JP2010101905A - Measurement system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To determine whether variations in the measured values by a combination balance and a weight sorter are abnormal. <P>SOLUTION: Each time a combination product is discharged from the combination balance 1, the weight sorter 4 performs remeasurement. A CPU 9 calculates a subtracted value, namely, the difference between the total measured weight of the discharged combination product and the remeasured value by the weight sorter 4 of the discharged combination product. The CPU 9 calculates an operation value, namely, the total value or the average value of the subtracted values where an article measured by one meter corresponds to a nonparticipation combination article by one meter for each different combination article, thus calculating the standard deviation of the total value or average value. When at least one of all the standard deviations is smaller than a predetermined boundary value, the CPU 9 determines that the variations of values measured by the weight sorter 4 are normal. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

The present invention comprises a combination weigher and a reweighing means, such as a weight sorter, for reweighing the combination articles weighed and discharged by the combination weigher, for example, a weight sorter, and re-runs based on the respective weighed values. The present invention relates to a weighing system capable of diagnosing the weighing state of a weighing means, warning if there is a failure, and returning to a normal state if it is abnormal.

Generally, a zero point abnormality or span fluctuation may occur during use of a measuring instrument. The combination weigher variously combines each weighing signal from a plurality of weighing instruments, selects a combination whose total value is acceptable from the combination of these weighing signals, and constitutes the selected combination. The product is discharged from the measuring instrument. Therefore, if a zero point abnormality or span fluctuation occurs in the weighing instrument of the combination weigher, the total weight (combination weighing value) of the combination articles discharged from the combination weigher may deviate from an allowable value.
Conventionally, a combination article discharged from a combination weigher is re-weighed by a weight sorter to check whether the combination weighing value is an acceptable value, and among a plurality of measuring instruments provided in the combination weigher. A method for automatically detecting a defective measuring instrument has been proposed (Japanese Patent Laid-Open No. 62-284211).

This method generates a combination weighing value of the discharged combination article every time the combination article is discharged from the combination scale, and the reweighing value every time the discharged combination article is re-weighed by the weight sorter. A combination weigher that weighs the discharged combination article each time it is determined that the combination weighed value and the reweighed value match within a predetermined allowable range. The count value of the counter corresponding to the weighing instrument is increased, and the count value indicating the maximum value is detected as a defective weighing instrument.

Next, determination of whether or not the combination weighing value and the re-weighing value match will be described. FIG. 1A shows the allowable weight range of the combination weighing value, where WT is the combination target weight and the lower limit weight, and WU is the upper limit weight. When the combination weighing value and the reweighing value are within the range from the target weight WT to the upper limit weight WU, it is determined as an appropriate amount, and when it is outside the allowable weight range, it is determined as an inappropriate amount. Here, when both the combination weighing value and the re-weighing value are within the allowable weight range, it is determined that the both match, and either one is within the allowable weight range and the other is outside the allowable weight range. It is determined that there is a mismatch at some point.
And each time the judgment is made sequentially and the judgment that there is a discrepancy is made several times, the count value of the counter corresponding to the defective measuring instrument increases, and the count value corresponds to the normal measuring instrument. In this state, the counter value becomes at least one larger than the count value of the counter. Therefore, it is possible to detect a counter having the maximum value among the count values of a plurality of counters corresponding to each measuring instrument, and determine that the measuring instrument corresponding to the detected counter is abnormal.

FIG. 1 (a) is a diagram showing variations when the true weight value of the combination article is Was1, and the combination weigher and the weight sorter generate the combination weighing value and the reweighing value in a normal state, respectively. is there. If the standard deviation of the combination metric value is σr, the combination metric value varies between Was1 + 3σr and Was1−3σr (p to q) with a probability of 99.7%. If the standard deviation of the reweighing value is σx, the reweighing value varies between Was1 + 3σx and Was1-3σx (r to s) with a probability of 99.7%. Thus, as shown in FIG. 1 (a), when the intervals between q and s and WU are relatively wide and the intervals between p and r and WT are relatively wide, the combination weighing value and the reweighing value are Even if there is variation, it is possible to accurately determine that the combination weigher and the weight sorter that are normal are normal.

However, in FIG. 1 (b), when the true weight value of the combination article is very close to the target weight Wt as in Was2, even if the combination weighing value and the reweighing value are very slightly different, In some cases, one of the measured values is determined to be an appropriate amount larger than Was2, and one of the measured values is determined to be an appropriate amount smaller than Was2. In this case, the determination results of the two do not match. Such a phenomenon occurs with a high probability.
That is, there is a problem that discrepancies occur frequently in the determination results even though the scales of both the combination weigher and the weight sorter are normal. And when it is close to Wu, such as Was3, and when the difference between Wt and Wu is set small, the probability that the same phenomenon will occur increases. Therefore, in the conventional method, even if the weighing instrument (weighing hopper) of the combination weigher is not defective, it becomes easy to determine the defect, which hinders the correct quality determination.

Furthermore, if the set range of the upper limit weight value Wu with respect to the target weight value Wt is increased too much, as shown in FIG. 1 (c), the true weight value is Was4 and one of the measuring instruments of the combination weigher is Was4-we. Even when there is a relatively large bias error we such that> WT, there is a problem that the determination results of both scales always coincide and no alarm is issued.
Therefore, if the set interval of the upper limit weight value WU with respect to the target weight value WT is increased too much, an abnormality can be detected only when the combination weigher or the weight sorter has a large weighing error, which is also inconvenient.
Therefore, in the conventional method, there is a problem that the abnormality or normality of each measuring instrument of the combination weigher and the weight sorter cannot be accurately determined due to the presence of an error due to variations in the combination weighing value and the reweighing value.

An object of the present invention is to provide a weighing system that can accurately determine abnormality and normality of variation in measurement values of a combination weigher and a reweighing unit.

  A weighing system according to an aspect of the present invention includes a plurality of weighing instruments, and various combinations of the weights of articles weighed by these weighing instruments, and the total weight of these combinations falling within an allowable weight range. A combination weigher that discharges a combination article, a reweighing means for reweighing each time the combination is discharged from the combination scale, a total weight of the discharged combination article, and the discharged A calculating means for calculating a subtraction value which is a difference from a reweighing value in the reweighing means of the combination article, and a total value of the subtraction values corresponding to the combination articles in which the article weighed by one measuring instrument does not participate, or An arithmetic value, which is an average value, is calculated for each combination article in which the one measuring instrument is different, and at least one of all the standard deviations is calculated by calculating means for calculating the total value or the standard deviation of the average value. Forecast Is smaller than prescribed boundary value, and a first determination means for determining the variation of the weight value is normal for the re-metering means, is intended to comprise.

  The first determination means may determine that the variation of all the measuring devices or the re-weighing means is abnormal when all the standard deviations are equal to or greater than the boundary value.

  When the first determination means determines that the variation of the re-weighing means is normal, when one of the standard deviations is smaller than the boundary value, and the other standard deviation is equal to or greater than the boundary value The weighing system may include second determination means for determining that the measuring instrument that does not participate in the creation of the standard deviation smaller than the boundary value is abnormal in variation.

  A weighing system according to another aspect of the present invention includes a plurality of weighing instruments, various combinations of the weights of the articles weighed by these weighing instruments, and a total weight of these combinations falling within an allowable weight range. A combination weigher that discharges a combination article composed of composing articles, a reweighing means that re-weighs each time the combination product is discharged from the combination scale, a total weight of the discharged combination product, and this discharge A calculating means for calculating a subtraction value which is a difference from a reweighing value in the reweighing means for the combination article, and a total value of the subtraction values corresponding to the combination articles in which the article weighed by one measuring instrument does not participate Alternatively, an arithmetic value that is an average value is calculated for each combination article in which the one measuring instrument is different, and a calculation means that calculates a total value or a standard deviation of the average value, each standard deviation, and each standard Standard for deviation The door, and display means for displaying, in which comprises.

  A weighing system according to another aspect of the present invention includes a plurality of weighing instruments, various combinations of the weights of the articles weighed by these weighing instruments, and a total weight of these combinations falling within an allowable weight range. A combination weigher that discharges a combination article composed of composing articles, a reweighing means that re-weighs each time the combination product is discharged from the combination scale, a total weight of the discharged combination product, and this discharge A calculating means for calculating a subtraction value that is a difference from a reweighing value in the reweighing means for the combination article, and a total value of the subtraction values corresponding to the combination articles in which the article weighed by a predetermined number of measuring instruments does not participate Alternatively, the calculation value, which is an average value, is calculated for each different combination of the predetermined number of weighing instruments, and calculation means for calculating the standard deviation of the total value or the average value, and all of the standard deviations are predetermined. Border When abnormal, the a first determining means for determining the variation in the weighing values of the re-metering means or all of the measuring instrument is abnormal, it is to provided.

  When the first determination means has one of the standard deviations smaller than the boundary value, the variation in the measurement value is abnormal in any or all of the predetermined number of measuring instruments corresponding to the standard deviation. It can be determined.

According to the present invention, it is possible to determine whether or not there is an abnormality in the variation of the measurement value of the combination weigher or the reweighing means.

(A), (b), (c) is a figure for demonstrating the conventional method of determining an abnormal measurement hopper. It is a figure used in order to demonstrate embodiment of the measurement system which concerns on this invention, and is a figure which shows the combination of each measurement hopper selected for the combination, the measurement value by each measurement hopper, and the re-measurement value. It is a figure used in order to explain the measuring system concerning the embodiment, and is a figure showing the combination of each measurement hopper selected for combination, the measurement value by each measurement hopper, and the re-measurement value. It is a figure for demonstrating the conventional method of determining an abnormal measurement hopper. (A) is a diagram showing a failure area, an abnormal area, and a normal area of a combination weighing value set in the combination weigher according to the embodiment, and (b) is a reweighing value set in the weight sorter according to the embodiment. It is a figure which shows a failure area | region, an abnormal area | region, and a normal area | region. It is a figure which shows the signal which the combination scale and weight sorter which concern on the embodiment generate | occur | produce. It is a figure which shows the state in which the combination article discharged | emitted from the combination scale which concerns on the embodiment is conveyed to a weight sorter. It is a figure which shows the discharge | emission time interval which the measurement system which concerns on the same embodiment produces | generates, conveyance time, and arrival time interval, (a) is a figure for demonstrating a normal state, (b) demonstrates an abnormal state. FIG. It is a block diagram which shows the structure of the measurement system which concerns on the same embodiment. It is a flowchart which shows the reading process of the group data from the combination scale by the weight sorter which concerns on the embodiment. (A) is a flowchart showing the arrival time interval counting process by the weight sorter, (b) is a flowchart showing the arrival time interval storage process by the weight sorter. It is a flowchart which shows the designation | designated process of the head group data by the weight sorter based on the embodiment. It is a flowchart which shows the count process of the standby timer by the weight sorter based on the embodiment. It is a flowchart which shows the count process of the conveyance timer by the weight sorter which concerns on the embodiment. It is a flowchart which shows the collation process of the data of the combination weigher and weight sorter | selector which concerns on the same embodiment. It is a flowchart which shows the process which memorize | stores the reweighing value and set data by the weight sorter | selector which concerns on the same embodiment as a set. It is a flowchart which shows the process which memorize | stores the reweighing value and set data by the weight sorter | selector which concerns on the same embodiment as a set. It is a flowchart which shows the process of determination calculations, such as abnormality of the weight sorter by the weight sorter concerning the embodiment. It is a flowchart which shows the process of determination calculations, such as abnormality of the combination scale by the weight sorter which concerns on the embodiment. It is a block diagram which shows the state which connected each apparatus of the measurement system which concerns on the embodiment with a communication line.

First, the basic concept of the abnormality diagnosis / recovery method using the weighing system according to the present invention will be described.
[1. (Basic concept of abnormality diagnosis and recovery method)
[1-1. Combination weighing on combination scales)
The combination weigher includes a plurality of n weighing instruments (weighing hoppers), and once a plurality of weighing hoppers are put into a plurality of weighing hoppers and weighed by a substantially constant number or a substantially constant volume, a plurality of known heavy objects are measured. Is selected, an appropriate combination is selected from the plurality of known weight objects, and the combination is selected so that the total weight becomes a preset target weight WT, and the known weight object constituting the selected group is selected. It is a scale that collects a plurality of articles to make a combined article having a substantially constant weight.
In this case, the total weight value of the combined article can be selected to a value close to the target weight WT with a high probability, and is about 1/2 of the total number n of total weighing hoppers (in some cases, about 1/3 when n is large). The total weight value is set to the target weight value by the combination of the number of weighing hoppers). Accordingly, each weighing hopper is designed and adjusted so that an article having a weight of about WT / m (m = n / 2) is loaded.

Next, the basis of a countermeasure method for extracting the bias error while suppressing the influence of the variation error included in each weighing value obtained by the combination weighing machine and the reweighing means such as the weight sorter and the weight inspection machine. The concept is explained below with examples. An example of a combination weigher composed of six weighing hoppers A, B, C, D, E, and F will be given. In this combination weigher, a combination of three weighing hoppers on average is selected from six weighing hoppers to produce one set of combination articles. For example, when a 100 g combination article is desired, a volume or number of articles corresponding to about 33.3 g is put into one weighing hopper.
The combination in which 3 weighing hoppers are selected from 6 weighing hoppers
There are 20 types of 6C3 = of which 10 are combinations of which scale A participates. This is shown in FIG. The remaining 10 patterns in which the weighing hopper A does not participate in the combination are shown in FIG. However, when 20 combination articles that can be continuously formed even if the weighing hoppers and article input devices are evenly arranged are picked up, the 20 types of combinations shown in FIGS. 2 and 3 cannot be made. As described above, the combination weighing value corresponding to the target weight is not established with the three weighing hoppers each time, but the total weight value of the articles put into the three weighing hoppers becomes the target weight. Since the articles are put into the weighing hopper almost uniformly, the probability that a combination is established with three units is extremely high.
Therefore, in order to facilitate understanding of the following description, it is assumed that the combination is always formed by the articles in the three weighing hoppers, and if many combinations are made, the types of combinations shown in FIG. 2 and FIG. Think of it with the same probability. However, the present invention does not assume that the combination result of the combination weigher satisfies this condition. Many various combination results may be temporarily stored, and one that satisfies the above conditions may be selected from them, or a sample that does not satisfy the conditions may be handled.

[1-2. (Extraction of bias error in combination weighing and reweighing means)
Basically, the present invention presupposes that the combination weigher and the reweighing means do not cause an abnormal condition at the same time and that all the weighing hoppers of the combination weigher do not become abnormal at the same time. When the combination weigher is normal, the combination weigher indicates the abnormal state / error amount of the reweighing means, and when the reweighing means is normal, the reweighing means and the normal weighing of the weighing scales of the combination weighers are normal. The amount of error of the abnormal combination weigher is detected by the hopper weighing value, and the abnormal weighing is corrected by the normal weighing value. If the measurement value of the combination weigher is correct for the abnormal state / error amount of the reweighing means, the deviation from the value is corrected.
For each weighing hopper of the combination weigher, the articles weighed by each weighing hopper are not individually weighed by the reweighing means, and the combined articles discharged from the normal and abnormal weighing hoppers are only weighed by the reweighing means. Therefore, the error amount of the specific balance of the combination weigher cannot be detected by simply comparing the reweighing value of the reweighing means and the combination weighing value of the combination weigher. Therefore, the following method is adopted.

