JP6305781B2 - Combination scale - Google Patents

Description

  The present invention relates to a combination weigher that puts a weighed article into a packaging machine or the like.

  FIG. 14 is a diagram showing a schematic configuration of a conventional example of a combination weigher. Articles to be weighed from the supply device 2 at the upper center, for example, articles such as confectionery, are supplied onto the conical top cone 3a of the main feeder (dispersing feeder) 3, and the top cone 3a is vibrated by the vibration of the top cone 3a. A plurality of linear feeders (straight forward feeders) 4 arranged on the periphery at equal intervals are fed onto linear feeder troughs (straight forward trays) 4a. A plurality of supply hoppers 5 and a plurality of weighing hoppers 6 are provided at the conveyance end of the plurality of linear feeder troughs 4a for conveying the articles outward by vibration. Articles conveyed by the vibration of each linear feeder trough 4a are put into each supply hopper 5, respectively. In the plurality of supply hoppers 5, the articles are temporarily held, and then the discharge gate 5 a is opened to be supplied to each weighing hopper 6.

  The weight of the articles supplied to each weighing hopper 6 is weighed by a plurality of weight sensors 11 respectively connected to each weighing hopper 6, and a combination calculation is performed based on these measured values by a control unit (not shown). .

  In this combination calculation, whether the weighing values of the weighing hoppers 6 are variously combined and the combined weight, which is the combined weight, is equal to or more than the target combination weight Wt and the allowable upper limit value Wu, and matches the target combination weight Wt. The closest combination of the weighing hoppers 6 is selected as the optimum combination. The discharge gate 6a of the weighing hopper 6 selected by the combination calculation is opened, and the articles are put into a packaging machine (not shown) through the collecting chute 7 and the discharge chute 8 and packaged as a combination product.

  The weighing hopper 6 that has been selected as the optimum combination by the combination calculation and has been emptied by discharging the articles is supplied with articles from the supply hopper 5 thereabove, and when the supply hopper 5 becomes empty, the upper portion of the supply hopper 5 The linear feeder trough 4 a is activated, and the articles on the linear feeder trough 4 a are supplied to the supply hopper 5.

  Assuming that the target combination weight in the combination calculation is Wt as described above and the target combination number of the weighing hoppers 6 selected in the combination calculation is M, the weight of the articles supplied to the weighing hopper 6 is the target supply weight Wt / M is preferred.

When the total number of the weighing hoppers 6 participating in the combination, that is, the total number of the weighing hoppers 6 of the combination weigher is S, the target combination number M is M = S / 2 when S is an even number, and when S is an odd number. Is set to M = (S−1) / 2 or (S + 1) / 2, the number of combinations _S C _M for selecting M units from S units is maximized. As a result, when selecting the optimum combination in the combination calculation, the number of combinations to be selected becomes the largest, so that the selection probability of the combination close to the target combination weight Wt is increased, and the yield is improved. Therefore, it is preferable that the supply weight of the articles supplied to each weighing hopper 6 is a target supply weight Wt / M obtained by dividing the target combination weight Wt by the target combination number M.

  Since articles are supplied from the supply hopper 5 to the weighing hopper 6, if the weight of the articles supplied to the supply hopper 5 by the linear feeder 4 is controlled to be the target supply weight Wt / M, it is inevitably necessary. Articles having the same weight are also supplied to the weighing hopper 6.

  For this reason, the conveyance force of the linear feeder 4 which conveys articles | goods, specifically the linear feeder 4 so that the weight of the articles | goods supplied to the supply hopper 5 by the linear feeder 4 may become target supply weight Wt / M. The vibration intensity and driving time are adjusted.

  In Patent Document 1 described later, each weight hopper 6 (combination hopper) is supplied with each article so that the weight of the article is a target supply weight = Wt / M (a value obtained by dividing the set weight by the number of combinations). The conveyance force of the feeder 4 corresponding to the weighing hopper 6 is set, and for each weighing hopper 6, a predetermined number of supply weights of articles is set as the number of samples, and the supply weight of articles for the number of samples is added and added. The average value of the supply weight of the article is calculated by dividing the supply weight by the number of samples, and the next conveying force by the feeder 4 so that the calculated average value of the supply weight is close to the target supply weight Wt / M. Techniques for controlling are disclosed.

