, the statistical average value of the winding time for one yarn splicing by yarn exchange < t w >
and its dispersion Δtw〉 are calculated.

In addition, the statistical average value of the yarn splicing time required for one yarn splicing t
k〉 is nk・=2, that is, yarn splicing 1 due to pipe yarn exchange
For the data when there is one yarn splicing in the middle and one yarn splicing in the middle, in the same way as above, convert the data into histograms to find the distribution curve opening, and use the distribution function to calculate the statistical value of the winding time in the case of nk・-2. The average value <t//w> is determined and calculated as the difference from the statistical average value <t'□> of the hoisting time when nk·=1.

That is, ぢ7tk〉=〈t//w〉−〈t~〉.

Here, the yarn splicing time associated with pipe yarn exchange has a statistical distribution and can be treated as the same as one statistical yarn splicing time tk〉, so it is the statistical average value of the net winding time of one pipe yarn. is determined as <tw>=<tw'>−<tk>.

At the same time, judging from the above distribution state, statistical variation (
Judgment constants Kh and Ks are appropriately selected to judge whether the thread is abnormally replaced or exceeds the value below ±Δ<t w> ).

The determination constant Kh is used to determine when the winding time for the tube yarn is short, and as shown in FIG. 5, the tube yarns B6 and B7 are exchanged with half balls and the time until the tube yarn is replaced is short.

For B22, these are judged as half-ball exchanges,
It is appropriately set empirically within a range where Kh<tw><<tw>−Δ<tw> holds true.

In addition, the determination constant 5 is for determining the case where the time until the pipe thread is replaced is long, in other words, there is a stopper on the way, and as with the determination constant Kh above, it is 8 tw〉〉〉〉 tw〉
+Δ×tw〉 is appropriately set empirically.

Then, using Ks<tw> as a determination condition, a weight with a longer hoisting time than this is determined to be an abnormal weight.

There are various causes of the abnormality in this case, as described above, and these are determined by the following program.

(■) Cause analysis based on the main observation In the main observation, the statistical mean values 〈tw〉±∆〈tw〉, 〈tk〉 and the judgment constant Kh obtained from the preliminary observation are used.
, Ks is basically used to determine the presence or absence of a half-weld exchange and a stop spindle, and if there is a stop spindle halfway, a warning stop spindle, a full bobbin spindle, and a stop of the automatic thread splicing machine are used. Determine specifically the cause of the outage.

The specific information obtained for each unit, each section, and each spindle within the monitoring time TM through this main observation is as follows.

Information; (a) Total stop time Tsk of automatic yarn splicing machine and its number N
5k (b) Total time Trb, Tru for each cause of warning stop.

Trm, Trx, and each number of times NrbtNru,
Nrm, Nrx and warning stoppage rate R, (=Ncx/Tne
t) (c) Total number of half-weld exchanges Nh, total number of tube yarn exchanges Nbc
and the half-beam exchange rate Rh (=Nh/N5o)), the number of operations Nu of the slab catcher, and the thread cutting rate Ru (Nu /
T net) (E) Total number of thread splicing "kc," number of machine troubles during thread splicing "Nmt" and machine trouble rate Rmt (=Rmt/Nko); and) Total dead time T8, net wire winding time Tnet (
TM-T,, Nko<tk>), operating rate E (Tne
(t/TM) (G) Number of pipe threads without abnormal stoppages other than thread splicing Nn (excluding half-weld replacements) (H) Number of full stoppages Nd and time Td Processing is performed according to the flowchart shown in the figure.

(1) Initialization Values <tk> and <tv> obtained through preliminary observation.

Set Kh, and other initial constants.

Then, various counters and flags are reset and monitoring begins.

(ii) Determining the status of the automatic yarn splicing machine Now, in a specific section of a specific unit, when IRPk is reached when winding the j-th pipe yarn, the winding spindle number Si is read and a status signal regarding the automatic yarn splicing machine is sent. It is determined by the signal from the gate 36 shown in FIG. 4 whether the machine is simply passing through, stopping for splicing, or abnormally stopping the splicing machine.

It is determined whether the output of the gate 36 is simply passing through the yarn splicing machine with a pulse width of about 100 m5ec or whether it continues for longer than the time required for yarn splicing (usually 5 to 5 seconds), and the output is stopped at that time. If so, it is determined that the automatic yarn splicing machine is being spliced, and if the output does not continue to change any longer, it is determined that the automatic yarn splicing machine has stopped due to some other cause.