FIGS. 2 and 3 are divided into a combination group (group selected for the combination) in which the weighing hopper A participates and a combination group that does not participate, but all the combinations are represented by the two diagrams. And all the weighing hoppers of A to F participate in the 20 sets in the same number.
2 and 3, wa1, wa2, ..., wa10, wb1, ..., wb4, wc1, ..., wd2, ..., we3, ..., wf4, ..., wb11 , ..., wc11, ..., wd11, ..., we12, ..., wf13, ..., wf20 are measurement values of the weighing hoppers A, B, C, D, E, and F of the combination weigher. Wc1, Wc2,..., Wc10,..., Wc19, Wc20 are remeasurement values obtained by the reweighing means for 20 sets of combination articles. In the combination weigher, the combination weighing value of the combination article (within the allowable weight range) for which the combination is established and the reweighing value obtained when the combination article is weighed by the reweighing means are compared for the same combination article as To find the difference.
Paying attention to the group in which weighing hopper A participates,

Wc1− (wa1 + wb1 + wc1) = wya1
・
・
Wc10− (wa10 + we10 + wf10) = wya10 (1.2.1)

These differences only have meaning as errors caused by variations if both the combination weigher and the reweighing means are normal. However, when any weighing hopper or re-weighing means has a zero point error or span movement, the deviation error of the weighing value based on it is also included. (1.2.1) The first subtracting means according to claim 1 performs an operation such as an equation.
In the present invention, for example, the presence / absence evaluation of the error for the weighing hopper A is performed by the total value of the differences wya1, ..., wya10 or the average value thereof. The reason is that each weighing value of each weighing hopper of the reweighing means and the combination weigher has a variation error, but the purpose is to detect a bias error as small as possible contained in the variation error.
In formula (1.2.1), assuming that the variation of each weighing hopper and re-weighing means of the combination weigher is zero, the weighing hopper A of the combination weigher has a weight error of wa1, wa2,. .., And weighing hopper B is weighing Eb1, Eb2,... For weighing values of wb1, wb2,..., And weighing hopper F is weighing wf1, wf2,. Assume that Ef1, Ef2,... Have values, and the reweighing means has errors of Ecw1, Ecw2,... With respect to the measured values of Wc1, Wc2,. Therefore, wya1, wya2,... In equation (1.2.1) is the result of adding the deviation error of the weighing value of each weighing hopper in the combination weigher and the deviation error of those combination articles in the reweighing means, Respectively

wya1 = {Ecw1- (Ea1 + Eb1 + Ec1)}
・
・
wya10 = {Ecw10− (Ea10 + Ee10 + Ef10)} (1.2.2)
It is expressed.
When the total value Swya of the above errors is obtained,
Swya = {Ecw1- (Ea1 + Eb1 + Ec1)} + {Ecw2- (Ea2 + Eb2 + Ed2)} +... + {Ecw10− (Ea10 + Ee10 + Ef10)}
= Ecw1 + Ecw2 + ... + Ecw10-{(Ea1 + Ea2 + Ea3 + .. + Ea10) + (Eb1 + Eb2 + Eb3 + Eb4) +.
It becomes.

(1.2.3) Organize the formula in consideration of the weighing characteristics of the combination weigher and reweighing means.
The bias error is a value obtained by adding the zero point and the span movement amount. However, since the error due to the zero point movement is not affected by the measured weight, even if an article having a different weight is measured each time, it appears as a fixed value. (1.2.3) The first total value calculation means described in claim 2 performs the calculation of the equation or the like.
However, the amount of the span error varies depending on the article weight value. In the combination weigher, if the number of articles selected for the combination is the standard number of combinations, there is a high probability that each weighing hopper is filled with articles of approximately equal weight. Even if the weights of the articles weighed by the hopper are slightly different, they appear with almost the same error. For example, if a combination of two weighing hoppers is selected, the weight of each weighing hopper will be larger than when three weighing hoppers are used, and the amount of span error that appears will be much different than when three weighing hoppers are used. Become.
Accordingly, the combination article to be calculated for the above error addition is selected as the standard combination number, in this example, the combination established by three weighing hoppers. Strictly speaking, if the measurement value of each weighing hopper participating in the combination spreads over a certain range, the set is not adopted as the data for the evaluation calculation.
The span error in the reweighing means is basically considered that the same amount appears for each reweighed value, because the operation of the combination weigher produces a combined article of approximately the same weight each time.

Here, assuming that the re-weighing means is Ecw, the weighing hopper A is Ea, the weighing hopper B is Eb,... Ecw1 = Ecw2 = ... = Ecw, Ea1 = Ea2 = ... = Ea,..., Ef1 = Ef2 = ... = Ef10 = Ef It becomes.
The evaluation of the total value of errors or the average value is the same for other weighing hoppers other than the weighing hopper A at the same time. However, when a limited number of combination weighing values are adopted, not all weighing hoppers participate in them. However, even if the number is not the same, it is desirable to evaluate every weighing hopper with an average value of a certain number or more. Therefore, in practice, the minimum number appropriate for evaluation is determined as P, and a condition is set such that evaluation is performed when all of the weighing hoppers reach a number equal to or greater than P. Of course, all the weighing hoppers may participate in the same number, and the data of the same combination number and different combinations may be selected from the temporary data storage memory.

[1-3. Difference between abnormal state level evaluation of the present invention and the conventional method]
The difference between the weighing hopper abnormal state level evaluation of the present invention and the conventional method will be numerically described.
In the conventional example, it is determined whether or not the combination weighing value of one set of combination articles obtained by the combination weigher and the reweighing value by the reweighing means (check scale) of the combination article is an appropriate amount. It was determined whether or not they matched, and a defective weighing hopper was detected.
Each weighing hopper of the combination weigher has a variation error even if the operation is normal. The variation is set to the same standard deviation value σdw in each of the weighing hoppers A, B, C,. Even if the reweighing means is normal, there is a variation represented by the standard deviation value σcw. In this case, for example, when one of the sets in FIG. 2 as one of the sets including the measurement values of the weighing hopper A, the variation of wya1 in Wc1− (wa1 + wb1 + wc1) = wya1 is expressed by the standard deviation σt1.
σt1 = (3 · σdw2 + σcw2) ^1/2 (1.3.1)
It is.

As specific numerical examples, when the standard deviation σdw = 0.2 g of the weighing value of each weighing hopper of the combination weigher and the standard deviation σcw = 0.3 g of the reweighing value of the reweighing means, σt1 = 0.458 g 3σt1 = 1.37 g.
In order to stably evaluate the presence or absence of the deviation error of the weighing hopper A in this state, the boundary level of the evaluation cannot be set smaller than the width of ± 1.37 g.
On the other hand, when the total value of the deviation wyak and the variation error of the average value regarding the 10 sets of combination articles in FIG. 2 including the measurement value of the weighing hopper A are described, the total measurement value of the 10 sets of combination measurements is The standard deviation σT of the difference between this total measured value and the total value of the reweighed values by the 10 sets of reweighing means is as follows:

σT2 = σdw ² + σdw ² + σdw ² + σdw ² +... + σcw ² + σcw ² +... + σcw ² = 30 · σdw ² + 10 · σcw ²
Therefore
σT = (30 · σdw ² + 10 · σcw ² ) ^1/2 (1.3.2)
It is expressed.
Since this is a variation of the total value of the 10 sets of deviations, σt = σT / 10 (1.3.3) is obtained by taking the average value and expressing the variation of the average value of one set by the standard deviation.

In this case, σt = 0.147 g. Therefore, σt <σt1. That is, the standard deviation σt of the weighing data based on 10 sets of combination articles is σt = 0.147 g, 3σt = 0.441 g, and 3σt1 = 1.37 g in the case of the standard deviation of weighing data based on one set of combination articles. In comparison, the width of the boundary value can be set narrow by 1.37 g−0.441 g = 0.929 g.
The main object of the present invention is to detect the bias error and to determine the magnitude, and to detect the bias error as small as possible. If there is a deviation error Ea in the weighing hopper A with respect to the variation error, the average value is also Ea. Therefore, when evaluating the existence of the deviation error in the presence of the variation error, a smaller deviation is performed on the average value processed value. Evaluation becomes possible.

2 and 3, the deviation error of each weighing hopper of the combination weigher is Ea, Eb, Ec, Ed, Ee, Ef, the deviation error of the reweighing means is Ecw, the combination weigher and the reweighing means When the standard deviation of the variation obtained by subtracting (adding) the measured value of σT is σT and the average value is σt as described above, the combined measured value of the 10 sets of combined articles in which the weighing hopper A participates and the reweighing means The total value Swya of the measurement value difference wyak (k is the number of the set data. K = 1, 2, 3,..., 10) is
Swya = 10 Ecw− {10 · Ea + 4 · (Eb + Ec + Ed + Ee + Ef) ± σT}
(Refer to the formula (1.2.3)). Average value Awya is calculated by dividing Swya by 10
Awya = Ecw− {Ea + (2/5) · (Eb + Ec + Ed + Ee + Ef) ± σt}
As described above, it can be expressed by the sum of the deviation error and the variation error of the combination participation weighing hopper of the reweighing means and the combination weigher.
Accordingly, when the presence or absence of a bias error in the weighing hopper A is determined by a boundary value, the presence of the bias error can be clearly distinguished from the variation error by setting the boundary value to a level exceeding ± 3σt = ± 0.441 g. .
If the number of data sets is increased to 10 or more, 3σt becomes a smaller value, and a smaller deviation can be detected.

Next, the combination in which the weighing hopper B participates is selected from the 20 sets, and as in the case of the weighing hopper A, the combination weighing value of the combination weigher and the reweighing value group by the reweighing means of the combination article are noted. The difference wybk is obtained, the average value Awyb of the total value Swyb of wybk is obtained, and the magnitude is compared with the set boundary value. In this way, the differences wyak, wybk, wyck, wydk, wyek, wyfk are calculated in the order of the weighing hoppers A, B, C, D, E, and F. The first subtraction means), and for each difference, the total value Swya, Swyb, Swyc, Swyd, Swye, Swyf is calculated (the first total according to claim 2 is to calculate each total value) Value calculation means). And each average value Awya, Awyb, Awyc, Awyd, Awye, Awyf is calculated | required (It is the 1st average value calculation means of Claim 2 which calculates each average value.).
In the combination weigher, a method called a memory hopper, which stores items that could not participate in the combination in a separate hopper corresponding to the weighing hopper, and participates in the combination calculation together with the articles staying in the weighing hopper for the next combination calculation There are things. In this case, depending on the combination method, an article in a weighing hopper may be combined with an article in a memory hopper previously weighed by the weighing hopper, and multiple weighing results from the same weighing hopper may be included in one combination. May exist. In such a case, the average value Awyx of wyxk is, for example, when the average value is obtained for the weighing hopper A (x is a to f), it is not included in the number of combination articles that the weighing hopper A participates, but participates in the combination. Divide by the number of items weighed by weighing hopper A. In addition, a plurality of articles weighed by the same weighing hopper may be removed in advance from a combination weighing value that participates in one combination.

[1-4. Method for judging abnormality when assuming that only one abnormality of weighing hopper of combination weigher occurs at the same time]
[1-4-1. (Bias error judgment method for each weighing hopper of combination weigher)
The determination order is performed from the re-weighing means in the embodiment, but will be described first from the abnormality determination calculation of each weighing hopper of the combination weigher.
Assuming that 20 different combinations shown in FIG. 2 and FIG. 3 are adopted from the combination weighing values of the combination weigher, the sum of the difference between the combination weighing value and the re-weighing value of the combination in which the weighing hopper A participates The average value Awya of is obtained by dividing the total value Swya by 10
Awya = Ecw− {Ea + (2/5) · (Eb + Ec + Ed + Ee + Ef)} ± σt
Similarly, for the other weighing hoppers, as in the weighing hopper A, the average value is obtained.
Awyb = Ecw− {Eb + (2/5) · (Ea + Ec + Ed + Ee + Ef)} ± σt
Awyc = Ecw− {Ec + (2/5) · (Ea + Eb + Ed + Ee + Ef)} ± σt
Awyd = Ecw− {Ed + (2/5) · (Ea + Eb + Ec + Ee + Ef)} ± σt
Away = Ecw− {Ee + (2/5) · (Ea + Eb + Ec + Ed + Ef)} ± σt
Awyf = Ecw− {Ef + (2/5) · (Ea + Eb + Ec + Ed + Ee)} ± σt (1.4.1.1)
It is.

However, the bias error values Ecw, Ea,..., Ef represent positive or negative numerical values, and the standard deviation σt represents a positive numerical value. When applied to an actual example, when all possible combinations of combination data are prepared, the combination data in which all the weighing hoppers participate in the same number is selected as the combination to be used in the calculation ( (Select combination data shown in FIGS. 2 and 3). As another method, the minimum number of each weighing hopper participating in the combination data is set in advance, and when all the weighing hoppers participate in the combination at least the set minimum number or more, A, B, C,... There is also a method in which an average value for each weighing hopper designated in the order of is obtained, and the total error value is divided by the total number of designated weighing hoppers included in a plurality of sets of data taken up.
In such a case, the number of participants of each designated weighing hopper and the number of participants of other weighing hoppers included in the specified weighing hoppers are different between the respective sets of designated weighing hoppers. The weighting factor (2/5) for the error of each weighing hopper other than the hopper is not uniform, but in any case, the weighting factor of the other weighing hopper is smaller than the weighting factor of the designated weighing hopper = 1. Choose as follows. In other words, the combination data of the specified weighing hopper and any other weighing hopper must not be the same number, and the combination data should be selected so that the number of other weighing hoppers participates is less than the specified weighing hopper participation number. And
That is, Awya,..., Awyf in equation (1.4.1.1) each include errors of all weighing hoppers, but speaking of Awyx, Awyx has a weighting factor of 1 for the weighing hopper x and is the maximum. And all other weighing hoppers are 2/5.
Therefore, when the deviation error of the weighing hopper x is large and the other weighing hoppers are small deviation errors that are almost negligible compared to the value, the weighing hopper having an abnormal error and the error amount are determined in Away to Awyf. Let's look for the one with the maximum absolute value. Let Awyx be the maximum. The weighing hopper having an abnormal error is limited to one unit, and the error weight of the other weighing hoppers is smaller than the error weight of the weighing hopper x in Awyx. Therefore, the error of the weighing hopper x has almost Awyx. Ex is expressed, and it can be determined by comparing with an abnormal boundary value set separately whether this Ex is abnormal or not.

For example, when the deviation error Ecw = 0 of the re-weighing means, the deviation error Ea = e (e> 0) of the weighing hopper A, and the deviation error Eb = Ec = Ed = Ee = Ef = 0 of the other weighing hoppers, The average value of
Awya = −e ± σt
Awyb = Awyc = Awyd = Awye = Awyf = − (2/5) · e ± σt
It becomes. If the abnormal boundary value R = e is set and the condition of Awyx ≧ R is abnormal, the weighing hopper A can be determined to be abnormal.

Next, the concept of setting the boundary value for determining the presence or absence of a bias error will be described. Assuming that the variation range of the measurement value difference average value Awyx is up to 3σt, when the value of Awyx is within the range of −3σt ≦ Awyx ≦ + 3σt, it is possible to identify whether the value of Awyx is a bias error or a variation error. Absent. Therefore, the boundary value for determining whether or not there is a bias is set to a value larger than 3σt, and the abnormality determination margin value rd with respect to the magnitude of the bias error itself is added to determine the abnormal boundary set value Red as follows.
Red = 3σt + rd (1.4.1.2)
If | Awyx | ≧ Red is satisfied for the weighing hopper x, it is determined that the weighing hopper x is abnormal. It is defined that self-recovery is possible within the abnormality determination area, and the self-recovery function is activated. However, nσt (n is an arbitrary real number) may be used instead of 3σt. Further, the failure boundary value Rtd defined as an error that is difficult to recover from itself is set to a value larger than the abnormal boundary value Red. FIG. 5A shows the state of each boundary value level.
Bias error (bias error considering variation error when expressed accurately) | Awyx |
Rtd> | Awyx | ≧ Red (1.4.1.3)
Is satisfied, the automatic zero point correction command is first output from the reweighing means to the specific weighing hopper x of the combination weigher (the weighing hopper determined to be abnormal among the specified weighing hoppers). The automatic zero correction is applied to the specific weighing hopper x of the combination weigher. The reweighing means waits for the arrival of an automatic zero correction completion signal from the combination weigher.

Next, the combination weighing value obtained by the combination weighing after the automatic zero point correction and the reweighing value are re-determined by, for example, the reweighing means. When | Awyx | is within the range of the error abnormal boundary value, an automatic span correction command and span coefficient data (described later) are output from, for example, the reweighing means, and the span is applied to the specific weighing hopper x in the combination weigher. Apply correction. After the span correction is completed, the combination weigher outputs a span correction completion signal to the reweighing means.
The reweighing means waits for a span correction completion signal from the combination weigher. The transmission of the signal A for fetching the next determination data is made after the elapse of a certain time Tw as in the case of zero point correction. If correction is continued in the order of zero correction and span correction, but an abnormality is still detected, a failure alarm is output. However, from the beginning
| Awyx | ≧ Rtd (1.4.1.4)
If the above holds, the procedure is to output a failure alarm, and the operator does not perform the self-return operation but waits for inspection and repair by the operator. Depending on the specifications, an operation stop command for the combination weigher may be output to stop the operation. here,
| Awyx | <Red (1.4.1.5)
If is true, the combination weigher is normal. The span correction method will be described later.