JP 59-622 A

  Even if the linear feeder 4 is given the same operation amount to vibrate so that the weight of the article supplied to the weighing hopper 6 becomes the target supply weight Wt / M, it is controlled so as to have the same conveying force. However, the weight of the articles supplied to the supply hopper 5 and thus the weight of the articles supplied to the weighing hopper 6 varies depending on the distribution method of the articles on the linear feeder trough 4a and the situation of the behavior due to the properties of the articles. . This variation is a random variation that cannot be predicted.

  In the above-mentioned Patent Document 1, the number of samples of the supply weight of the articles supplied to the weighing hopper is added, and the added supply weight of the articles is divided by the number of samples to obtain the articles supplied to the weighing hopper 6. The average value of the supplied weight is calculated, but this number of samples is not particularly specified. In general, the number of samples is set to a relatively large number so as to suppress the influence by smoothing the unpredictable random variation, and the average value of the supply weight of the article is calculated with the set number of samples. Is done.

  However, in the combination weigher, fluctuations in the supply weight of the articles supplied to the weighing hopper may include fluctuations with a relatively short cycle in addition to fluctuations due to random fluctuations as described above. The relatively short fluctuation of this cycle will be described.

  In the combination weigher, as shown in FIG. 14, an article is supplied from a supply device 2 onto a conical top cone 3a. The supply of this article is performed by detecting the amount of the article on the top cone 3a. When it is detected that the lower limit has been reached by the container, supply of articles from the supply device 2 to the top cone 3a is started, and the amount of articles on the top cone 3a has reached the upper limit by the detector. Is detected, the supply of articles from the supply device 2 is stopped intermittently.

  Since the articles are intermittently supplied in this way, the thickness of the overlapping article layers on the top cone 3a, that is, the article layer thickness varies. Further, the weight of the articles supplied to each linear feeder tray 4a due to variations in the layer thickness due to uneven distribution of articles on the top cone 3a and differences in the behavior of articles at each location on the top cone 3a. In this case, the fluctuations tend to be relatively short, that is, the fluctuations tend to increase or decrease as a whole while being randomly varied each time. For example, if the article supplied from the supply device 2 to the top cone 3a has a biased drop position on the top cone 3a, the article staying amount is biased at a part of the top cone 3a. The amount of change in the weight of articles supplied to the weighing hopper 6 corresponding to is also increased. Further, if the supply amount of each article supplied to the entire top cone 3a is too large or too small and there is an excessive difference from the discharge amount from the linear feeder 4, the entire weighing hopper 6 The uniformity of the supply weight of the articles to each of them decreases, and the amount of weight fluctuation of the entire weighing hopper 6 becomes too large.

  As described above, the weight of the articles supplied to the weighing hopper 6 may not only include unpredictable random variations but also tend to increase or decrease with a relatively short period.

  Therefore, if the average value of the supply weight of the article is calculated and controlled while maintaining a relatively large number of samples so as to smooth out random fluctuations that cannot be predicted and suppress the influence, the above comparison If there is a trendy weight fluctuation with a short period, the number of samples is too large, the weight fluctuation is smoothed, and the error of the supply weight of the article from the target supply weight Wt / M increases. There are things to do.

  The present invention has been made in view of the above-described circumstances, and in the case where the weight of an article supplied to the weighing hopper is controlled based on the supply weight corresponding to the number of samples supplied to the weighing hopper, the weighing is performed. An object of the present invention is to improve the accuracy of control so that the number of samples can be appropriately changed in accordance with a change in the weight of an article supplied to a hopper.

  In order to achieve the above object, the present invention is configured as follows.

(1) The combination weigher of the present invention includes a plurality of feeders that respectively supply articles, and a plurality of feeders that respectively supply the articles via the plurality of feeders and that measure the weight of the supplied articles. With a weighing hopper,
In a combination weigher that performs a combination calculation based on the weight of the article and a target combination weight that are respectively weighed by a plurality of weighing hoppers,
Predicted value calculation means for estimating and calculating a predicted value related to the supply weight of the weighing hopper based on the supply weight of the article for two or more samples supplied to the weighing hopper;
A sample number varying means for varying the number of samples based on a supply weight of the article supplied to the weighing hopper;
Based on the predicted value of the weighing hopper estimated and calculated by the predicted value calculating means, the supply weight of the article supplied to the weighing hopper is defined based on the target combination weight. so that the target supply weight, and control means for controlling the supply amount by the feeder,
The number of samples is 3 or more,
The predicted value calculation means estimates and calculates a change amount of the supply weight per sample number as a trend variation amount based on the supply weight of the sample supplied to the weighing hopper. At the same time, based on the trend variation amount, the predicted value related to the supplied weight reflecting the trend variation amount is estimated and calculated .