There is an easy way to determine this because the time t from the previous interruption to the current interruption is abnormally longer than the yarn splicing time, and this method is used to determine if the yarn splicing machine has stopped abnormally. I'll decide.

0i1) In case of abnormal stop of yarn splicing machine The stop time tsk of the yarn splicing machine is equal to the interruption interval t corresponds to

Here, the cumulative stop time Tsk and the number of times Nsk thereof are determined.

(Tsk = Tsk + 1 to +Nsk, N) and sets a flag to stop the yarn splicing machine (Flagk←1) 6V) The yarn splicing machine is just a passing signal, and the automatic yarn splicing machine 2 is the i-th spindle. If the yarn simply passes through Si, the yarn splicing will not be continuous, so even if the yarn splicing has been continuous until then, the number of consecutive yarn splicings by cause n.

B. Clear ncutncmtncx and reset the flag (p/agk←0).

(■ Processing in the case of yarn splicing First, for the j-th pipe yarn B, calculate the number of yarn splices nbk
Add 1” to (j) (nbk(j) ←rlbk(j
)+1).

Furthermore, 1” is added to the cumulative number of yarn splices Nkc (Nko4-
Nko+1).

Next, determine what was the cause of this thread splicing.

That is, in cases other than the cause of pipe yarn exchange, the presence or absence of operation of the slab catcher 8 is determined by the presence or absence of a yarn cut signal, and if the slab catcher 8 is not activated, it is determined that it is a machine trouble.

Depending on whether the cause of yarn splicing is due to pipe yarn replacement, uncut by the slab catcher, or machine trouble, the cumulative number of yarn splices by cause Nbo, Nu, Nm
Add "1" to any one of t.

Next, the number of continuous yarn splices n.

Add 1” to X.

Here, continuous yarn splicing means that yarn splicing is performed every traverse of the automatic yarn splicing machine 2, and the causes thereof include pipe yarn exchange (ncb) and operation of the slab catcher 8 (n
ou), due to machine trouble (nom),
Number of continuous yarn splices n for each of these causes.

1 depending on the cause of the
” will be added.

Number of continuous yarn splicings for each of the above causes n.

b, nCu, nCm. For ncx, set a preset constant, for example "3", and set "cb, ncu".
Either of jnctlncx - is set value "3"
When this is reached, the warning stop requirements are met and a flag is raised.

That is, if Fla-gr<-1 is set and the subsequent process involves a stoppage on the way, it is determined as a warning stoppage.

(vb) Determination of whether or not to replace half-balls Whether winding of the j-th yarn has been completed is determined based on the presence or absence of a yarn exchange signal.

When the end of winding is confirmed, the counting of the winding time tj for the pipe yarn Bj is stopped, and at the same time, the counting of the winding time tj for the next pipe yarn Bj is stopped.
Start counting the hoisting time tj+1 of j+1.

Regarding the j-th winding time tj, in the next process, it is first determined whether a half ball is to be replaced.

This determination of half-ball replacement is made by comparing the winding time of the j-th yarn with the lower limit of the net winding time of a normal yarn obtained through preliminary observation.

That is, tj tk<<t>-Δku> -0-1-■ However, tk is the total time required for splicing the j-th pipe yarn, and if there are nbk splices, then ``bk yarn splicing times (tkl, tk2......

. . . , tkn), but in reality, using the statistical average value of time <tk>, tk = nbk x t
k> and determined using the following formula.

t゛-nbkkutk〉Kh・く涙〉・・・・・・00
If it is determined by the formula that half-ball exchange is "present,"1'' is added to the cumulative number of half-ball exchanges Nh.

If it is not a half-ball exchange, then it is determined whether or not there is a stopper on the way.

□ In addition, in the observation example of 40 count cotton yarn, ∆kutw〉≦(0,1~0.15)〉, and the cases where the number of yarn splices nbk is 2 or more are statistically less than 20%, In cases where nbk is 3 or more, the distribution is 10 or less, so Kh is set to 3/4.

In this case, if there is little thread remaining, it will not be determined as a half-weld, but as explained later, the data will be statistically processed and those with a high frequency will be determined to have an "abnormality" in the thread exchange mechanism, so it is not practical. There are no problems above.