[1-4-2. (Bias error judgment method of reweighing means)
The reweighing means and the combination weigher shall not cause any abnormality or failure at the same time. However, when there is a large error in the re-weighing means or in any one weighing hopper of the combination weigher in the average value Awyx of the weighing value difference obtained by the equation (1.4.1.1), this Awyx Since the absolute value becomes large, it is impossible to determine whether the cause of the large error is the reweighing means or the weighing hopper of the combination weigher. Note that the average value Awyx of the measurement value difference can identify the measurement hopper having the abnormality when any of the measurement hoppers is known to be abnormal.
Therefore, in order to determine which of the reweighing means and the weighing hopper of the combination weigher is abnormal, it is necessary to determine whether the reweighing means is abnormal or not. To do.
Assume that a specific unit for each weighing hopper is divided into groups that do not participate in the combination, and the total value or average value of the weighing value differences is obtained for each group to determine whether there is an error in the reweighing means.
In FIG. 2 and FIG. 3, there are 10 combination groups from 11 to 20 in which the weighing hopper A does not participate. Taking these 10 sets and calculating the non-participating total value Swyna of the weighing hopper A,

Swyna = {Wc11− (wb11 + wc11 + wd11)} + {Wc12− (wb12 + wc12 + we12)} +... + {Wc20− (wd20 + we20 + wf20)} = 10Ecw-6 (Eb + Ec + Ed + Ee + Ef) ± σT
It becomes. Therefore, if you calculate the average value by dividing Swyna by the number of pairs,
Awyna = Ecw−3 / 5 · (Eb + Ec + Ed + Ee + Ef) ± σt
It is guided.
Similarly, the average values of the combination groups in which the weighing hoppers B, C, D, E, and F do not participate are as follows.
Awynb = Ecw−3 / 5 · (Ea + Ec + Ed + Ee + Ef) ± σt
Awync = Ecw−3 / 5 · (Ea + Eb + Ed + Ee + Ef) ± σt
Awynd = Ecw−3 / 5 · (Ea + Eb + Ec + Ee + Ef) ± σt
Awine = Ecw−3 / 5 · (Ea + Eb + Ec + Ed + Ef) ± σt
Awynf = Ecw−3 / 5 · (Ea + Eb + Ec + Ed + Ef) ± σt (1.4.2.1)

When the re-weighing means has a large error and each weighing hopper of the combination weigher has no error, all of Awynx have a large value (abnormal value). When the re-weighing means does not have a large error and any one weighing hopper of the combination weigher has a large error, at least one of Awynx has a normal value. Therefore, it is possible to determine the failure / abnormality of the reweighing means by examining the magnitude of Awynx under the condition that the reweighing means and the combination weigher do not fail or become abnormal at the same time. Any one of Awynx does not include an error of the weighing hopper that is faulty or abnormal, so that there is at least one that takes a small absolute value if the reweighing means has no error.

Regarding the setting of the magnitude of the abnormality determination boundary value Rec, assuming that the variation of the average value of the combination measurement value by the reweighing means and the combination weigher is σt in standard deviation, the determination boundary value expects a variation error up to 3 sigma. In addition, the abnormality determination margin value rc of only the bias value is added.
Rec = 3σt + rc (1.4.2.2)
And set.
On the other hand, an appropriate numerical value satisfying Rtc> Rec is set as the failure boundary value Rtc of the reweighing means. About all | Awynx |
| Awynx | ≧ Rtc (1.4.2.3)
If the above holds, the reweighing means determines that a failure has occurred and outputs an alarm signal. In this case, the self-recovery function does not work and the operator waits for inspection / repair. The combination weigher may be automatically stopped. If at least one | Awynx | does not satisfy the formula (1.4.2.3), the re-weighing means is not a failure, and the process proceeds to the next determination. For all | Awynx |
Rtc> | Awynx | ≧ Rec (1.4.2.4)
Is satisfied, it is determined that the reweighing means is in an abnormal state. First, a supply stop command signal for the combined article is output from the reweighing means to the combination weigher, for example. This signal causes the combination weigher to stop discharging at least one combination article. As a result, the reweighing means eventually has at least one time interval for arrival of the combined article, and the automatic zero point correction command of the reweighing means itself works to execute automatic zero point correction.
However, if the supply capacity of the combination weigher is low and the automatic zero point correction of the reweighing means can be operated normally between the combination articles supplied to the reweighing means, the combination articles are temporarily stopped with respect to the combination weigher. It is not necessary to issue a command signal. For example, when a zero correction is required, a timer is activated after the reweighing means has completed the weighing of one combination article, and the weighed combination article completely leaves the weighing conveyor of the reweighing means. Allowing for a stable time allows automatic zero correction for the reweighing means. After transmitting this command, at least after the confirmation of the execution of the automatic zero correction operation of the reweighing means itself, the next determination data (set data) starts to be collected.

As a result, the next | Awynx | is re-determined by the re-weighing means. When it is determined that | Awyx | is within the range of the error abnormal boundary value, the automatic span correction function of the reweighing means itself is activated. However, for at least one | Awynx |
| Awynx | <Rec (1.4.2.5)
When the above holds, it is determined that the reweighing means is not in an abnormal state.
When all Awynx are less than the failure boundary value Rtc but larger than the abnormal boundary value Rec, an instruction is given to the combination weigher to create a mode in which one or a plurality of combination articles are not made continuously, and the combination article to the reweighing means A predetermined time interval is given to the supply, and the reweighing means gives a margin for applying automatic zero correction by its function. As a result, automatic span correction is applied when an abnormality is further determined. If the result of span correction is still abnormal, a failure alarm is output. Depending on the specification, operation stop of the combination weigher may be commanded.
FIG. 5B shows the failure boundary value Rtc and abnormal boundary value Rec of the reweighing means, and FIG. 5A shows the failure boundary value Rtd and abnormal boundary value Red of the combination weigher.

[7-4-3. (Optimal boundary value setting method)
In the determination method described above, the larger the number of sets of the calculation target (1.4.1.1) and (1.4.2.1), the smaller the variation of the average value as already explained. Since the participation ratios of the weighing hoppers are also aligned, it is possible to make a highly accurate determination. That is, the absolute value of the boundary value can be set small by calculating using a large number of sets. In order to reduce the influence of the variation error, it is only necessary to make a determination on the condition that the number of combination participation of the weighing hopper is a numerical value equal to or greater than a certain set number (standard number) in advance. However, the longer the number of measurements, the longer it takes to determine. On the other hand, we want to deal with large abnormalities quickly, and on the other hand we want to make highly accurate determinations. Next, the method will be described.
As a countermeasure, we will adopt a method to detect abnormalities quickly while taking advantage of the characteristics of the average method described above. In this case, since the total number of participations of the individual weighing hoppers is 6, the standard deviation σT of the entire two sets is determined as an error average value in the case of the minimum participation number 2 of the designated weighing hoppers.
σT = (6 · σdw ² + 2 · σdw ² ) ^1/2 (1.4.3.1)
It becomes. The standard deviation σt2 of one set of average values is
σt2 = σT / 2 = ((3/2) · σdw ² + (1/2) · σcw ² ) ^1/2 (1.4.3.2)
It becomes. When 3σt2 is used as the boundary value, the abnormal boundary value Red of the combination weigher is
Red = rd + 3σt2 (1.4.3.3)
And When rd = 3σt2,
Red = 6σt2 (1.4.3.4)
It is. This value is set as Red2.

Similarly, when determining with 3 pairs of minimum weighing hoppers, the standard deviation σT of the entire 3 groups is
σT = (9 · σdw ² + 3 · σcw ² ) ^1/2
Therefore, the standard deviation σt3 of one set of average values is
σt3 = σT / 3 = (σdw ² + (1/3) · σcw ² ) ^1/2 <σt2
It becomes. Therefore,
Red3 = 6σt3 <Red2
Thus, as the number of weighing hoppers participating increases, a smaller boundary value can be set. In this way, the optimum boundary value for the determination at that time is determined in accordance with the current participation number of the participation weighing hoppers while considering the variation error. While the abnormal boundary value is large while the number of data is small, only a large abnormality can be determined. However, as the number of data increases, the reliability of the determination criterion increases, so that a small abnormality can be determined. The failure boundary value can be set similarly.
Further, the boundary value setting method in the case of the re-weighing means can be determined by the same method, and detailed description thereof is omitted.
In determining the boundary value as another method, it is not always necessary to directly correspond to the standard deviation value. The total value of the combination weighing value and the total value of the reweighing value of the reweighing means or the total of the combination weighing values One or a plurality of appropriate numerical values corresponding to the dispersion of the average value of the total value of the value and the reweighing value of the reweighing means may be prepared and applied to the boundary value according to the number of combinations. These variations in the average value are such that when the number of measurement values increases and the number of measurement values for which the average value is calculated can be increased, the variation in the average value of the measurement values becomes smaller. A plurality of boundary values that are reduced according to the increase in the number of calculation objects (number of measurement values) are prepared, and the boundary values are selectively applied according to the number of measurement values.
In order to set σtn, it is necessary to obtain the variation error σdw of the weighing value of each weighing hopper of the combination weigher and the variation error σcw of the reweighing value of the reweighing means prior to the operation operation.

[1-4-5. (Automatic standard deviation setting method)
Regarding how to obtain the variation error σcw in the operation condition of the reweighing means, in the test mode before the operation operation (the main operation operation), the reweighing means can be used for a representative arbitrary combination article having a known weight under the operation conditions. There is a method of obtaining the standard deviation σcw of the re-weighing value of the re-weighing means by flowing a plurality of times above. This is a method of obtaining by a conventional method.
On the other hand, it is not easy to obtain the variation error of each weighing hopper under the operating conditions of the combination weigher.
Therefore, for the combination weigher, as a test operation mode before the operation operation, after all the weighing hoppers are charged with the amount of articles to be input during the actual operation, the measurement value (this is referred to as the operation clock amount value). First, measure the measured value at the timing of acquisition, then determine the measured value after the weighing hopper has stabilized sufficiently (this is called the stable clock value), and for each weighing hopper, After obtaining the deviation of the stable clock amount value and obtaining a plurality of deviations, the standard deviation value of the variation of each weighing hopper can be obtained. In the test operation, the above measurement / calculation sequence is automatically operated. These obtained standard deviation values represent the standard deviation value σdw of the variation of each weighing hopper of the combination weigher during the actual operation.
Each weighing hopper of the combination weigher is designed and adjusted so that the weighing values are evenly distributed. Instead of obtaining σdw for each weighing hopper, the average value of the standard deviation value σdw of each weighing hopper is used to represent a typical combination weigher. Use the standard deviation value of the weighing hopper. According to the ninth aspect of the present invention, the standard deviation calculating means for the weighing hopper calculates the standard deviation value σdw of the variation in the measured value as described above.

In the weighing system of the combination weigher and the reweighing means, a method for automatically obtaining the standard deviation value of the variation in the measured value of the reweighing means in the test operation mode will be described.
As described above, since the combination weigher has obtained a stable clock value in the test operation mode as described above, when the combination calculation is performed using these values and the measurement value of the combination article is obtained, this measurement value is approximately the true value of the combination article. Represents the weight value. This combination article is conveyed to the reweighing means, and the reweighing means is operated under the same conditions as the operation including the preceding conveyor, and weighs the combination article. In this way, the reweighing value by the reweighing means has the same variation as during operation. After obtaining a true combination weighing value from the combination weigher, a deviation is obtained from the reweighing value of the reweighing means.
In this way, a plurality of deviations between the combination weighing value based on the stable clock value supplied from the combination weigher and the reweighing value in the reweighing means are obtained, and the standard deviation value σcw of the deviation variation is obtained at an appropriate weighing number, This value represents the standard deviation value σcw of the reweighing value during operation of the reweighing means. Of course, this does not include the variation of each combination metric value.
In this way, the standard deviation values σdw and σcw of the variation of the operation clock amount value can be respectively determined automatically in the test operation mode simultaneously in both the combination weigher and the reweighing means. Assuming that the standard deviation value of the weighing value of each weighing hopper of the combination weigher is σdw and the standard deviation value of the reweighing value of the reweighing means is σcw, this is transmitted to the reweighing means at the end of the test operation. In the means, σcw and σdw are stored in the memory. In the calculation for determining the magnitude of the error amount, the stored values σcw and σdw of the memory are used, and σT and σtn are calculated corresponding to the number of participation of the weighing hopper in the combination metric value used for the calculation at that time point, The boundary values such as Red and Rcd are obtained again and the set values are changed.

[1-5. (Judgment method when the weighing hopper of the combination weigher causes multiple abnormalities at the same time)
In the conventional method, it is not possible to determine the abnormality of a plurality of weighing hoppers at the same time. Next, in the above section, the average error value Awyx of the weighing hopper x is applied to the determination of one abnormal weighing hopper, but the determination method based on this value can be applied when abnormality occurs in a plurality of weighing hoppers at the same time. Will be explained.
[1-5-1. Abnormality judgment method of reweighing means)
Again, it is assumed that the reweighing means and the combination weigher do not cause any abnormality or failure at the same time. The average error of a group in which any two weighing hoppers in the combination weigher are not participating in the combination is obtained. An error between the combination weighing value and the re-weighing value in which the weighing hoppers x and y do not participate is expressed as wynxyr. However, r represents the group number of a combination measurement value. 2 and FIG. 3, there are 6C2 = 6C4 = 15 combinations in which any two of the six weighing hoppers do not participate (or combinations in which four of the six hoppers participate). For example, the groups in which the weighing hoppers A and B do not participate in the combination are the groups 17, 18, 19, and 20 in FIG.
Wc17− (wc17 + wd17 + we17) = wynab17
Wc18− (wc18 + wd18 + wf18) = wynab18
Wc19− (wc19 + we19 + wf19) = wynab19
Wc20− (wd20 + we20 + wf20) = wynab20 (1.5.1.1)
It becomes. Therefore, the total error value Swynab of the weighing hoppers A and B non-participating weighing values is
Swynab = (Ecw17 + ... + Ecw20)-{(Ec17 + Ec1 8 + Ec19) + (Ed17 + Ed18 + Ed20) + ... + (Ef18 + ... + Ef20)} = 4Ecw- {3 (Ec + Ed + Ee + Ef)}
It becomes. The error average value Awynab of the weighing hoppers A and B non-participating weighing values is
Awynab = Ecw−3 / 4 (Ec + Ed + Ee + Ef) (1.5.1.2)
It becomes. Similarly,
Awayac, Awynad, ... 15 ways to Awynef are obtained.

As a general expression, if the error average value without participation of the weighing hoppers x and y is Awynxy, the reweighing means has a large error, and when each weighing hopper of the combination weigher has no error, all of Awynxy are large. Take the value (abnormal value). When the reweighing means does not have a large error and any one or two weighing hoppers of the combination weigher have a large error, since there is an Awynxy excluded from those weighing hoppers, at least within the Awynxy One of them takes a normal value.
Therefore, under the condition that the reweighing means and the combination weigher are not in an abnormal state at the same time, the reweighing means is checked by checking the magnitude of Awynxy, that is, by determining whether or not all Awynxy are abnormal values. Can be determined.

[1-5-2. (Bias error judgment method for each weighing hopper of combination weigher)
If the two weighing hoppers of the combination weigher become abnormal at the same time, it is necessary to verify whether the Awyx of the above-mentioned formula (1.4.1.1) can be used for the determination.
In Awyx, when the abnormal boundary value is e, the error Ecw = 0, Ea = e, Eb = −5 / 2e, Ec = Ed = Ee = Ef = 0 when Aweya = 0, Awyb = − Since it is 5 / 2e, the error of the weighing hopper A cannot be detected from the value of Awya. Further, when the errors Ecw = 0, Ea = e, Eb = −e, Ec = Ed = Ee = Ef = 0 of the reweighing means, Awya = 3 / 5e, Awyb = −3 / 5e, Awyc =. Since = Awyf = 0, any Awyx is less than the abnormal boundary value. Therefore, the magnitude of the error of the weighing hopper x cannot be determined directly from Awyx.
Therefore, first, it is assumed that a weighing hopper knows which weighing hopper has a large error by Awyx, and then takes a procedure of checking whether the weighing hopper has an error and determining whether it is abnormal.