  The predicted value related to the supplied weight may be a predicted value of the supplied weight itself, or a predicted value of an error of the supplied weight from the target supplied weight.

  The sample number varying means preferably varies the number of samples based on fluctuations in the supply weight of the article supplied to the weighing hopper.

According to the combination weigher of the present invention, the predicted value related to the supply weight is calculated based on the supply weight corresponding to the number of samples of the supply weight of the articles supplied to the weighing hopper, and supplied to the weighing hopper based on the calculated predicted value. The supply amount of the feeder is controlled so that the supply weight of the article to be supplied becomes the target supply weight. However, since the number of samples is variable based on the supply weight of the article supplied to the weighing hopper, It is possible to control by changing to an appropriate number of samples so as to suppress fluctuations in the supply weight of the articles supplied to the hopper.
In addition, based on the supply weight of three or more samples of the supply weight of the articles supplied to the weighing hopper, the amount of change in the supply weight per sample number is estimated and calculated as the amount of trend fluctuation. Since the estimated value related to the supply weight reflecting the amount is estimated and calculated, and based on this predicted value, the supply amount by the feeder is controlled so that the supply weight of the articles supplied to the weighing hopper becomes the target supply weight. When the supply weight supplied to the weighing hopper fluctuates every time but changes as a whole or changes in a decreasing direction, that is, when there is a gradual increase or decrease, control that follows the change Is possible.

(2) The combination weigher of the present invention includes a plurality of feeders for supplying articles, a plurality of supply hoppers to which the articles are respectively supplied via the plurality of feeders, and the articles from the plurality of supply hoppers. And a plurality of weighing hoppers for weighing each of the supplied articles.
In a combination weigher that performs a combination calculation based on the weight of the article and a target combination weight that are respectively weighed by a plurality of weighing hoppers,
Each of the plurality of supply hoppers can respectively weigh the supplied article.
Predicted value calculation means for estimating and calculating a predicted value related to the supply weight of the supply hopper based on the supply weight of the article for two or more samples supplied to the supply hopper;
A sample number varying means for varying the number of samples based on a supply weight of the article supplied to the supply hopper;
Based on the predicted value of the supply hopper estimated and calculated by the predicted value calculation means, the supply weight of the article supplied to the supply hopper is defined based on the target combination weight. so that the target supply weight, and control means for controlling the supply amount by the feeder,
The number of samples is 3 or more,
The predicted value calculation means estimates and calculates a change amount of the supply weight per sample number as a trend variation amount based on the supply weight of the number of samples of the supply weight of the article supplied to the supply hopper. At the same time, based on the trend variation amount, the predicted value related to the supplied weight reflecting the trend variation amount is estimated and calculated .

  The predicted value related to the supplied weight may be a predicted value of the supplied weight itself, or a predicted value of an error of the supplied weight from the target supplied weight.

  The sample number varying means preferably varies the number of samples based on fluctuations in the supply weight of the article supplied to the supply hopper.

According to the combination weigher of the present invention, the predicted value related to the supply weight is calculated based on the supply weight corresponding to the number of samples of the supply weight of the articles supplied to the supply hopper configured to be capable of being measured, and based on the calculated prediction value Thus, the supply amount by the feeder is controlled so that the supply weight of the articles supplied to the supply hopper becomes the target supply weight. The number of samples is based on the supply weight of the articles supplied to the supply hopper. Therefore, it is possible to control by changing to an appropriate number of samples so as to suppress fluctuations in the supply weight of the articles supplied to the supply hopper.
Also, based on the supply weight of three or more samples of the supply weight of the article supplied to the supply hopper, the amount of change in the supply weight per sample number is estimated and calculated as a trend fluctuation amount. Since the estimated value related to the supply weight reflecting the amount is estimated and calculated, and based on this predicted value, the supply amount by the feeder is controlled so that the supply weight of the articles supplied to the supply hopper becomes the target supply weight. When the supply weight supplied to the supply hopper fluctuates every time, but changes in the overall increase direction or in the decrease direction, that is, if there is a gradual increase or decrease change, control that follows the change Is possible.