(Determination of viD mid-stop spindle If half ball exchange is “no”, then presence or absence of mid-stop spindle is determined.

In this case, the determination condition is based on a comparison of the yarn winding time with the upper limit of the statistical average value of the net winding time of normal yarns.

t J t k><t>+Δ<tw>...
. . .■, and if considered in the same manner as the above equation 0, using the determination constant KS, -1·-nbk tk>>K8<tw> . . .■ can be obtained.

When the weight is stopped midway, when the weight is at full capacity or when the weight is stopped at warning, it takes a considerable amount of time for the platform holder to take recovery measures (approximately 5 minutes including the patrol time of the platform holder).
minute) is required, so there is no problem even if the above-mentioned judgment constant KS is made a little large.

Note that, like Kh, this determination constant Ks can be determined empirically and statistically depending on the management status of the pipe yarn, and for 40 count cotton yarn, Ks may be set to 1.3.

If there is a stopping spindle on the way, the stopping time t of the j-th pipe is calculated using the following formula.

t,=i j −nbk<t k>−<t,>
-”(E) The deadweight time t5 for the pipe system B calculated using the above formula (■) is added to the cumulative deadweight time T5 (i.e.,
T5←Ts+ts).

If there is no stopping spindle on the way, wl'' is added to the number Nn of pipe yarns without stopping spindles other than yarn splicing (Nn←Nn+1).

(vii) For pipe threads that have a dead weight in the middle as determined by the counting formula ■ by cause of dead weight in the middle,
The cause of the stoppage in the middle is a warning stoppage, a full winding stoppage, and a stoppage due to the stoppage of the automatic yarn splicing machine.

G) Determination of warning stop The warning stop is determined by the condition of ``Yarn splicing is present'' and the number of continuous yarn splices ``Yes'' for each traverse of the automatic yarn splicing machine (Flagr = 1). If the condition for the intermediate stop is satisfied according to the above formula 0, the warning stop is present.
Then, 1" is added to the cumulative number of warning stoppages NrX, the number of times due to pipe thread replacement Nrb, uncut Nru due to the slab catcher, and machine trouble Nrm due to other causes, as applicable.

Then, reset plugr. P) Distinguishing the stop spindle due to the stoppage of the automatic yarn splicing machine After the stoppage of the automatic yarn splicing machine is confirmed, that is, when Flagk = 1 and the spindle is determined in the middle, it is determined as the stop spindle due to the stoppage of the yarn splicing machine. It is determined.

(c) Determination of full winding stop The full winding dead weight corresponds to the above-mentioned warning dead weight and does not correspond to the winding dead weight due to the stoppage of the yarn splicing machine, that is, Fla
When gk=O, it is determined that there is a full-volume stoppage, and 1'' is added to the cumulative number of full-volume stoppages Nd.

(Nd4-Nd+1). In the above process, the determination of the presence or absence of the spindle during half-ball exchange for the pipe yarn Bj is completed, and in the post-processing process, in order to determine the state of the next supply pipe yarn B.・After initializing the number of yarn splicing nbk, warning stop flash agr, stop flag Flagk of the yarn splicing machine, etc. and determining whether or not the monitoring time TM has been reached, if the hinitaring time TM has not been reached, the monitoring Continue.

By executing the above process, it is possible to programmatically distinguish between normal winding and abnormal winding that may be due to some cause.In addition, when a warning stop or gt picture scroll stop is encountered for the same pipe thread, the stop time is determined. If it is long and is exchanged as a half ball, the above program will determine that it is a warning stop or a full stop, but the probability that both will occur at the same time is the half ball exchange rate (Nh / Nbc) 5-10%,
The incidence of a warning stop or a full stop is extremely small, about 10%, at most 1%, so in practice it can be ignored.

In addition, in the case of a bobbin thread breakage, the thread is automatically replaced and the thread is replaced.

If necessary, this can be separated by installing a known bobbin thread trim detector.

By the way, when an abnormality is actually predicted and you want to know in a short time which winding plate weight has an abnormality, set the monitoring time TM to the extent that data that is normally distributed can be obtained. For this purpose, a practically accurate determination can be made by monitoring for about 2 to 3 hours.

Of course, it goes without saying that continuous monitoring for a longer period of time may be carried out if necessary.