Since it is assumed that there is a large error in the weighing values of the two weighing hoppers at the maximum, the amount of error is set to Ex and Ey for the two weighing hoppers with errors, and the weighing values of the other weighing hoppers. And the error Ecw of the reweighing value of the reweighing means are 0. Since the amount of error is determined by an absolute value, a weighing hopper having the maximum error is found by comparing the magnitudes of absolute values of Awyx, Awyy, and other error average values (typically expressed as Awyz). Like that. For this purpose, | Awyx |, | Awyy |, and | Awyz | If it can be proved that at least one of | or | Awyy | is necessarily larger than | Awyz | for errors Ex and Ey of arbitrary magnitude, it is possible to first find the one having the maximum error average value. it can. This makes it possible to identify the weighing hopper having the maximum error. In order to simplify the explanation, when the error of the weighing hopper x and the weighing hopper y is replaced with any real number x and y that are not 0,

| Awyx | = | x + (2/5) y |
| Awyy | = | y + (2/5) x |
| Awyz | = | (2/5) · (x + y) |
For any real number x, y that is not 0,
| X + (2/5) y |> | (2/5) · (x + y) |
| Y + (2/5) x |> | (2/5) · (x + y) |
Or
| X + (2/5) y | or | y + (2/5) x |> | (2/5) · (x + y) | (1.5.2.1)
It is only necessary to show that
If the weighing hopper having the maximum error can be identified, a method for examining the case of only one abnormal weighing hopper by removing the weighing value in which the weighing hopper participates can be applied. The necessary and sufficient condition for the above inequality to hold is
(X + (2/5) y) ² or (y + (2/5) x) ² > {(2/5) (x + y)} 2 (1.5.2.2)
Is established, or
(X + (2/5) y) 2 <{(2/5) (x + y)} 2, and
(Y + (2/5) x) 2 <{(2/5) (x + y)} 2 (1.5.2.3)
It is sufficient to prove that these two inequalities do not hold simultaneously.

(Proof)
P = (x + (2/5) y) ² − {(2/5) (x + y)} ² = (x / 25) · (21x + 12y)
Therefore, P> 0 when x> 0 and y> 0, and P> 0 when x <0 and y <0. Therefore,
| X + (2/5) y |> | (2/5) (x + y) | and | y + (2/5) x |> | (2/5) (x + y) | (1.5.2.4)
Is established. next,
(A) x> 0, 21x + 12y <0,
Or (b) x <0, 21x + 12y> 0
Is established, P <0, that is, | x + (2/5) y | <| (2/5) (x + y) | is established. Similarly,
Q = (y + (2/5) x) ² − {(2/5) (x + y)} ²
= (Y / 25). (21y + 12x)
Therefore, when x> 0 and y> 0, Q> 0, and when x <0 and y <0, Q> 0. And
(C) y> 0, 21y + 12x <0, or (d) y <0, 21y + 12x> 0
Is established, Q <0, that is, | y + (2/5) x | <| (2/5) (x + y) |

In the above formula, when x> 0, y> 0, or x <0, y <0, P> 0 and Q> 0.
| X + (2/5) y |> | (2/5) (x + y) |
And | y + (2/5) x |> | (2/5) (x + y) |
Holds at the same time.
Next, (c) or (d) is established when (a) is established, and (c) or (d) is not established when (b) is established,
| X + (2/5) y | <| (2/5) (x + y) |
And | y + (2/5) x | <| (2/5) (x + y) |
Are not established at the same time.
When (a) is satisfied, that is, when x> 0, since y <− (21/12) x <0, equation (c) is not satisfied.
With respect to y> − (12/21) x in (d), since y is y <− (21/12) x in (a), equation (d) does not hold.
When (b) is satisfied, that is, when x <0, since y> − (12/21) x> 0, equation (d) is not satisfied.
With regard to y <− (12/21) x in (c), since y is y> − (21/12) x in (b), equation (c) is not satisfied.
Therefore, | x + (2/5) y | <| (2/5) (x + y) | and | y + (2/5) x | <| (2/5) (x + y) | When x and y are any real numbers other than 0,
| X + (2/5) y |> | (2/5) (x + y) |, or
| Y + (2/5) x |> | (2/5) (x + y) | (1.5.2.5)
Either inequality is established.
From the above proof, when the absolute value of the average error value is examined under the condition that there is a large error for the two weighing hoppers in the most cases, Awyx taking the largest value is the value of the weighing hopper having the two large errors. It can be determined that it corresponds to one of the weighing hoppers x.

Next, in (1.4.1.1), in determining the error average value Awyx for each specified weighing hopper in the error between the combination weighing value of the combination weigher and the reweighing value of the reweighing means, the combination of the designated weighing hopper in the data used for the calculation Since the number of participants is chosen to be greater than the number of participants of other weighing hoppers, expressing Awyx in a more general form,
Awya = Ecw− {Ea + (n / m) · (Eb + Ec + Ed + Ee + Ef)} ± σt
Similarly, when the average value of other weighing hoppers is obtained in the same manner as the weighing hopper A, Awyb = Ecw− {Eb + (n / m) · (Ea + Ec + Ed + Ee + Ef)} ± σt
Awyc = Ecw− {Ec + (n / m) · (Ea + Eb + Ed + Ee + Ef)} ± σt
Awyd = Ecw− {Ed + (n / m) · (Ea + Eb + Ec + Ee + Ef)} ± σt
Away = Ecw− {Ee + (n / m) · (Ea + Eb + Ec + Ed + Ef)} ± σt
Awyf = Ecw− {Ef + (n / m) · (Ea + Eb + Ec + Ed + Ee)} ± σt
However, m>n> 0 (1.5.2.6)
It is.
As above, | Awyx | = | x + (n / m) y |
| Awyy | = | y + (n / m) x |
| Awyz | = | (n / m) (x + y) |
For any real number x, y that is not 0 in
| X + (n / m) y | or | y + (n / m) x |> | (n / m) (x + y) | (1.5.2.7)
It is sufficient to show that

As above, P = x. (Mn) / m @ 2 {(n + m) x + 2ny}
Q = y. (Mn) / m @ 2 {(n + m) y + 2nx}
Since m>n> 0, mn> 0
Therefore, if x> 0, y> 0, P> 0, Q> 0, and x <0, if y <0, P> 0, Q> 0.
| X + (n / m) y | and | y + (n / m) x |> | (n / m) (x + y) |
Further, since (mn) / m2> 0, similarly to the conditional expressions (a), (b), (c), and (d),
(A) x> 0, (m + n) x + 2ny <0 or (b) x <0, (m + n) x + 2ny> 0,
(C) y> 0, (m + n) y + 2nx <0 or (d) y <0, (m + n) y + 2nx> 0
Here, since m + n> 2n, similarly,
| X + (n / m) y | or | y + (n / m) x |> | (n / m) (x + y) | (1.5.2.8) can be proved.

When the weighing hopper x can be specified, an average error value is obtained in the combination group excluding the weighing hopper x. When the weighing hopper x is the weighing hopper A, the total value Swybna of the combination group (set) of the non-participation of the weighing hopper A and the participation of the weighing hopper B, the total value Swycna of the combination group of the non-participation of the weighing hopper A and the participation C ········································ Swifna total value of the combination group of weighing hopper A non-participation / F participation, and average value Awydna,.
2 and FIG. 3, the weighing hopper A non-participating group is a combination of 11 to 20, of which the weighing hopper B participating group is 11, 12,...
Swybna = 6Ecw− {6Eb + 3 (Ec + Ed + Ee + Ef)}
Therefore,
Awybna = Ecw− {Eb + ½ (Ec + Ed + Ee + Ef)}
Similarly for each weighing hopper,
Awycna = Ecw− {Ec + ½ (Ec + Ed + Ee + Ef)}
Awydna = Ecw− {Ed + 1/2 (Eb + Ed + Ee + Ef)}
Awyena = Ecw− {Ee + 1/2 (Eb + Ec + Ed + Ef)}
Awyfna = Ecw− {Ef + ½ (Ec + Ed + Ee + Ef)}
... (1.5.2.9)
Therefore, the idea that only one weighing hopper has a large error can be applied. Among these average values, the one having the maximum absolute value is found, and it is determined whether or not it is larger than the boundary value. If it is larger than the boundary value, the weighing hopper is determined to be abnormal. If it is the weighing hopper B that takes the maximum value regardless of normality / abnormality, this time, the error average value Awyanb of the group in which the weighing hopper A participates and the weighing hopper B does not participate is obtained. From FIG. 2 and FIG.
Awyanb = Ecw− {Ea + (1/2) (Ec + Ed + Ee + Ef)}
... (1.5.2.10)
Therefore, the value of Ea is obtained from this equation, and abnormality / normality is determined by the boundary value.

Regarding the possibility that three weighing hoppers may cause an abnormality at the same time, using the same error average value Awyx, identify weighing hoppers with a maximum of three errors and if the difference is larger than the boundary value. It is possible to determine whether or not.
In order to identify three weighing hoppers with large errors, if one of the three can be identified first, the second and first weighing hopper can be identified based on the same idea as above, and the error Normality can also be determined.
To do so, let x, y, z be any non-zero real number,
| X + (2/5) (y + z) |, | y + (2/5) (x + z) |, | z + (2/5) (x + y) | are all simultaneously | (2/5) (x + y + z) | We just need to prove that smaller things don't happen.
As with the two units,
{X + (2/5) (y + z)} 2 <{(2/5) (x + y + z)} 2 (A)
{Y + (2/5) (x + z)} 2 <{(2/5) (x + y + z)} 2 (b)
{Z + (2/5) (x + y)} 2 <{(2/5) (x + y + z)} 2 (c)
... (1.5.2.11)
What is necessary is just to prove that these three formulas do not hold simultaneously.
We abbreviate the calculation in the middle, and for (b), (b) and (c) respectively,
(A): (a1) x> 0, 21x + 12 (y + z) <0 or (b1) x <0, 21x + 12 (y + z)> 0
(B): (a2) y> 0, 21y + 12 (x + z) <0 or (b2) y <0, 21y + 12 (x + z)> 0
(C): (a3) z> 0, 21z + 12 (x + y) <0 or (b3) z <0, 21z + 12 (x + y)> 0
Because
(A1) (a2) (a3), (a1) (a2) (b3), (a1) (b2) (b3), (a1) (b2) (a3) and (b1) (b2) (b3) , (B1) (b2) (a3), (b1) (a2) (a3), (b1) (a2) (b3), it is only necessary to prove that each of the three sets does not occur simultaneously. .

It is clear that (a1), (a2) and (a3) do not hold at the same time.
(A1) (a2) (b3)
Under x> 0, y> 0, z <0,
From (a1), z <− (21x + 12y) / 12
From (a2), z <− (12x + 21y) / 12
When z> -12 (z + y) / 21 in (b3),
| 12 (z + y) / 21 | <| (21x + 12y) / 12 |
| 12 (z + y) / 21 | <| (12x + 21y) / 12 |
This is not true. Therefore, (a1), (a2), and (b3) do not hold simultaneously.
For (a1), (b2), and (b3), x <−12 (y + z) / 21 from (a1) under x> 0, y <0, z <0, but in (b3) x> − (12y + 21z ) / 12 must hold.
However, since | 12 (y + z) / 21 | <| (12y + 21z) / 12 |, the above inequalities do not hold simultaneously. Therefore, (a1), (b2), and (b3) do not hold simultaneously.

For (a1), (b2), and (a3), y <− (21x + 12z) / 12 from (a1) and y> −12 (x + z) / from (b1) under x> 0, z> 0, and y <0. 21 must hold at the same time,
Since | 12 (x + z) / 21 | <| (21x + 12z) / 12 |
The above inequalities do not hold at the same time. Therefore, (a1), (b2), and (a3) do not hold simultaneously.
Since (b1) can be proved in the same manner, it is omitted.
From the above proof, it can be determined that the weighing hopper corresponding to the largest of all Awyx absolute values may have a large error, so a set of error average values excluding this weighing hopper is obtained, Furthermore, since the largest absolute value among them can be determined as a weighing hopper that may have a large error, by calculating the error average value excluding the weighing hopper, the magnitude of the error of the remaining weighing hopper, It is obtained from the average error value and determined whether or not there is an abnormality. When the error value of the third unit is obtained, the error value is obtained in order of the second and first units based on the error value, and abnormality / normality is determined.

[1-6. Abnormality judgment due to variation)
[1-6-1. Causes of abnormal variations in combination weighers and reweighing means)
A method for determining an abnormality of the weighing hopper based on the magnitude of the variation itself is also conceivable. For combination weighers and reweighing means, unpredictable floor vibrations, electric noise, attachment and detachment of articles to / from the weighing conveyor of the weighing hopper and reweighing means, and temporary items on the weighing hopper In the case where the ink adheres and leaves after a while, an abnormal state different in nature from drift in a long time may occur, such as unstable output due to trouble of the load cell for some reason.
Variation due to floor vibration or electrical noise is likely to occur at the same time on the combination weigher and reweighing means. There is a high probability that abnormal variations will occur. These phenomena can be judged more accurately by observing the amount of variation than by observing the average drift amount of the measured value. For example, when a large positive and negative error is repeated, the average value cannot be detected, but an abnormality can be determined by looking at the amount of variation.

[1-6-1. Abnormal variation judgment method of combination weigher and reweighing means]
Therefore, as variation determination data, the variation in error values of the combination groups in which the weighing hoppers A, B, C,. For example, when taking a combination weighing value group in which weighing hopper A does not participate from FIGS. 2 and 3,
Wc11− (wb11 + wc11 + wd11) = wyna1
Wc12− (wb12 + wc12 + we12) = wyna2
・
・
Wc20− (wd20 + we20 + wf20) = wyna10 (1.6.2.1)
Therefore, the standard deviation σna is obtained from the error values wyna1, wyna2,..., Wyna10 as a value representing the variation amount. If σna is called the standard deviation when the weighing hopper A is not participating, the standard deviations σnb, σnc, σnd, σne, and σnf when the weighing hoppers B, C, D, E, and F are not participating are obtained in the same manner.
On the other hand, as the abnormal boundary value, the standard deviation σcw of the reweighing means and the standard deviation σdw of each weighing hopper of the combination weigher as the standard deviation at normal time are appropriately set in the larger direction. The total value with the defined abnormal range value s is Rσ,
Rσ = σT + s (1.6.2.2)
Set. In accordance with the number of combinations used, Rσ also recalculates the portion of σT. Further, since this evaluation examines the variation of the specific weighing hopper, an error due to the same combination, for example, data of wyna1 may be included many times in the evaluation calculation group for obtaining the standard deviation.

Next, an abnormality determination method for the reweighing means and the combination weigher when only one weighing hopper of the combination weigher may cause an abnormality at the same time will be described.
When at least one of σna to σnf is smaller than the boundary value Rσ, it is determined that the variation of the reweighing means is normal. When all of σna to σnf are equal to or larger than Rσ, it is determined that all the weighing hoppers or reweighing means of the combination weigher are abnormal in dispersion.
Under the condition that the variation of the re-weighing means is normal, when the σna of the combination weigher not participating in the weighing hopper A is a normal value and the other σnb to σnf are abnormal values, it is normal if the weighing hopper A is not participating Since it is abnormal if it is a participation, the weighing hopper A determines that the variation is abnormal. Similarly, it is determined that the weighing hopper x is abnormal when the non-participating σnx is a normal value and the standard deviation excluding σnx is normal.
When the operator wants to determine the magnitude of the variation, the standard deviation σna to σnf, the standard deviation σcw of the variation of the reweighing means obtained in the test operation mode as the standard value, and each weighing hopper of the combination weigher If the standard value σT calculated from the representative standard deviation σdw of the variation and the number of adopted data is displayed on the display operation unit, the degree of variation can be easily determined.