   (3) In another embodiment of the combination weigher of the present invention, the sample number variable means includes a supply weight of the article supplied to at least one of the weighing hopper and the supply hopper, and the target supply. The number of samples is varied based on the weight.

  The sample number variable means varies the number of samples based on the supply weight of the article supplied to the weighing hopper and the target supply weight, or the supply weight of the article supplied to the supply hopper Preferably, the number of samples is variable based on the target supply weight, but based on the supply weight of the article supplied to both the weighing hopper and the supply hopper, and the target supply weight. The number of samples may be varied.

  The target supply weight of the weighing hopper and the target supply weight of the supply hopper are preferably the same.

  According to this embodiment, the number of samples is determined based on the supply weight of articles supplied to at least one of the weighing hopper and the supply hopper, and the target supply weight to at least one of the weighing hopper and the supply hopper. Therefore, control is performed by changing the number of samples to an appropriate number so that the deviation between the supply weight of the article supplied to at least one of the weighing hopper and the supply hopper and the target supply weight becomes small. Is possible.

( 4 ) In another embodiment of the combination weigher of the present invention, the predicted value calculation means supplies the article supplied to the hopper of at least one of the weighing hopper and the supply hopper as the predicted value related to the supply weight. The estimated value of the error from the target supply weight of the supply weight is estimated and calculated.

  According to this embodiment, in the predicted value calculation means, based on the supply weight of articles for the number of samples, the target supply weight of the supply weight of articles supplied to at least one of the weighing hopper and the supply hopper is calculated. Since the estimated value of the error is estimated and calculated, the control means may control the supply weight of the article to at least one of the weighing hopper and the supply hopper so as to cancel the calculated estimated error value. Good.

( 5 ) The combination weigher of the present invention includes a plurality of feeders that respectively supply articles, and a plurality of feeders that respectively supply the articles via the plurality of feeders and that measure the weight of the supplied articles. With a weighing hopper,
In a combination weigher that performs a combination calculation based on the weight of the article and a target combination weight that are respectively weighed by a plurality of weighing hoppers,
Predicted value calculation means for estimating and calculating a predicted value related to the supply weight of the weighing hopper based on the supply weight of the article for two or more samples supplied to the weighing hopper;
A sample number varying means for varying the number of samples based on a supply weight of the article supplied to the weighing hopper;
Based on the predicted value of the weighing hopper estimated and calculated by the predicted value calculating means, the supply weight of the article supplied to the weighing hopper is defined based on the target combination weight. Control means for controlling the supply amount by the feeder so as to achieve a target supply weight ,
The predicted value calculating means, based on the supplied weight of the number of samples of the article to be supplied to said metering hot path, the mean value for supplying the weight of the weighing hot path, and estimates calculated as the predicted value Yes,
A standard deviation of the supply weight of the article supplied to the weighing hopper is calculated for each different number of samples, and the supply is a difference between the supply weight of the article supplied to the weighing hopper and the target supply weight A standard deviation and supply weight error calculating means for calculating an average value of weight errors;
Display means for displaying an average value of the standard deviation and the supplied weight error together with the number of samples .