(III) The data obtained through the execution of processes beyond statistical processing of data are printed out together with necessary information such as the year and month of monitoring, the pipe number, unit number, section number, key number, etc.

The data obtained in this way is classified by weight (by unit,
Section by section) undergoes statistical processing by computer,
Finally, it is determined whether the weight is abnormal or not.

That is, the amount of data obtained as described above is expressed as the half ball replacement rate Rh (Nh/Nbc), the warning stop rate Rr (Nc x/T
net ), yarn breakage rate Ru (Nu/'rne
t), machine trouble rate Rmt (mt/Nkc), etc., and then the presence or absence of abnormality is statistically determined for each weight.

For weights that are determined to have an abnormality, the yarn exchange mechanism, automatic yarn heeling machine, slab catcher, etc. are each diagnosed and maintenance is performed at an early stage.

As explained in detail above, the present invention detects the winding time of a pipe yarn and the number of splices of a pipe yarn as information for each spindle, and calculates the net winding time and splicing time of a normal pipe yarn through preliminary observation. The statistical average value tw〉.

Using < t k >, the presence or absence of half-bead exchange and the presence or absence of a dead weight on the way (abnormal dead weight) is determined for each individual pipe yarn based on the above spindle-specific information, and the weights with a high probability of half-beam exchange or dead weight on the way are identified. The present invention provides a method for monitoring an automatic thread winding machine based on determining that the spindle is abnormal, and a monitoring device for carrying out this monitoring method.

According to the monitoring method according to the present invention, it is possible to accurately detect abnormal weights by cause in the main observation using accurate and practical judgment conditions obtained from the preliminary observation, so there is no need for a detector for each cause. There are advantages to not doing so.

Furthermore, the monitoring device according to the present invention simply uses two types of detectors, and can be classified and analyzed into a large amount of useful information through computer program processing, which greatly simplifies the processing circuit for detected information. This has the advantage that information processing costs are extremely low.

[Brief explanation of drawings]

Fig. 1 is an explanatory diagram schematically showing the configuration of one section of an automatic thread winding machine, Fig. 2 is a schematic perspective view showing the structure at each winding time,
Figure 3 A90, C, and 2 are waveform diagrams showing detection output signals for the automatic yarn splicing machine, Figure 4 is a configuration explanatory diagram of the monitoring device according to the present invention, and Figure 5 shows the winding state of the pipe yarn. FIG. 6 is a flowchart of the preliminary observation, FIG. 1 is a histogram of data obtained by the preliminary observation, and FIGS. 8(1) and 8(■) are flowcharts of the main observation. 1... Take-up spindle, 2...1 moving yarn splicing machine (2-L...2-4... switch lever), 5... tube yarn, 8... slab catcher, 10... cheese, 11...Magnet,
13...Reed switch, 18...Reed relay, 30
... Yarn splicing signal processing circuit, 31 ... Encoder, 32 ...
latch circuit, 33... control circuit, 36... gate circuit,
40... Thread cutting signal processing circuit, 41... Chatter free circuit, 43... Latch circuit 11.44... Non-zero detection circuit.

Claims (1)