When there is a variation abnormality in the two weighing hoppers, at least one of the abnormal weighing hoppers is always added to σnx. Therefore, σnx always becomes an abnormal value, and it is possible to distinguish the abnormality between the reweighing means and the combination weigher. Absent.
As a countermeasure, first, in order to identify two abnormal weighing hoppers, error data of a combination group of two weighing hoppers not participating is obtained. For a combination weigher with 6 weighing hoppers, 6C2 = 15 calculations are required. If the standard deviation of non-participating weighing hoppers x and y is σnxy, if all of the 15 σnxy are abnormal, all weighing hoppers or reweighing means of the combination weigher can be judged as abnormal variations, and any one is normal For example, it can be determined that either or both of the weighing hoppers x and y of the combination weigher are abnormal. If it can be specified as one of the weighing hoppers x and y, the standard deviation of the weighing hopper x non-participation, y participation, and the weighing hopper x participation and y non-participation is examined among 15 standard deviations.
However, for the abnormal boundary value, it is necessary to obtain a new value as a reference value σT of a normal value in advance by obtaining a new value on the condition that the two weighing hoppers do not participate, and set a new Rσ. When the standard deviation of the weighing hopper x non-participation and the weighing hopper y participation is in a normal range, it can be determined that the weighing hopper y is normal and the weighing hopper x is abnormal. When the standard deviation of the weighing hopper x non-participation and the weighing hopper y participation is in an abnormal range, the weighing hopper y is abnormal but it cannot be determined what the weighing hopper x is. In that case, the standard deviation of the weighing hopper x participation and the weighing hopper y non-participation is checked, and if it is in a normal range, it can be determined that the weighing hopper x is normal and if it is in an abnormal range, the weighing hopper x is determined to be abnormal.

[1-7. Failure / Abnormal Logic Judgment Method)
Further, as a method different from the method described so far, a method of specifying an abnormal weighing hopper earlier than the conventional method is shown, although one combination data is used according to the conventional method. It is important to identify and correct abnormal weighing hoppers at an early stage in the sense that mass production of abnormal combination articles will not increase loss.
Examples of discrimination by the conventional method are shown in FIGS. (A) indicates that the weighing hoppers participating in the first set of combination articles are A, B, C, and the second set is A, D, E. The abnormal weighing hopper is A, and all other weighing hoppers are normal. Since the abnormal weighing hopper A participates in the first set, the determination result is abnormal when the combination article is weighed by the reweighing means. Separately, an abnormal counter is provided for each weighing hopper, and when the first set is determined to be abnormal, the abnormal counters for the weighing hoppers A, B, and C participating in the first set are incremented by one. Here, the numerical values of the abnormality counters are compared with each other, and if there is one of the maximum number among them, the weighing hopper corresponding to the abnormality counter is regarded as abnormal.
In (a), in the first set, 1 is in three counters, so an abnormal weighing hopper cannot be determined. The weighing hoppers A, D, and E participate in the second set. When it is determined that the weighing hoppers A, D, and E are also abnormal, the abnormality counters corresponding to the weighing hoppers A, D, and E are incremented by one. Here, when the number of abnormal counters is compared again, the counter corresponding to the weighing hopper A is 2 and the others are 1 at most, and it can be determined at this point that the weighing hopper A is abnormal.
However, when combinations appear as shown in (b), up to four sets cannot be determined because the counters corresponding to the weighing hoppers A and C indicate 2. For the first time in the fifth set, the weighing hopper A counter is 3, the weighing hopper C counter 2, and the others are 1 or 0, and it is found that the weighing hopper A has the maximum counter, and the weighing hopper A can be determined to be abnormal.

A temporary memory (second storage unit) and a storage memory (first storage unit) that store normal / abnormal determination results for all weighing hoppers are prepared. As an initial setting, all the storage memories for each weighing hopper are set to 1. When a combination weighing value that is determined to be abnormal for the first time after detection detection operation is detected, the memory corresponding to the weighing hopper participating at that time is 1 (first logical value) in the temporary memory, and does not participate The memory of the weighing hopper is set to 0 (second logical value). When it is determined that the combination weighing value is normal, the memory corresponding to the weighing hopper that participated at that time is set to 0, and the memory corresponding to the weighing hopper that did not participate is set to 1. A logical AND is performed for each weighing hopper between the temporary memory and the storage memory, and the result is stored in the storage memory. At the same time, it is checked whether or not a certain weighing hopper is 1 alone. It is the measurement value difference determination means that determines whether the combination measurement value is normal or abnormal.

In the example of (a), at the end of the first set, the temporary memory is set to 1 for the weighing hoppers A, B, and C, and the memory of the other weighing hoppers is set to 0. That is, at the time of abnormality determination, the logical value of the weighing hopper that participated in the combination is set to 1, and the logical value of the weighing hopper that did not participate is set to 0. The temporary memory becomes 111000, and when it is ANDed with the storage memory 111111, it becomes 111000, and 111000 is stored in the storage memory.
Next, because the abnormality is also determined in the second set, the temporary memory becomes 100110. When ANDing between this value and the storage memory, a result of 100,000 is obtained, and the abnormality of the single weighing hopper A is found, It is determined that the weighing hopper A is abnormal. In this case, the determination speed is the same as that of the conventional method.
Next, in the case of (b), the saved memory result of the first set is 111000, but the second set is normal, so the temporary memory is 0 for the weighing hopper that participated at that time, corresponding to the weighing hopper that did not participate Set the minutes to 1. That is, at the time of normal determination, the logical value of the weighing hopper that participates in the combination is set to 0, and the logical value of the weighing hopper that has not participated is set to 1. Accordingly, the temporary memory becomes 100101, and when an AND operation is performed with the storage memory, a result of 100,000 is obtained, so that an abnormality of the single weighing hopper A can be detected.
However, in order to determine that the weighing hopper is abnormal when the logical value 1 of the single weighing hopper is obtained, it is required that there is an abnormality determination at least once.

Since this judgment method is based on the assumption that only one weighing hopper of the combination weigher is abnormal at the same time, when it is judged abnormal, one of the weighing hoppers participating in the combination at that time is abnormal. At the same time, the determination that the weighing hoppers not participating are normal is performed at the same time. In the case of normal determination, it is determined that only the weighing hoppers participating in the combination at that time are normal.
Therefore, even if 1 remains in the logical value of only the specific weighing hopper due to the continuation result of only normal determination, it is considered as 1 of the undetermined result. If the logical value 1 remains in any weighing hopper as a result of only normal determination, the determination is continued. If the normality determination is continued and all the weighing hoppers become the logical value 0 as a result of the determination by the logical operation, it is determined that all the weighing hoppers are normal, and the calculation is restarted from the point where 111111 is set in the first storage memory.
In this manner, the abnormal weighing hopper can be determined earlier than the conventional method by excluding the normal weighing hopper in logic.
However, 111111 is set in the first storage memory, the weighing hopper determined to be abnormal is stored in the temporary memory as 1, and a logical AND (logical product) of the logical values of the storage memory and the temporary memory is performed. The stored logical value is stored in the storage memory, and the abnormal weighing hopper whose logical value is 1 is determined. Instead, 000000 is set in the first storage memory, and the weighing hopper determined to be abnormal is set to 1 Is stored in a temporary memory, and a logical OR (logical sum) of the logical values of the storage memory and the temporary memory is performed, and the logical value obtained thereby is stored in the storage memory. May be determined.

[1-8. (Auto span correction method)
Next, as a self-recovery method as well as diagnosis of failure and abnormality, the span of the abnormal reweighing means is corrected with the weight value of the normal combination weigher, and the weight value of the normal reweighing means and the combination weight The method of correcting the span of the abnormal weighing hopper of the combination weigher with the measured value of the normal weighing hopper can be returned to normal.
For this purpose, the span coefficient is corrected by comparing the combination measurement value of one or a plurality of combination articles with the re-measurement value of the re-measurement means. A method using a plurality of measurement values is suitable as a correction method because the variation error can be reduced as described above.
[1-8-1. (Span correction of reweighing means)
A zero point correction is performed for a specific weighing hopper, but it is determined that the abnormality of the specific weighing hopper has not been removed. That is, the span correction condition is satisfied. Next, after the zero point correction is performed once by the reweighing means (a CPU for instructing zero point correction is provided in the reweighing means. However, a CPU for instructing zero point correction and span correction may be provided in the combination scale). As a result of the measurement, the first set of reweighing values Wc1 shown in FIG. 2 can be obtained and corrected with one of the reweighing values Wc1. When the correction coefficient is j, the combination weighing value and the memory value (measurement value) of the weighing hopper constituting the combination and the corresponding re-weighing value of the re-weighing means are transmitted and read from the combination weigher.
j = (wa1 + wb1 + wc1) / Wc1 (1.8.1.1)
Ask for. The reweighing value of the reweighing means for the same combination article is Wc1, and the combination weighing value of the combination weigher is wa1 + wb1 + wc1. The combination weighing value wa1 + wb1 + wc1 of the combination weigher is normal, the reweighing value Wc1 of the reweighing means is abnormal, and the ratio is the above equation. If j <1, it means that the span of the reweighing means is larger than the normal value. If the current setting span Kc of the reweighing means is multiplied by j and Kc · j is used as a new span Kc, the reweighing means can be returned to the normal span.

In order to perform span correction with higher reliability, calculation is performed as follows. When using 10 sets of measurement values from 1 to 10 in FIG.
j1 = (wa1 + wb1 + wc1) / Wc1
j2 = (wa2 + wb2 + wd2) / Wc2
・
・
j10 = (wa10 + we10 + wf10) / Wc10
Therefore, as an average value,
j = (j1 + j2 + ... + j10) / 10 ... (1.8.1.2)
Is used. As described above, the use of many sets of data reduces the influence of the variation error, and can reduce the correction error.
In this way, the span correction means according to claim 21, wherein the current setting span Kc of the re-measuring means is multiplied by the average correction coefficient j and corrected as Kc · j as a new span Kc.
Further, by displaying this ratio on the display / operation unit, the operator can recognize the degree of the span error. Since the span is an important factor, the operator can manually instruct span correction by judging this value.
The operator can instruct span correction as a new span Kc using the correction coefficient j.

[1-8-2. Span correction of combination weigher (in case of one abnormal weighing hopper)]
It is also possible to correct the span of the specific weighing hopper of the combination weigher with the measured value of the reweighing means. The invention according to claim 18 automatically corrects the span of the specific weighing hopper. When the specific weighing hopper of the combination weigher is abnormal, if the zero point correction is instructed once by the re-weighing means and the same weighing hopper is abnormal even though the combination weigher corrects the corresponding weighing hopper to zero (as described above (When correction conditions are in place)
For example, it is assumed that the weighing hopper A has a span abnormality in FIG. Reference is made to the reweighing value Wc1 of the reweighing means and the weighing values wa1, wb1, and wc1 for each weighing hopper of the combination article in which the weighing hopper A corresponding to Wc1 participates. Since the weighing hoppers B and C of the combination weigher are normal and the re-weighing means are also normal, if the correct weighing value of the weighing hopper A is wa1c, the weighing value of these weighing hoppers is as follows: wa1c = Wc1− (wb1 + wc1) The weight value of the weighing hopper A in the normal case is obtained by calculation. The second subtraction means performs the subtraction of wa1c = Wc1− (wb1 + wc1). Here, if the correction coefficient is ia, ia is
ia = wa1c / wa1 (1.8.2.1)
It becomes. In the measurement hopper A, the normal measurement value is wa1c and the abnormal measurement value is wa1, so if ia <1, it means that the span of the measurement hopper A is large. Therefore, if the current set span coefficient of the weighing hopper A is Ta, the weighing hopper A of the combination weigher can be returned to a normal span if Ta · ia is used as the new span Ta of the weighing hopper A. This new span coefficient is sent to the weighing hopper A of the combination weigher. Thus, the span correction means according to claim 18 corrects the span as Ta of the new weighing hopper A using Ta · ia.

Since this calculation also includes a variation error in the data of one combination article, it is more appropriate to perform the calculation with a plurality of combinations. Take the 10 sets of FIG. 2 as an example.
For each pair from the first to the tenth wa1c = Wc1- (wb1 + wc1)
wa2c = Wc2- (wb2 + wd2)
・
・
wa10c = Wc10− (we10 + wf10)
Each correction coefficient ian is
ia1 = wa1c / wa1
ia2 = wa2c / wa2
・
・
ia10 = wa10c / wa10
It is. And as an average value,
ia = (ia1 + ia2 + ... + ia10) / 10 ... (1.8.2.2)
And ia is multiplied by the span Tda of the current weighing hopper A of the combination weigher in the same manner as described above to perform correction.
Further, by displaying this ratio on the display / operation unit, the operator can recognize the degree of the span error. Since the span is an important factor, the operator can manually instruct span correction by judging this value.

[1-8-3. Span correction of combination weigher (in case of two abnormal weighing hoppers)]
When there are two weighing hoppers A and B as the abnormal weighing hoppers, first, in order to correct the span of the weighing hopper A, in the example of FIGS. Select the weighing value. 5 to 10 corresponds to that. When calculating in the same way as above,
wa5c = Wc5-(wc5 + wd5) ia1 = wa1c / wa5
・ ・
・ ・
wa10c = Wc10− (we10 + wf10) ia10 = wa10c / wa10
It becomes. In addition, the weighing hopper B is a combination of combination articles that participate in the weighing hopper B and do not participate in the weighing hopper A 11
To 16
wb11c = Wc11− (wc11 + wd11) ib11 = wb11c / wb11
・ ・
・ ・
wb16c = Wc16- (we16 + wf16) ib16 = wb16c / wb16
It becomes. Therefore,
ia = (ia5 + ia6 +... + ia10) / 5
ib = (ib11 + ib12 +... ib16) / 5 (1.8.3.1)
It becomes. The span coefficients of the weighing hoppers A and B of the combination weigher are corrected by ia and ib, respectively.

[2. Signal transmission and data verification method between combination weigher and reweighing means]
[2-1. About collation of mutual data)
A weighing system comprising a combination weigher 1 and a reweighing means (for example, a weight sorter) 4 normally has a packaging machine 2 and a transport conveyor 3 between the two 1 and 4 as shown in FIG. 1 to a plurality of combination articles P between the flow of the combination article P discharged from the predetermined weighing hopper H and assembled and the combination article P measured by the re-weighing means. Such a system is generally present. FIG. 7 shows the presence of four combination articles between the combination scale 1 and the reweighing means 4.
In order to allow the mutual diagnosis / self-recovery system described above to operate normally in such a weighing system, it is sent from the combination weigher 1 through the transmission line to the reweighing means prior to the arrival of the combination article, Construction of a system capable of reliably grasping that data such as measurement values for a certain combination article stored in the memory belong to the same combination article as the measurement value obtained by the reweighing means when the combination article arrives is required. Otherwise, even if the measured values of both are compared, the calculation result is meaningless. A signal generation method in the combination weigher for realizing this, a signal reference method in the reweighing means, and a signal transmission method between them will be described with reference to FIG.

If the center of the weighing system is the combination weigher 1 with respect to the combined article processing capacity per unit time, the packaging machine 2 and the reweighing means 4 that can cope with the processing capacity of the combination weigher 1 (the combined article discharge capacity) are selected. For example, when a combination article of maximum M pieces / min is discharged from the combination balance 1 to the packaging machine 2 according to the specifications, the packaging machine 2 has the same capability, in other words, a combination article assembled at a time interval of maximum 60 / Msec. The product is packaged and sent to the transport conveyor 3, and the transport conveyor 3 sends the combined article to the reweighing means 4 at the same time interval as the combined article is discharged from the packaging machine 2.
On the way from the combination weigher 1 to the reweighing means 4, if the timing at which the combination article is discharged from these devices greatly deviates from the discharge timing of the combination weigher 1 due to the transport route of the packaging machine 2, the transport conveyor 3, etc. An abnormal narrowing or spreading of the combination article input interval to the re-weighing means 4 occurs. As a result, there may be a case where two vehicles are placed on the weighing conveyor V due to the positional relationship with the combined article conveyed forward and backward, and in this case, accurate weighing cannot be performed. Therefore, the system is constructed so that the processing / transport time does not vary as much as possible in the intermediate packaging process or transport path.
In other words, when the transport route up to the re-weighing means 4 including the packaging machine 2 is a single device, this device carries in the combination article at a time interval corresponding to the maximum capacity set in the specification of the combination weigher 1 ( It is designed so that the combined article can be sent out (output) at almost the same interval even if it is input). By doing so, the re-weighing means 4 does not generate two pieces (two combination articles are put on the weighing conveyor V at the same time), and accurate weighing is possible.
In addition, if the combination article falls off the line including the conveyor 3 and the combination article falls out, the combination weighing value in the combination scale 1 cannot be collated by the re-weighing means 4, and measures are also taken against this. It is noted that when the combination article comes off, the discharge time interval of the combination article from the combination scale 1 and the arrival time interval of the combination article to the reweighing means 4 are different.
If the process of each device in the line is operating normally, any combination article discharged from the combination scale 1 will be transported to the reweighing means 4 in substantially the same time. About each time interval with the combination goods before and behind that, when the discharge time in the combination scale 1 and the arrival time in the reweighing means 4 are compared, it becomes substantially the same length.