According to the present invention, in the predictive value calculating unit, based on the supplied weight of the sample number of the article, the average value for the supply weight weighing hot path, for example, an average value of the supply weight metering hot path, next metering hot It can be estimated and calculated as a predicted value of the supply weight of the pad.
In addition, since the standard deviation of the supply weight of the articles supplied to the weighing hopper represents the degree of random variation of the supply weight, according to the present invention, both the random variation of the supply weight and the supply weight error are considered. Thus, for example, the number of samples at the beginning of the operation operation can be selected and set.
(6) The combination weigher of the present invention includes a plurality of feeders for supplying articles, a plurality of supply hoppers to which the articles are supplied via the plurality of feeders, and the articles from the plurality of supply hoppers. And a plurality of weighing hoppers for weighing each of the supplied articles.
In a combination weigher that performs a combination calculation based on the weight of the article and a target combination weight that are respectively weighed by a plurality of weighing hoppers,
Each of the plurality of supply hoppers can respectively weigh the supplied article.
Predicted value calculation means for estimating and calculating a predicted value related to the supply weight of the supply hopper based on the supply weight of the article for two or more samples supplied to the supply hopper;
A sample number varying means for varying the number of samples based on a supply weight of the article supplied to the supply hopper;
Based on the predicted value of the supply hopper estimated and calculated by the predicted value calculation means, the supply weight of the article supplied to the supply hopper is defined based on the target combination weight. Control means for controlling the supply amount by the feeder so as to achieve a target supply weight,
The predicted value calculation means is configured to estimate and calculate an average value related to the supply weight of the supply hopper based on the supply weight of the article for the number of samples supplied to the supply hopper, as the predicted value.
The standard deviation of the supply weight of the article supplied to the supply hopper is calculated for each different number of samples, and the supply is the difference between the supply weight of the article supplied to the supply hopper and the target supply weight A standard deviation and supply weight error calculating means for calculating an average value of weight errors;
Display means for displaying an average value of the standard deviation and the supplied weight error together with the number of samples .
According to the present invention, the predicted value calculation means calculates the average value related to the supply weight of the supply hopper, for example, the average value of the supply weight of the supply hopper, based on the supply weight of the articles for the number of samples. It can be estimated and calculated as a predicted value of weight.
In addition, since the standard deviation of the supply weight of the articles supplied to the supply hopper represents the degree of random variation of the supply weight, according to the present invention, both the random variation of the supply weight and the supply weight error are considered. Thus, for example, the number of samples at the beginning of the operation operation can be selected and set.

(7) In another embodiment of the present invention, a display unit may be provided that displays the trend variation amount estimated and calculated by the predicted value calculation unit together with the number of samples for each different number of samples.

  According to this embodiment, when the number of samples is different, the trend variation amount indicating the trend variation of the supply weight of the article supplied to at least one of the weighing hopper and the supply hopper is displayed on the display means. As a result, it is possible to grasp how much of the trend changes, and for example, it is possible to select and set the number of samples at the beginning of the operation operation.

  (8) In the embodiment of the above (7), a standard deviation for calculating a standard deviation of a supply weight of the article supplied to at least one of the weighing hopper and the supply hopper for each different number of samples. A calculation means may be provided, and the display means may display the standard deviation calculated by the standard deviation calculation means together with the number of samples.

  Since the standard deviation of the supply weight of the articles supplied to at least one of the weighing hopper and the supply hopper represents the degree of random variation of the supply weight, according to this embodiment, the random variation of the supply weight For example, it is possible to select and set the number of samples at the beginning of the operation operation in consideration of both trending fluctuations.

  (9) In the embodiment of the above (6), for each of the different numbers of samples, a standard deviation of the supply weight of the article supplied to at least one of the weighing hopper and the supply hopper is calculated, Standard deviation and supply weight error calculation for calculating an average value of a supply weight error that is a difference between the supply weight of the article supplied to at least one of the weighing hopper and the supply hopper and the target supply weight Means and a display means for displaying the average value of the standard deviation and the supply weight error together with the number of samples.

  Since the standard deviation of the supply weight of the articles supplied to at least one of the weighing hopper and the supply hopper represents the degree of random variation of the supply weight, according to this embodiment, the random variation of the supply weight Considering both the supply weight error, for example, it is possible to select and set the number of samples at the beginning of the operation operation.

  According to the present invention, the predicted value related to the supply weight is calculated based on the supply weight corresponding to the number of samples of the supply weight of the articles supplied to the weighing hopper or the supply hopper configured to be capable of being measured, and based on the calculated predicted value. Thus, the supply amount by the feeder is controlled so that the supply weight of the articles supplied to the weighing hopper or the supply hopper becomes the target supply weight, but the number of samples is supplied to the weighing hopper or the supply hopper. Since it varies based on the supply weight of the article, it is possible to control by changing to an appropriate number of samples so as to suppress fluctuations in the supply weight of the article supplied to the weighing hopper or the supply hopper.

FIG. 1 is an external view of a combination weigher according to an embodiment of the present invention. FIG. 2 is a block diagram showing a schematic configuration of a main part of the control system of the combination weigher. FIG. 3 is a diagram illustrating an example of changes in the number of articles supplied to the weighing hopper and the supply weight. FIG. 4 is a diagram illustrating a feedback control model of the weight of articles supplied to the weighing hopper. FIG. 5 is an operation sequence diagram showing the acquisition timing of the weight supplied by the weighing hopper during the adjustment operation. FIG. 6 is a diagram showing a section average standard deviation and a section trend fluctuation amount for each number of samples subjected to the aggregation processing by the adjustment operation. FIG. 7 is a flowchart of the counting process during the adjustment operation. FIG. 8 is a flowchart of the counting process following FIG. FIG. 9 is a flowchart of the counting process following FIG. FIG. 10 is a flowchart of the counting process following FIG. FIG. 11 is a flowchart during operation. FIG. 12 is a flowchart at the time of operation following FIG. FIG. 13 is a diagram corresponding to FIG. 3 of another embodiment. FIG. 14 is a schematic configuration diagram of a conventional combination weigher.