[Claims] 1. An abnormal weight is determined by performing preliminary observation (a), main observation (a), (b), and integral determination (C) as follows: 1. A method for monitoring an automatic yarn winding machine equipped with an automatic yarn splicing machine and an automatic yarn winter changing mechanism; (a); Calculate the statistical values of the following quantities (■, ■, ■) through preliminary observations. ■Statistical average value of yarn splicing time tk ■Statistical average value of half-weld determination winding time Kh-tk (however, K
h: Half-replacement judgment constant obtained statistically, empirically, and empirically) ■5.tw for the average value of the abnormal stop weight winding time. (However, 8: Abnormality/stopped spindle judgment constant obtained statistically and empirically) (b); The following judgment is made regarding the number of splices nj and winding time tj of the pipe yarn B obtained by observing each spindle. (b) -1°(b)
-Do step 2; (b)-1... Judgment condition (i-ni・tkk
When h-1w) is satisfied, it is determined that there is a half-ball exchange (b)-2-... Judgment condition (tj-nj-1k>K
s-tw) is established, it is determined that the abnormal stoppage is present.Next, by repeating these judgments within the set observation time TM, the number of half-ball replacements Nh, the number of abnormal stoppages Ns, and In addition to measuring the cumulative deadlock time Ts, the cumulative value Nu of the number of slab catcher operations, the cumulative value Nmt of the number of machine troubles, and the cumulative value Nox of the number of warning deadlocks are measured. (C); The above quantities Nh, Ns, T, , N obtained for each weight
u. After normalizing Nmt and Nox, the presence or absence of an abnormality is determined for each weight based on the statistical distribution of these quantities. 2 Using the cumulative number of half ball exchanges and tube yarn exchanges Nh and Nbo counted for each spindle in this observation (b), Rh = Nh/
2. A method for monitoring an automatic thread winding machine according to claim 1, wherein after normalizing as Nbo, statistical processing is performed to detect abnormal spindles due to half-weld replacement. 3 Cumulative value N of the number of times the slab catcher operates and cumulative value Tn of the net hoisting time counted for each weight in this observation (b)
2. The automatic yarn winding machine monitoring method according to claim 1, wherein abnormal spindle weight due to malfunction of the slab catcher is detected by normalizing it as 9=Nu/Tnet using Et and then performing statistical processing. 4 After normalizing as Rmt-Nmt/Nkc using the cumulative number of machine troubles Nmt counted for each spindle in this observation (b) and the cumulative value Nkc of the number of yarn splices, statistical processing is performed to identify abnormal spindles mainly due to misknots. A method for monitoring an automatic thread winding machine according to claim 1, which detects. 5 Cumulative value NCx of the number of warning stoppages calculated for each spindle in this observation (b) and cumulative value T of net and top times. The method for monitoring an automatic yarn winding machine according to claim 1, wherein the abnormal weight is determined by statistical processing after normalization is performed as Rr=Nox/Tne, t using et. 6. In a monitoring device for an automatic yarn winding machine that is equipped with an automatic yarn splicing machine that runs repeatedly and has a slab catcher and a yarn exchange mechanism for each spindle, a magnet is fixed to the winding lever that operates when exchanging the yarn of the automatic yarn splicing machine. However, while a reed switch is provided for each spindle to be opened and closed by the magnet, if the magnet moves along with the automatic yarn splicing machine, it can be determined whether it is simply passing through the automatic yarn splicing machine by identifying the waveform of the output detection signal pulse. , a yarn splicing signal processing circuit that determines whether the stopping force for yarn splicing is a stop to replace the incoming yarn or an abnormal stop of the automatic yarn splicing machine, and a yarn cutting signal processing circuit that detects uncut from the operation of the slab catcher. A circuit is installed, and the output from each signal processing circuit is used as input information to calculate the statistical average value tk of yarn splicing time and the statistical average value of half-ball judgment hoisting time, which have been set in advance in preliminary observations. A determination method for determining the presence or absence of half-ball replacement and the presence or absence of abnormal suspension based on the value Kn・t and the statistical average value K・tw of the abnormal stoppage hoisting time, and a single stage of the determination determines “Yes”. The number of half-ball exchanges Nh for the set determined as
The number of abnormal spindle stops N5 and the cumulative spindle time T8 are measured values Nu
means for detecting the cumulative number of machine troubles Nmt; means for counting the cumulative value Ncx of the number of warning stoppages; and the above-mentioned quantities Nh, NS, TS, N obtained for each weight.
A monitoring device for an automatic winding machine, characterized in that it includes means for normalizing u, Nmt, and Nox and then reprinting the presence or absence of an abnormality on a weight 1 based on the statistical distribution of these quantities. 7. A monitoring device for an automatic thread winding machine according to claim 6, characterized in that the reed switch <1 is disposed in a cylinder made of a non-magnetic material at a position facing the magnet for each spindle. . 8 The inherited signal processing circuit includes an encoder that converts detection signal pulses for each spindle into a binary code signal corresponding to the spindle number, and a circuit that groups the output signals and simultaneously interrupts the impuputer. Of the signals that
7. A monitoring device for an automatic yarn winding machine according to claim 6, characterized in that it comprises a gate for outputting a status signal of the automatic yarn splicing machine. 9 The yarn cutting signal processing circuit has a reed switch that is opened and closed by the operation of the slab catcher, and a chatter-free circuit that stabilizes the opening and closing signal of the reed switch,
7. A monitoring device for an automatic yarn winding machine according to claim 6, wherein the output signals are grouped and then set in a latch and an interrupt is issued to a computer.