Based on the above idea, it is always diagnosed whether or not the weighing system is operating normally, and only correct set data is adopted in the determination calculation. If not, a function for temporarily suspending the mutual diagnosis function of the present invention is also constructed. What is necessary is that one or a plurality of combined articles in the middle of transportation is dropped from the conveyor 3 due to some inconvenience, normal packaging cannot be performed due to a trouble in the packaging machine 2, or the packaged combined articles When the conveyance interval approaches an abnormality, a weighing system that does not cause inconvenience in data processing by detecting the abnormality is constructed.

[2-2. Preparation for data transmission of combination weigher and preparation for data reception of reweighing means)
Data exchange between the combination weigher and the reweighing means for preventing inconvenience in data processing will be described with reference to FIG.
(1) The discharge time interval according to claim 24, 31 is a time interval at which a combination of articles in the combination balance is determined or a discharge time interval of the combination article from the combination balance (time interval with the previous combination article). The measurement means measures the value on the combination weigher side, and creates a set of measurement data (weighing hopper number and weighing data for each weighing hopper) (measurement data) in the combination weigher.
As for the discharge time interval of the combination article from the combination weigher, the combination is determined, and the interval at which the discharge gate of the weighing hopper or memory hopper (the hopper for temporarily storing the article with the measured value) is opened is measured. That's fine. Then, as the timing of data output from the combination weigher, this discharge time interval value is added to the data as Tdj, and is output to the transmission line together with the discharge of the combination article.
(2) The re-weighing means sends a signal A to the combination weigher at a timing convenient for reading the output data from the combination weigher. When the combination weigher reads the signal A, when the combination is completed after the signal A is read, or when the combination article is discharged from the combination weigher, the confirmation signal B is first returned to the re-weighing means, and the data corresponds to the above data. Send the set to the re-weighing means. The time delay when the process of outputting the set data after issuing the confirmation signal B is small compared to the discharge time interval of the combination article discharged from the combination scale.

(3) The re-weighing means starts the timer TI at the timing of receiving data from the combination weigher immediately after receiving the signal B. For this TI, a standard time length TiH until the combination article discharged from the combination weigher reaches the reweighing means, for example, until the weighing by the reweighing means is completed is set. There is also a method of measuring the time until the combined article arrives at the reweighing means by installing a photo sensor PH for detecting the combined article immediately before the weighing conveyor of the reweighing means. Assume that counting up to the completion time (this time is the time of arrival at the re-weighing means) is described as a timer value Ti. This timer value Ti is a conveyance time. Further, the time when the combination article discharged from the combination weigher reaches the reweighing means may mean the time when the combination article detection photosensor PH detects the combination article, for example, instead of the measurement completion time. Good.
(4) Each time the combination article is discharged (output), the combination weigher has a measurement value Tdj of the discharge time interval between the discharge timing of the previous combination article, the combination measurement value, the participation measurement hopper number, and each measurement hopper. Is sent to the re-weighing means as a set (set data). For this purpose, a timer TDJ for counting the discharge time interval of the combination article is prepared in the combination weigher. The discharge time interval value is Tdj. Since the combination weighing value can be obtained by adding the weighing value of each weighing hopper, it is not always necessary to send it to the reweighing means. If any trouble occurs in the self-diagnosis of the combination weigher itself, a failure signal is added to normal group data, or a failure signal is sent specially when a failure occurs.
(5) When the re-weighing means receives the failure signal, if it is already after reading a sufficient amount of data, the input data immediately before the failure may be used for the computation for mutual diagnosis. If a failure occurs when the amount is small, the diagnostic calculation and data are reset. Data received from the combination weigher in a normal state is stored in the memory register of the reweighing means in the order received.
(6) Two types of timers, TI and TJ, are provided in the reweighing means. The reweighing means sequentially reads and stores the data output from the combination weigher, but collates the combination article / attachment data output from the combination weigher with the combination article / measurement data weighed in the reweighing means for the same combination article. There is a need. The timer TI is provided to check whether the combination article discharged from the combination scale has reached the re-weighing means within a standard conveyance time range that is substantially in line with the normal sequence of the packaging and conveyance process.
On the other hand, the timer TJ measures the arrival time interval Tj at which the combination articles have sequentially reached the reweighing means, and maintains the discharge time interval Tdj when the combination articles discharged from the combination scale are discharged smoothly. It is provided to confirm that it has been transported through the transport line and has reached the re-weighing means.

First, the data of the same combination article is collated in the arithmetic circuit of the reweighing means by the functions of the two timers TI and TJ, and subsequent data from the next combination article from the combination weigher is obtained by the function of the timer TJ. The discharge time interval Tdj and the arrival time interval Tj of the combined article at the re-weighing means are checked, and if both timer count values Tdj and Tj are substantially the same, the combined article is smoothly combined without any trouble such as dropout or stagnation. Judge what is being transported from the scale to the reweighing means.
With respect to the timer TI, ± TE is set as a difference between the standard conveyance time TiH for the combination article discharged from the combination scale to reach the reweighing means and the allowable time difference with respect to TiH. The conveyance time until the combination article discharged from the combination scale is completely weighed in the reweighing means is substantially constant if the system is determined. First, the reweighing means requests the combination weigher a signal B at the moment when the combination weigher ejects the combination article as a data collation procedure for using the data of the combination weigher for calculation.
When the reweighing means receives the confirmation signal B indicating that the combination article has been discharged from the combination weigher, the timer TI is started. The re-weighing means monitors the value of the timer TI and the generation of the measurement completion signal. If the completion of the measurement occurs within the time of Ti-TE and Ti + TE, the timer TI is discharged from the combination weigher together with the confirmation signal B. It is considered that there is a high possibility that the combined article has been smoothly transported in the middle of the process and has reached the reweighing means, and the next confirmation operation is started. This is shown in FIG. If there is no measurement completion within the allowable time difference, the line is determined to be abnormal.
However, in order to finally determine that the operation is smooth, the following procedure is executed. Here, TE is an allowable maximum variation time value generated when the combination article output from the combination weigher is transported through the intermediate packaging machine and the transport path. For example, as shown in FIG. 8 (b), the combination article P1 that precedes the combination article P2 discharged from the combination weigher together with the confirmation signal B is delayed due to some trouble on the line, so that the combination article P2 is exactly right. It is also conceivable to arrive at the re-weighing means at substantially the same timing as the timing at which the data should arrive and generate a weighing completion signal. Therefore, in the re-weighing means, the time interval (arrival time interval) Tj of the completion of the weighing of the sequentially arrived combination articles is counted by the timer TJ. Thereby, at the timing of completion of weighing of a certain combination article, it is possible to measure the arrival time interval value Tj with the combination article that has been weighed immediately before the combination article.

Based on the above preparation, one of the combination articles from which the reweighing means generates a weighing completion signal within the allowable error ± TE of the timer TI started after receiving the confirmation signal B. The arrival time interval value Tj of the combination article that has previously generated the weighing completion signal in the reweighing means is measured by the timer TJ and compared with the discharge time interval value Tdj from the combination scale belonging to the combination article P2 obtained from the combination scale. If the combination article that caused the completion of weighing at a time within the allowable error of the timer TI is the combination article P1 instead of P2, the value Tj of the timer TJ of the reweighing means is the combination that is further forward. The arrival time interval value with the article P0 is shown. Since P1 is delayed due to some line trouble, the value of Tj in this case should be considerably larger than the discharge time interval value when P1 and P2 are discharged from the combination scale. Since data belonging to the combination article P2 is referred to as data from the combination weigher, the discharge time interval value Tdj included in the data is a discharge time interval time value with respect to the combination article P1.
Therefore, the difference between the values of Tdj and Tj shown in FIG.
This difference is a small value if the transport process is operating smoothly, and is less than the same tolerance TE as the timer TI. When the absolute value of the difference between the values of Tdj and Tj is larger than TE, the line is abnormal because the combination article P2 is not entered at the timing of the timer TI but the combination article P1 output immediately before is entered. to decide. If the difference is within TE, the combination article is P2, and the line is finally determined to be smooth. When it is determined that there is an abnormality, an alarm signal is issued, or a signal A is generated from the re-weighing means and a retry is performed. When it is determined that the operation is smooth, the re-weighing value at the time when the completion of the measurement is generated in the allowable time between the data read immediately after the confirmation signal B and the timer TI belongs to the same combination article, and the verification operation is completed.

Thereafter, when the next measurement completion signal is generated, the timer value Tj by the timer TJ and the measurement data (group data) sent immediately after the measurement data (group data) sent immediately after the confirmation signal B are displayed. The discharge time interval value Tdj included is compared. If the difference is less than the tolerance TE, it is assumed that the combined article has flowed smoothly, and the data of the combination weigher and the reweighing means are stored in the memory register as belonging to the same combined article. Further, the same operation / determination is performed every time the measurement completion signal is generated, the timer value Tj is compared with the expected reading data Tdj, and if the absolute value of the difference is less than TE, the operation of storing in pairs is continued. If the condition above the allowable value is established, the line is immediately alerted as abnormal or returns to the beginning to retry the data request.
In addition, when a double ride (when two or more combined articles are put on the weighing conveyor V of the reweighing means 4 at the same time) is detected by the function of the reweighing means itself, it is determined as abnormal as described above. Start by retrying the data request. That is, the signal A is generated. When the data collection is retried, if there is a sufficient number of data collected so far, it may be used for the evaluation calculation. If there are fewer, reset and use the data after retry.

[3-1. (Configuration of weighing system)
Next, an embodiment of a weighing system according to the present invention will be described with reference to the drawings. As shown in FIG. 9, the weighing system includes a combination weigher 1, a packaging machine 2 for packaging the combination articles weighed and discharged by the combination scale 1, and a weight selection of the combination articles packaged by the packaging machine 2. A transport conveyor 3 for transporting to the machine 4 and a weight sorter 4 for sorting the weight of the combined articles transported by the transport conveyor 3.
The combination weigher 1 includes n (for example, six) weighing hoppers H1, H2,..., Hn, and each weighing hopper is supported by load cells LC1, LC2,. Each load cell has a weighing signal line a1, a2,..., An, operational amplifiers AMPd1, AMPd2,. And the CPU 6 via the I / O circuit 5. As a result, the analog weighing signal generated by each load cell is amplified by the operational amplifier, converted into a digital weighing signal by the A / D converter, and input to the CPU 6 through the I / O circuit. A storage unit (memory element) 7 such as a ROM or a RAM is connected to the CPU 6 via a bus line (not shown).

The combination weigher 1 measures the weight of the articles with the respective weighing hoppers H1 to Hn, and the CPU 6 variously combines the weighing signals of the articles weighed with the respective weighing hoppers H1 to Hn, and is preset from these combinations. A set of total weights (a set within an allowable weight range) that is equal to or close to the target weight being set can be selected. Then, the gate of the weighing hopper in which each of the selected articles is accommodated is opened to drop each article on the collective chute S provided below. The articles dropped on the collective chute S are collected together into a combined article P and packaged in a single bag by the subsequent packaging machine 2. Each packaged combination article is transported by the transport conveyor 3 and sent to the weight sorter 4, and the weight is weighed when being transported by the weighing conveyor V provided in the weight sorter 4. The combined weight of the combined articles passing on the weighing conveyor V is measured by the load cell LCc. The analog weighing signal of the combined article is amplified by the operational amplifier AMPc, converted into a digital weighing signal by the A / D converter A / Dc, and input to the CPU 9 through the I / O circuit 8. A storage unit (memory element) 10 such as a ROM or a RAM is connected to the CPU 9 via a bus line (not shown).

The weight sorter 4 determines whether the combination weighing value of the combination article discharged from the combination balance 1 is within the allowable weight range of the target weight, and the combination article within the allowable weight range and the outside of the allowable weight range. It is provided to sort out the combination articles. By the way, when the weight sorter 4 determines that the combination weighing value is out of the allowable weight range, the number of weighing hoppers weighing the combination articles is almost always plural, so which weighing hopper is defective. It is necessary to judge. And since there is no problem in each weighing hopper and there is a problem in the weight sorter 4, it is necessary to determine which one is defective.
Therefore, the weighing value of the article obtained by weighing with the combination weigher 1 using the weighing value of the article weighed by each weighing hopper and the reweighing value data of the combination article weighed by the weight sorter 4 is used. Therefore, the reweighing value obtained by the weight sorter 4 is determined. Then, the weighing value of the article obtained by weighing with the combination weigher 1 is determined by the weighing value of the article obtained by weighing with the combination weigher 1 and the reweighing value obtained by the weight sorter 4. And it is set as one of the objectives to detect abnormality or failure of each weighing hopper or the weight sorter 4 using these determination results. One of the purposes is to return the weighing hopper or the weight sorter 4 in an abnormal state to a normal state.

This weighing system transmits various data such as the transmission and reception of signals between the combination weigher 1 and the weight sorter 4 and the weighing values of each article obtained by the combination weigher 1 to the weight sorter 4. Therefore, SIOd and SIOc, which are I / O circuits for serial communication, are connected to the CPU 6 of the combination weigher 1 and the CPU 9 of the weight sorter 4 respectively, and the SIOd and SIOc are connected to each other by a bidirectional serial communication line 11. Connected.

Next, the abnormality and failure diagnosis of the combination weigher 1 and the weight sorter 4 and the measures for returning the abnormal state will be described. Here, the number of weighing hoppers is six. As a precondition, both the combination weigher 1 and the weight sorter 4 do not cause an abnormality or failure, and either one may cause an abnormality or failure.
[3-2. Conveyance of combined articles from the combination scale 1 to the weight sorter 4 and transmission of data]
FIG. 6 is a diagram illustrating preparation for transmission of the combination data of the combination weigher 1 and preparation for reception of the combination data of the weight sorter 4. The combination weigher 1 selects combination articles whose total weight is equal to or close to the target weight (within an allowable weight range) and sequentially discharges them to the packaging machine 2 side through the collecting chute S. It can be measured by TDj (discharge time interval measuring means). The timer TDj is provided in the CPU 6 of the combination scale 1. The timer TDj is discharge time interval measuring means according to claim 24. When the combination weigher 1 discharges the combination article, the combination weighing value of the combination article, the number of the weighing hopper that weighed the articles constituting the combination article, the measurement value (weight value) of each article, and the discharge time interval Tdj (group data) is transmitted as one group to the CPU 9 of the weight sorter 4 through the serial communication line 11. At this time, the combined article is packaged by the packaging machine 2 and conveyed by the conveyor 3 onto the weighing conveyor V of the weight sorter 4. The combination article on the weighing conveyor V is weighed by the weighing conveyor when the photosensor PH is shielded from light, and a reweighing value is obtained.

[3-3. Checking the combination data generated by the combination weigher 1 and the reweighing value generated by the weight sorter 4]
Next, set data such as a weighing value of each article constituting the combination article transmitted from the combination scale 1 to the weight sorter 4, a reweighing value corresponding to the combination article generated by the weight sorter 4, Will be described. By checking the combination data and the re-weighing value to determine that both are data of the same combination article, it is possible to diagnose abnormality or failure of the weighing hopper or the weight sorter 4.
The flowchart shown in FIG. 10 shows the processing contents for storing the set data transmitted from the combination weigher 1 in the storage unit 10 of the weight sorter 4, and a program for performing this processing is stored in the storage unit 10. This process is executed every time the set data transmitted from the combination scale 1 is input to the weight sorter 4.
The data transmitted from the combination weigher includes set data and alarm data which will be described later. First, the data input to the CPU 9 by the CPU 9 of the weight sorter 4 is either set data or alarm data. (S100), and when it is determined to be set data YES, the set data is sequentially stored in the memory register R1 provided in the storage unit 10 in the input order (S102). Then, it is determined whether or not a signal B, which will be described later, has been received from the combination weigher 1, that is, whether or not the flag Fsb = 1 (S104). If Fsb = 1 and YES, the memory address is determined. The count value Cr11 of the designated storage counter CR11 is incremented by 1, and this process is terminated (S106). The signal B is a signal that is sent to the weight sorter 4 immediately before the combination weigher 1 sends the set data to the weight sorter 4.
However, if NO in step S100, it is determined whether or not the signal B has been received (S108). If YES, the flags Fsb and Fst are set to 1, respectively. Is finished (S110). And when it determines with NO instead of the signal B in step S108, it determines with it being alarm data, and memorize | stores this alarm data in the memory | storage part 10 (S112).
As described above, the weight sorter 4 can store one or a plurality of set data sequentially input immediately after the signal B is input as one group in the memory register R1, and alarm data is stored in a separate memory register. Can be remembered. Therefore, the signal B is a signal for instructing to process as a new group of set data. The alarm data is a signal indicating an abnormality or failure in the combination weigher 1, and predetermined processing is performed based on the alarm data.
If NO is determined in step S104, the weight sorter 4 is not receiving the signal B, and the set data input until the signal B is received is not used for the calculation. The count value Cr11 is fixed to 0. Furthermore, the storage unit 10 of the weight sorter 4 also includes a counter CR12 for specifying a memory address for reading the set data stored in the memory register R1.