(Embodiment 1)
FIG. 1 is a schematic configuration diagram of a combination weigher according to an embodiment of the present invention.

  The combination scale 1 of this embodiment includes a supply device 2 that supplies articles such as confectionery to be weighed by vibration, a conical top cone 3a and a top that distribute and convey the articles supplied from the supply device 2 in a radial manner. A main feeder (dispersing feeder) 3 having a vibrator 3b that vibrates the cone 3a, a linear feeder trough (straight forward tray) 4a that conveys articles supplied from the top cone 3a to the outside, and a linear feeder trough 4a are vibrated. A plurality of linear feeders (straight forward feeders) 4 each having a vibrator 4 b, a plurality of supply hoppers 5 for temporarily holding and discharging articles supplied from the respective linear feeders 4, and a supply from each supply hopper 5 A plurality of weighing hoppers 6 for containing and weighing the collected articles and articles discharged from the weighing hopper 6 are collected and supplied downstream. A plurality of collecting chutes 7, and a discharge chute 8 to discharge to the packaging machine (not shown) collects the articles supplied from the collecting chute 7.

  The main feeder 3, the linear feeder 4, the supply hopper 5, the weighing hopper 6, and the like are mounted on a central center base 22.

  Each supply hopper 5 and each weighing hopper 6 are provided with hopper gates 5a and 6a at the lower part, respectively, and the articles are held and discharged by closing and opening the hopper gates 5a and 6a.

  Further, the combination weigher 1 according to the present embodiment is housed in a control device 9, a level detector 10 made of, for example, an optical sensor that detects the amount of articles on the top cone 3 a, and a weighing hopper 6 as a control system. A weight sensor 11 including a load cell for detecting the weight of the article and an operation setting display 12.

  FIG. 2 is a block diagram showing a schematic configuration of a main part of the control system of the combination weigher 1 in this embodiment, and parts corresponding to those in FIG.

  As illustrated in FIG. 2, the control device 9 includes a control unit 21 configured by, for example, a microcomputer. The control unit 21 controls each unit and is given a load signal of a plurality of weight sensors 11 respectively connected to a main calculation control unit 17 that performs a combination calculation, a storage unit 18, and a plurality of weighing hoppers 6. / D conversion unit 15, an individual calculation unit 16 that individually calculates the weight of articles supplied to each weighing hopper 6, and a vibration control circuit unit 19 that controls vibrations of the plurality of linear feeders 4. . The storage unit 18 stores an operation program of the combination weigher 1 and operation parameters set by the operation setting display 12 and serves as a work area for operations of the main operation control unit 17. The operation parameters include vibration intensity (vibration amplitude), drive time, target combination weight Wt, and the like of the linear feeder 4 that conveys an article by vibration.

  In the combination calculation, the weighing values of the weighing hoppers 6 are variously combined, and the combined weight, which is the combined weight, is equal to or more than the target combination weight Wt and equal to or less than the allowable upper limit value Wu. The closest combination of the weighing hoppers 6 is selected as the optimum combination.

  The control unit 21 controls the driving of the supply device 2 based on the detection output of the level detector 10 in FIG. 1, and when the amount of articles on the top cone 3a reaches the lower limit, When the supply of the article is started and the amount of the article reaches the upper limit value, the article is intermittently supplied to the top cone 3a so that the supply of the article to the top cone 3a is stopped. The control unit 21 controls the opening and closing of the hopper gates 5a and 6a of the supply hopper 5 and the weighing hopper 6 of FIG.

  As described above, in the combination weigher 1, since articles are intermittently supplied from the supply device 2 onto the top cone 3a, the thickness of the layers of the overlapping articles on the top cone 3a, that is, the thickness of the articles varies. Further, the weight of the articles supplied to each linear feeder trough 4a due to the variation in the layer thickness due to the uneven distribution of the articles on the top cone 3a or the difference in the behavior of the articles at each part on the top cone 3a. In some cases, the fluctuations tend to increase or decrease with a relatively short period.