The flowchart shown in FIG. 11A shows the content of measuring the arrival time interval (measuring completion interval) Tj of the combined article conveyed to the weight sorter 4, and the flowchart shown in FIG. It is a figure which shows the memory | storage process of the timer TJ at the time of goods reaching the weight sorter 4 and completion | finish of measurement.
As shown in FIG. 11 (a), the timer TJ provided in the weight sorter 4 is continuously counting (S200), and the weighing conveyor V for the combined articles conveyed to the weight sorter 4 By measuring the time intervals of the measurement completion signals sequentially generated at this time, the arrival time interval Tj when the combination article reaches the weight sorter 4 is measured. Then, as shown in FIG. 11B, the CPU 9 of the weight sorter 4 stores the measured arrival time interval Tj in the memory register R2 (S220), and resets the timer TJ to 0 (S222). Note that the register R2 stores the latest one arrival time interval Tj.

The flowchart shown in FIG. 12 is a diagram showing a head data designation process for the combination weigher 1 when the weight sorter 4 transmits the signal A shown in FIG. The signal A is a signal that requests the combination weigher 1 to transmit the top data of the set data and subsequent data by sending it to the combination weigher 1 at a convenient timing. Signal for data recognition). It is a standby timer that measures the convenient timing of the weight sorter 4. The function of the standby timer is that when the weight sorter 4 generates an alarm signal and instructs the combination weigher 1 or the weight sorter 4 itself to perform zero point correction or span correction, the combination weigher 1 or the weight sorter 4 Until the correction operation is completely completed, the function of waiting for the collection of the next set data (abnormality / failure evaluation judgment data of the combination weigher 1 and the weight sorter 4) by the weight sorter 4 is achieved.

First, it is determined whether the CPU 9 of the weight sorter 4 is generating an alarm signal (S300). This alarm signal is a signal for warning an abnormality of the combination weigher 1 or the weight sorter 4 itself. When it is determined NO without generating an alarm signal, it is determined whether or not the signal B from the combination weigher 1 has been received, that is, whether or not the flag Fsb = 1 (S302). . If the signal B is not received and the flag Fsb = 0 and it is determined NO, it is determined whether the determination calculation is permitted, that is, whether the determination calculation permission flag Fj = 1. (S304). When the determination calculation is not permitted, that is, when the determination calculation permission flag Fj = 0 and it is determined NO, it is determined whether or not the count value Tw = 0 of the standby timer TW (S306). Then, when the count value Tw = 0 of the standby timer TW is determined to be YES, the signal A is transmitted to the combination weigher 1 and the process is terminated (S308).
However, when YES is determined in steps S300, S302, and S304, and when NO is determined in S306, the process returns to the start. As described above, it is determined that the alarm signal is not generated in step S300 and NO, the signal B is not received and the set data is not received in step S302, and the determination is made in step S304. The signal A can be transmitted to the combination weigher 1 on condition that the calculation is not permitted and that the count value Tw of the standby timer TW is 0 in step S306.
Upon receipt of this signal A, the combination weigher 1 transmits the signal B to the weight sorter 4 at a predetermined timing after receiving this signal A, and immediately after that, sets data relating to the combination article to the weight sorter 4. Send (not shown).

FIG. 13 is a diagram illustrating a flowchart for processing the count value Tw of the standby timer TW. When the weight sorter 4 issues a zero correction or span correction command, predetermined data Tw is set in the standby timer TW. The timer data Tw is counted down until it reaches 0 (S400, S402). When Tw = 0, the signal A is transmitted to the combination weigher 1 in step S306.
FIG. 14 is a diagram illustrating a flowchart for processing the count value Ti of the transport time measurement timer TI. As shown in step S108 of FIG. 10, the transport time measurement timer TI sets the flag Fst to 1 when the weight sorter 4 receives the signal B (S500), and starts counting the count value Ti. (S502). On the other hand, when the set data is transmitted from the combination scale 1 immediately after the weight sorter 4 receives the signal B, the set data is stored in the register R1 in step S102 shown in FIG. 10, and the count value Cr11 of the counter CR11 is stored. Increases by one.
Thereafter, the weight sorter 4 continuously reads and stores the set data transmitted from the combination weigher 1 while checking the count value Ti, and each time the set data is stored, the count value Cr11 is only 1. Continue to increase. At this time, the arrival time interval Tj of the combination article (a time interval at which the reweighing value of the combination article is completed) Tj is measured by the arrival time interval measurement timer TJ provided in the weight sorter 4, and this arrival time is measured. The interval Tj is stored in the memory register R2, the data stored in this R2 is updated, and each time the timer TJ is reset, the next count is started.

FIG. 15 is a flowchart showing a process for collating the set data received from the combination weigher 1 with the data obtained by the weight sorter 4. This flowchart is activated every time the re-weighing value of the combination article is completed and a measurement completion signal is generated (every time the combination article reaches the weight sorter 4 and generates an arrival signal), The conveyance time Ti of the article is checked. First, when the measurement completion signal is generated, it is determined whether or not the verification of the first set data has been completed (S600) and whether or not the signal B has been received (S602). Then, when the comparison of the first set data has not been completed and the flag Fj = 1 is determined as NO and the signal B has already been received and the flag Fsb = 1 is determined as YES, Ti _H -TE It is determined whether or not <Ti <Ti _H + TE (S604). Note that Ti is the transport time of the leading combination article, Ti _H is the standard transport time, and TE is the allowable time.
When the transport time Ti of the top combination article is within the allowable time and it is determined to be YES, the combination article may be the top combination article discharged immediately after receiving the signal B from the combination scale 1. Judgment is high. However, if it is determined in step S604 that the transport time Ti is outside the allowable time and NO, it is determined that the leading combination article has not been transported normally.
However, even if it is determined in step S604 to be YES and it is highly likely that the combined article that has reached the weight sorter 4 is the first one, the first combined article has dropped from the conveyor 3 due to some inconvenience. However, there is also a possibility that the combination article of the previous group being transported one time before arrives at the timing when the first combination article arrives at the weight sorter 4 and is weighed.

Therefore, first, the flag F _st is reset to 0 (S606), and the signal is transmitted from the combination weigher 1 simultaneously with the signal B in the memory register R1 provided in the weight sorter 4 by the count value Cr12 of the TDJ counter CR12. The set data thus read is designated, and the discharge time interval Tdj (discharge time interval with the previous combination article) included in the set data is read (S608). After reading this data, the count value Cr12 of the counter CR12 is incremented by 1 (S610). Next, the read discharge time interval Tdj is compared with the arrival time interval Tj stored in the counter TJ memory register R2 (S612). That is, it is determined whether or not Tdj−TE <Tj <Tdj + TE (time interval difference determination means according to claim 24).
Here, when it is determined YES in step S612, it cannot be said that the combination article discharged from the combination weigher 1 before the B signal is generated has arrived at the weight sorter 4 with its conveyance delayed. It is determined that the combination article re-weighed by the weight sorter 4 this time is the first combination article discharged from the combination scale 1 immediately after the signal B is generated. Then, the data collation flag Ff is set to 1 (S614), and thereafter, after completion of weighing, each subsequent set data is sequentially transferred from R2 to R3 of the memory register as a process normal process, and necessary combination weighing values are stored. Then, the process ends (S616).
However, if it is determined in step S602 that the signal B has not been received and the determination is NO, the process ends until the signal B is received.
Then, in step S604, the time the result of determination is NO, it is judged whether or not _{Ti ≧ Ti H + TE (S618} ). Ti <When it is determined that Ti _H are -TE NO, the process ends in step S618. If it is determined as YES in step S618, it is determined that an abnormality such as that the combined article has dropped from the conveyor 3 has occurred, and an alarm signal is output (S620), and the flags F _sb , F _st , F _f Are set to 0 respectively and the process ends (S622).
Also, when NO is determined in step S612, it is determined that the conveyance state is abnormal, and the processes of steps S620 and S622 are performed.

In step S600, when the collation of the first set data is completed and it is determined that Fj = 0 and YES, the combination article discharged from the combination scale 1 after the first combination article and the first combination The comparison with the combination article that has arrived at the weight sorter 4 later than the article is performed by comparing the discharge time interval Tdj of the combination article discharged from the combination scale 1 and the arrival time interval Tj (S624, S626). , S628). Note that steps S624, S626, and S628 are the same as the processes of steps S608, S610, and S612, and thus description thereof is omitted.
However, when it is determined NO in step S628 and it is determined that the conveyance state is abnormal, the processes of steps S620 and S622 are performed. When it is determined YES and it is determined that the conveyance state is normal, the process of step S616 is performed.

FIG. 16 and FIG. 17 are processes performed when the reweighing of the combination article by the weight sorter 4 is completed and it is determined that the conveyance (process) of the combination article is normal. Is moved to the register 3 and stored.
First, simultaneously with the reading of the discharge time interval Tdj in step S608, the combination participation weighing hopper data is also read from the register R1 (S700), and it is determined whether or not the number of participation weighing hoppers is a standard number, for example, 3 (= 6 units / 2). Determination is made (S702). When the number of participating weighing hoppers is determined to be NO instead of the standard number, the weight of the articles filled in the participating weighing hopper is too different compared to the case of the standard number, so the magnitude of the span error that appears. affect. In such a case, an error may occur in the span correction value. If it is not the standard number, the read participation weighing hopper data is deleted together with the weight value of the weight sorter 4 at this time, and the process is ended (S710). . As a safer method, although not shown in the flowchart, it is determined whether or not the weight values of the combination participating weighing hoppers are all within a certain weight value, and each weighing hopper has a weight value that is within an allowable range, Later, the span error can be evaluated to prevent variations in the span error.
However, if it is determined in step S700 that the number of participating weighing hoppers is equal to the standard number (3) and YES, the participating weighing hopper data is adopted for error calculation, so that it is paired with the reweighing value of the weight sorter 4 and the memory register. Store in R3 (S704). Then, 1 is added to the count value Cr21 of the memory address designating counter CR21 for storing data in the register R3 (S706).
In order to check the number of weighing hoppers participating in the combination in all the collected data, combination participation number counters Ca, Cb,... Are provided for each weighing hopper in advance. Each time, the weighing hoppers participating in the combination in the data stored in the register R3 are checked, and the corresponding counter is sequentially incremented by 1 according to the number of the weighing hoppers (S708).

Next, it is determined whether or not the participation weighing hopper number counter provided for each weighing hopper has reached the set number Mmin or more (S800). If the set number has not been reached and the determination is NO, the process is once ended and data input due to repeated weighing is awaited.
A bi-directional serial or parallel communication line 11 is installed between the combination weigher 1 and the weight sorter 4, and the weighing data and the like are sufficiently fast compared to the maximum processing capacity and line conveyance capacity of the combination weigher 1. It can be transmitted with. In addition to the signals shown in FIG. 6, the combination weigher 1 and the weight sorter 4 can exchange a failure signal, a zero correction instruction, signals A and B, a span correction instruction, and a span correction value through the communication line. .

In step S800, when each of the participating weighing hopper counters has reached the set number Mmin or more and it is determined to be YES, all data are referred to as the data collection is completed, and the other weighing hoppers are classified by the designated weighing hopper. The number of participants is counted (S802). To do this, first take up all the data of the set in which the weighing hopper A participates, count the number of participation of the weighing hopper B, the weighing hopper C,... It is determined whether or not the number of participants is included (S804).
A combination of structures in which a memory hopper is provided in addition to the weighing hopper, and the weighing value of the weighed article accommodated in the memory hopper and the weighing value of the article accommodated in the weighing hopper are also included in the combination. In the scale, there may be two or more items weighed by the same weighing hopper in one combination. Therefore, in the above count value, there is a count value equal to or greater than the count value of the specified weighing hopper. It is possible that
If there is even one such case, the judgment operation of the present invention does not hold, so the number of participation weighing hopper data collected is increased and the number of participations equal to or more than the designated weighing hopper is not included in the probability. Can be. Therefore, the value of Mmin is increased by 1 (S806). Increasing the value of Mmin gradually increases the possibility that it will be established, but if it is possible to combine only with the same weighing hopper due to some trouble, it will not be established easily. and if it is determined, the first data verification completion flag Ff to generate an alarm signal, the process ends the signal B received flag F _SB is reset to 0 (S810, S812, S814) . However, if it is determined as NO in step S808, the process ends.

Although not shown in FIG. 17, when it is determined in step S800 that each participating weighing hopper number counter is less than the set number Mmin and NO, the process ends. However, if a specific weighing hopper does not always participate in the combination due to some inconvenience, the program cannot proceed further, so if it is determined NO in the step S800 for a certain number of times, an alarm signal is generated. Processing may be performed.
However, even if the combination participation number of each weighing hopper is the same in all collected data, for example, when all the combinations in which the weighing hopper A participates are picked up, any one of the other weighing hoppers in the group in which the weighing hopper A participates. If the above number of data participates more than the number of data of the weighing hopper A, the error amount determination calculation does not hold.
Therefore, in the combination data once collected, in order from the weighing hopper A, the number of other weighing hoppers participating in the group of all the combination data in which the weighing hopper A participates is checked. Make sure that it is less. This operation is performed sequentially for all the weighing hoppers A to F.
If there is an occurrence of the same number of other weighing hoppers participating in a specific combination data group of a specified weighing hopper, the entire collected data is discarded and a new combination weigher 1 is added. Issue A signal and retry data collection. Such a phenomenon occurs when at least two weighing hoppers always participate in pairs in all combinations. However, if the number of data is increased to some extent, it is considered that such a phenomenon hardly occurs. The greater the number, the lower the probability of occurrence.
This check calculation is performed in step S804 by determining whether or not the number of designated weighing hoppers is the maximum in each designated weighing hopper group.

Another possible method is to continue collecting data until all kinds of different combinations of data can be collected. For example, when 10 weighing hoppers and 4 standard combinations are used, 10C4 = 210 combinations are used. In addition, since it is considered that many combinations are necessary until 210 patterns are arranged, it takes a long time when the processing capacity of the combination weigher 1 is slow. Therefore, this method may be applied when the number of weighing hoppers of the combination weigher 1 is small or when the processing capacity is high. In this case, it is not necessary to determine the number of other weighing hoppers participating in the combination group in which the designated weighing hopper participates. If it is determined in step S804 that conditions for performing the determination calculation are complete in the collected data and YES is determined, the value of Mmin is set to a constant value, the determination calculation permission flag Fj is set to 1, and the next error amount is set. Is started (S816, S818), and the processes of steps S812 and S814 are performed.

[3-4. Failure / abnormality judgment and automatic correction operation of combination scale 1 and weight sorter 4)
When the determination calculation permission flag F _j is set to 1, the program shown in the flowcharts of FIGS. 18 and 19 is started. In this program, Awynx is calculated by the equation (1.4.2.1) and Awyx is calculated by the equation (1.4.1.1) (S900). Awynx is an average value of the difference between the weighing values of the group in which the weighing hopper x is not selected as the combination. Awyx is the average value of the weighing value differences of the group in which the weighing hopper x is selected as the combination. This measurement value difference is a value obtained by subtracting the combination measurement value from the re-measurement value.
First, a failure and abnormality determination of the weight sorter 4 is performed by Awynx (S902). It is assumed that the abnormal boundary value Rec and the failure boundary value Rtc of the weight sorter 4 and the abnormal boundary value Red and the failure boundary value Rtd of the combination weigher 1 are already set. The abnormal boundary value Rec is a value obtained by calculation using the equation (1.4.2.2), and the failure boundary value Rtc is a value larger than Rec. The abnormal boundary value Red is a value obtained by calculation using the equation (1.4.1.2), and the failure boundary value Rtd is a value larger than Red.
Here, the number of weighing hoppers of the combination weigher 1 is 6, the minimum number of participation Mmin of each weighing hopper is 10, the standard deviation of the variation of each weighing hopper of the combination weigher 1 is 0.2 g, and the standard deviation of the weight sorter 4 is Assuming 0.3, σt = 0.147 g and therefore 3σt = 0.441 g are used according to the equation (1.3.3). If Mmin is made larger, a smaller 3σt value can be set. In addition, since the weighing hopper expected to be normal also has a slight deviation error, 3σt is expected as a margin value for determining the abnormality, and rc = rd = 3σt = 0.441 g (1.4.1.2) From the formula (1.4.2.2), Rec = Red = 0.882 g.
On the other hand, the value for failure determination is set to Rtc = Rtd = 3Rec = 3Red = 2.646 g, for example, assuming three times the abnormality determination boundary value. Instead of 3 sigma, n sigma (n is an arbitrary real number) may be applied.

Depending on the above set value, the failure / abnormality determination of the weight sorter 4 can be performed using the formulas (1.4.2.3), (1.4.2.4), and (1.4.2.5) (the bias error of the weight sorter 4). (S904, S906, etc.).
If it is determined that the weight sorter 4 is abnormal rather than malfunctioning, it is determined whether or not the zero correction flag F _CZ of the weight sorter 4 is 1 (S908). When the zero point correction flag F _CZ = 0 and it is determined as NO, the zero point correction is not performed. In order to correct the zero point of the weight sorter 4, a signal for temporarily stopping the discharge of the combination article is transmitted to the combination weigher 1 (S910). Then, the zero point of the weight sorter 4 is corrected in a state where the combination article is not sent to the weight sorter 4, and the zero point correction flag F _CZ = 1 is set (S912). Since the data immediately after the zero point or span correction command may not have been corrected by the weight sorter 4 or the combination weigher 1, the standby timer value is set in the standby timer until it reaches zero. Do not start data collection during this time. That is, the timing for sending the signal A from the weight sorter 4 is delayed by a certain time. Further, the determination calculation permission flag Fj = 0 is reset and the process ends (S914).

However, if it is determined in step S908 that the zero correction flag F _CZ = 1 and YES, it is determined whether or not the span correction flag FCS = 1 of the weight sorter 4 is determined (S916). That is, zero correction is performed on the weight sorter 4, but when it is determined that there is still an abnormality, span correction is performed, but it is determined whether or not the span correction has already been performed. Therefore, when it is determined that the span correction flag FCS = 0 and NO, since span correction is not performed, span correction calculation is performed and span correction is performed (S918). Then, the span correction flag F _CS = 1 is set, the zero point correction flag FCZ = 0 is reset (S920), and the process proceeds to step S914.
In step S916, when it is determined that the span correction flag F _CS = 1 of the weight sorter 4 and YES, zero point correction and span correction have been performed, but it is still determined that there is an abnormality. A signal is generated and the span correction flag F _CS = 0 is reset (S922, S924), and the process proceeds to step S914. When the alarm signal is generated, the operator checks and repairs the weight sorter 4.
If it is determined in step S904 that the weight sorter 4 has failed and YES, an alarm signal is generated (S926), and the flags F _j = 0, F _CZ = 0, F _CS = 0, F _dZ = 0. , F _dS = 0 (S928). The flag F _dZ is a zero correction flag for the combination weigher 1, and the flag F _dS is a span correction flag for the combination weigher 1.

If it is determined in step S906 that the weight sorter 4 is normal and NO, it is determined that the weight sorter 4 is normal and the span correction flag FCS = 0 of the weight sorter 4 is reset (S930). Then, it is determined whether or not each weighing hopper of the heavy combination weigher 1 is abnormal or malfunctioning (S932, S934).
As described above, when it is determined that the weight sorter 4 is not abnormal, a failure / abnormality of each weighing hopper of the combination weigher 1 is then determined by Awyx (S932, S934). For the failure / abnormality judgment of the combination weigher 1, the above boundary values Rtd and Red are used (1.4.1.3), (1.4.1.4) and (1.4.1.5) (combination weigher 1 The method / procedure described in (Method of judging deviation error of each weighing hopper)
If YES is determined in step S934, it is determined that any of the weighing hoppers provided in the combination weigher 1 is abnormal, and first, it is determined whether or not the zero correction flag F _dZ = 1 of the combination weigher 1 is set. Determination is made (S942). When the zero point correction flag F _dZ = 0 and NO is determined, the zero point correction command signal of the specific weighing hopper determined to be abnormal is output (S944), and the zero point correction flag F _dZ of the specific weighing hopper is set to 1 (S946). This makes it possible to correct the zero point of the specific weighing hopper having an abnormality. Then, the standby timer value is set in the standby timer (S948), the determination calculation permission flag F _j = 0 is reset, and the process ends (S950).

However, if it is determined in step S942 that the zero point correction flag F _dZ = 1 and YES, it is determined whether or not the span correction flag F _dS = 1 of the combination weigher 1 (S952). In other words, zero correction is performed on the specific weighing hopper, but when it is determined that there is still an abnormality, span correction is performed, but it is determined whether or not the span correction has already been performed. Therefore, when it is determined that the span correction flag F _dS = 0 and NO, the span correction is not performed, so the span correction calculation is performed and the span correction data and the span correction command signal are transmitted to the combination weigher (S954). . Then, the span correction flag F _dS = 1 of the specific weighing hopper is set, the zero point correction flag F _dZ = 0 of the specific weighing hopper is reset (S956), and the process proceeds to step S948.
In step S952, when it is determined that the span correction flag F _dS = 1 of the specific weighing hopper is YES and zero correction and span correction have been performed, it is determined that the abnormality is still abnormal. And the span correction flag F _dS of the specific weighing hopper is reset to 0 (S960, S962), and the process proceeds to step S948. When the alarm signal is generated, the operator checks and repairs the combination scale 1.

However, if it is determined in step S932 that the combination weigher 1 is broken and YES, an alarm signal is generated, the determination calculation permission flag F _j = 0 is reset, and the process ends (S936, S938). Since the alarm signal is generated, the operator checks and repairs the combination scale 1.
If NO is determined in step S934, it is determined that both the weight sorter 4 and the combination weigher 1 are normal, and the span correction flag F _dS of the combination weigher 1 is reset to 0 (S940). Then, the standby timer value is set in the standby timer (S942), the judgment calculation permission flag F _j = 0 is reset, and the process ends (S944).
In this way, after the zero point correction or the span correction of the weight sorter 4 or the combination weigher 1 is completed and the standby timer value has elapsed, the weight sorter 4 outputs the signal A to the combination weigher 1 and the first set data Collation and collection of set data.

[3-5. (Data display and correction operation)
As described above, according to the weighing system including the combination weigher 1 and the weight sorter 4 of this embodiment, the error amount of the other is quantitatively calculated on the assumption of one of the normal operations, and special human work is performed. Without adding, it is possible to automatically correct the zero point and span of the weighing hopper of the combination weigher 1 and the weight sorter 4.
Further, regarding the conveyance state of the combination article in the process of the combination weigher 1 and the weight sorter 4, the conveyance time Ti until the combination article discharged from the combination scale 1 reaches the weight sorter 4 and the standard conveyance time Ti _H Normality / abnormality of the transport state is quantitatively determined by comparing the discharge time interval Tdj from which the combination article is discharged from the combination weigher 1 and the time interval during weighing in the weight sorter 4, that is, the arrival time interval Tj. can do.
Accordingly, the status of the present measurement system expressed by these numerical values, for example, Awyx, Awynx, Ti, Tdj, Tj, etc., is sent to a remote place through a local area network or the Internet, and is monitored and displayed by a display operation unit installed there. The determination can be made by the user or the operator of the service company, and the zero point correction and the span correction can be performed not only automatically but also by a remote manual operation by the person determining the numerical value.
Of course, the display operation units 12 and 13 may be installed in the vicinity of the combination scale 1 or the weight sorter 4.

Next, the operation display unit 13 provided in the weight sorter 4 of the weighing system shown in FIG. 19 will be described with reference to FIG. FIG. 20 shows a constant weight filling / packaging / sorting system for articles comprising the combination weigher 1, the packaging machine 2, and the weight sorter 4, that is, the weighing system of this embodiment. Have been exchanged through.
In this weighing system, all error information and conveyance information are collected in the weight sorter 4. Therefore, a web server WS is installed in the weight sorter 4 and connected to an external communication network NW, and a controller such as a personal computer (operation display unit) 13 and a web browser WP are installed in the remote place. Thus, it is possible to access information collected in the weight sorter 4 by the operation 13 and to operate the system by information determination.

In this weighing system, first, the combination article discharge time interval Tdj from the combination scale 1 after the transmission of the signal B, and the combination article discharged from the combination scale 1 simultaneously with the signal B are conveyed to the weight sorter 4 and reach. The display operation unit 13 may display the transport time Ti and the time series data and the time series data for the combination article discharged from the combination scale 1 after the transmission of the signal B to reach the weight sorter 4. Therefore, by monitoring this display, the operator can check the normality / abnormality of the conveyance state of the combined article with the eyes of the operator.
On the screen of the display operation unit of the personal computer 13, the combination article designated at the head from the weight sorter 4 side, the discharge timing of the combination article discharged from the combination scale 1 after that, and the weight of the same combination article as those The weighing completion timing in the sorter 4, that is, the arrival timing of the combination article in the weight sorter 4 is displayed as an image arranged vertically so as to correspond to the same combination article in the form of pulses, for example, with time as the horizontal axis. Can do. As a result, the operator can easily grasp the understanding and monitoring of normality / abnormality of the conveyance state visually.
If the combination weighing value of the combination article conveyed from the combination weigher 1 to the weight sorter 4 and the data of the reweighing value of the combination article by the weight sorter 4 are accessed, the combination weighing value related to the same combination article A pair of reweighing values can be displayed on the display / operation unit of the personal computer 13, and the weighing state of the combination weigher 1 and the weight sorter 4 can be monitored by viewing this display.
Further, the error amount between the combination weigher 1 and the weight sorter 4, for example, (1.2.1), (1.4.1.1), (1.4.2.1), (1.5.1.1), (1.5.2.6), (1.5.2.9 ), (1.6.2.1), etc. can be displayed on the display / operation unit, so that the operator can quantitatively know the magnitude of the error of the combination weigher 1 or the weight sorter 4 from the displayed value. Can do.
The display operation unit 13 is operated to manually perform zero point correction and span correction operation, and transmission of set data from the combination scale 1 to the weight sorter 4 is started, that is, the weight sorter. The command to transmit the signal A from 4 to the combination scale 1 can also be performed manually. As a result, the operator can manually issue all commands from the sampling of the set data for correction to the correction operation when necessary, by operating the display operation unit of the personal computer 13.

Further, for example, since the error of the measured value calculated by the equation (1.2.1) or the like can be displayed on the display / operating unit 13, the operator himself can quantitatively determine the normality / abnormality of the combination weigher 1 or the weight sorter 4. If necessary, the zero correction of each weighing hopper of the combination weigher 1 or the weight sorter 4 can be performed manually by an operator. Further, when the above-mentioned various data collection and determination are instructed and the error cannot be corrected even by the zero point correction, the combination weigher 1 and the weight sorter 4 can be remotely controlled, for example, by the operator's judgment and instruction. It is possible to manually perform the span correction by manipulating the display operation unit 13 installed in
As a feature of span correction of this weighing system, the weighing hopper or weight sorter 4 having an abnormal span can be corrected using the weighing data of the other weighing hopper or weight sorter 4 having a normal span. As a result, it is possible to operate from the display operation unit installed near or remotely from the combination weigher 4 or the weight sorter 4 without performing any special field work such as adding a weight or providing a special weight adding mechanism. By operation, the span of an abnormal scale can be corrected.
However, in the above embodiment, all error information and conveyance information are collected in the weight sorter 4, but instead, they are collected in the combination weigher 1 so that the combination weigher performs the various processes described above. Good.

In the above embodiment, as shown in FIG. 20, a web server WS is installed in the weight sorter 4 and connected to an external communication network NW, and a controller such as a personal computer (operation display unit) 13 and the like are installed in a remote place. A web browser WP that is connected to the communication network NW is installed therein, and the configuration is such that the information collected in the weight sorter 4 by the operation of the personal computer 13 can be accessed and the system can be operated by information determination. Instead, a WWW server is installed in the weight sorter 4 or the combination scale 1 and connected to an external communication network, and a controller such as a personal computer (operation display unit) 13 is connected to a remote place and a communication network is connected therein. WWW browser is installed, and is collected in the weight sorter 4 or the combination scale 1 by operating the personal computer 13. And access The information may be configured to allow operation of the system according to the information determined.

1 Combination Weigher 2 Packing Machine 3 Conveyor 4 Weight Sorter 6, 9 CPU
12, 13 Operation display section V Weighing conveyor

Claims (6)

A combination having a plurality of measuring instruments, various combinations of the weights of the articles weighed by these measuring instruments, and discharging a combination article composed of articles constituting the total weight within the allowable weight range among these combinations A scale,
Reweighing means for reweighing each time the combination is discharged from the combination weigher;
A calculation means for calculating a subtraction value that is a difference between a total weight of the discharged combination product and a reweighing value in the reweighing means of the discharged combination article;
A calculation value that is a total value or an average value of the subtraction values corresponding to non-participating combination articles obtained by an article weighed by one measuring instrument is calculated for each different combination article. An arithmetic means for calculating the standard deviation of the total value or the average value;
First determination means for determining that the variation of the measurement value of the re-weighing means is normal when at least one of all the standard deviations is smaller than a predetermined boundary value;
Equipped weighing system.   2. The weighing system according to claim 1, wherein when all standard deviations are equal to or larger than the boundary value, the first determining unit determines that the variation of all the measuring devices or the re-measuring units is abnormal. .   2. The weighing system according to claim 1, wherein when the first determination unit determines that the variation of the re-measurement unit is normal, one of the standard deviations is smaller than the boundary value, and the other standard A measurement system comprising: a second determination unit configured to determine that the measuring instrument that does not participate in the creation of the standard deviation smaller than the boundary value has a variation abnormality when a deviation is greater than or equal to the boundary value. A combination having a plurality of measuring instruments, various combinations of the weights of the articles weighed by these measuring instruments, and discharging a combination article composed of articles constituting the total weight within the allowable weight range among these combinations A scale,
Reweighing means for reweighing each time the combination is discharged from the combination weigher;
A calculation means for calculating a subtraction value that is a difference between a total weight of the discharged combination product and a reweighing value in the reweighing means of the discharged combination article;
A calculation value that is a total value or an average value of the subtraction values corresponding to non-participating combination articles obtained by an article weighed by one measuring instrument is calculated for each different combination article. An arithmetic means for calculating the standard deviation of the total value or the average value;
Display means for displaying each standard deviation and a standard value for each standard deviation,
Equipped weighing system. A combination having a plurality of measuring instruments, various combinations of the weights of the articles weighed by these measuring instruments, and discharging a combination article composed of articles constituting the total weight within the allowable weight range among these combinations A scale,
Reweighing means for reweighing each time the combination is discharged from the combination weigher;
A calculation means for calculating a subtraction value that is a difference between a total weight of the discharged combination product and a reweighing value in the reweighing means of the discharged combination article;
A calculated value that is a total value or an average value of the subtracted values corresponding to a non-participating combination article in which an article weighed by a predetermined number of measuring instruments is calculated for each different combination, and the total An arithmetic means for calculating a standard deviation of a value or an average value;
A first determining means for determining that the variation of the measured values of the re-weighing means or all the measuring instruments is abnormal when all of the standard deviations are abnormal in a predetermined boundary value;
Equipped weighing system.   6. The weighing system according to claim 5, wherein when one of the standard deviations is smaller than the boundary value, the first determination means applies to any or all of the predetermined number of weighing devices corresponding to the standard deviation. A weighing system that determines that the variation in measurement values is abnormal.

Images