  In this case, a gradual increase / decrease fluctuation also occurs in the weight of articles supplied from the respective linear feeder troughs 4a to the respective supply hoppers 5, and hence the weight of articles supplied from the respective supply hoppers 5 to the respective weighing hoppers 6.

  Such a gradual increase / decrease fluctuation has a higher fluctuation speed and a larger fluctuation amount than a fluctuation caused by a change in ambient temperature or the like. The variation in the weight of the articles supplied to each weighing hopper 6 may include a gradual increase / decrease variation at a relatively fast speed (short cycle) as described above.

  In this embodiment, the following is performed in order to perform control that follows even when such a tendency of increase or decrease in the weight of the article supplied to the weighing hopper 6 is present.

  That is, the control unit 21 obtains the supply weight of the supplied article for each weighing hopper 6 every time the article is supplied, calculates an error from the target supply weight Wt / M, and calculates the calculated three times. The trend of fluctuations in the supply weight error corresponding to the number of samples is estimated, and the next supply weight error is predicted based on the estimated trending fluctuation. Then, the control unit 21 cancels the predicted value of the supply weight error, and calculates the operation amount of the linear feeder 4 so that the supply weight of the article supplied to the weighing hopper 6 becomes the target supply weight Wt / M. Then, the supply weight of the article by the linear feeder 4 is controlled. Further, the control unit 21 automatically changes the number of samples as will be described later.

  Here, the target supply weight Wt / M is the same as described above, and is a value obtained by dividing the target combination weight Wt in the combination calculation by the target combination number M of the weighing hoppers 6.

  In this embodiment, the driving time of the linear feeder 4 is set to a fixed time, and the operation amount for the linear feeder 4 defines the vibration strength of the linear feeder 4, that is, the amplitude of vibration.

  Prediction control for a gradual increase / decrease in the supply weight of articles to the weighing hopper 6 with a relatively short cycle will be described in detail. In this embodiment, the predictive control is performed for each weighing hopper 6 as described above.

  Each time an article is supplied to the weighing hopper 6 connected to the weight sensor 11, the control unit 21 calculates a supply weight error as shown in the following formula based on the supply weight measured by the weighing hopper 6.

Supply weight error = Supply weight-Target supply weight Wt / M
Then, an error variation function e, which is an approximation function representing a gradual increase / decrease variation of the supply weight error, is obtained as described below based on the supply weight error of the latest sample number N1 three times or more. Based on the error variation function e, the control unit 21 as the predicted value calculation means and the control means predicts the next supply weight error and controls the supply amount of the articles by the linear feeder 4 so as to cancel the error.

  The reason why the supply weight error of 3 times or more is used is that it becomes a difference in 2 times, and a gradual increase / decrease fluctuation cannot be obtained.

  The prediction of the supply weight error based on the error variation function e will be described with reference to FIG.

In FIG. 3, the vertical axis indicates the weight of articles supplied to the weighing hopper 6, and the horizontal axis indicates the number of times articles are supplied to the weighing hopper 6. FIG. 3 shows a change in the supply weight (W ₁ , W ₂ ,... W ₁₀ ) of the articles to the weighing hopper 6 from the supply frequency of 1 to 10 times when the supply weight is not controlled. The eleventh shows the predicted value based on the supply weight of 1 to 10 times. The target supply weight (Wt / M) is also shown.

  FIG. 3 shows an example in which the weight supplied to the weighing hopper 6 varies every time, but tends to increase from 1 to 10 times as a whole, that is, when there is a gradual increase fluctuation. .

  In this embodiment, a function approximation is performed on the basis of the latest supply weight error of the number of samples N1, and a linear error fluctuation function e representing the trend fluctuation of the supply weight error is obtained.

In Figure 3, the sample number N1 was 10 times, based on the supply weight error from the target supply weight of the feed weight W ₁ to _W-10 from one to 10 times (Wt / M), by the least square method An error variation function e = ax + b that is a straight line rising to the right indicated by a one-dot chain line is obtained. In this error variation function e, x is the number of articles supplied to the weighing hopper 6, and a is a linear gradient, as shown in FIG. It represents the amount of change in the supply weight, and represents the degree of trend fluctuation, which is called the trend fluctuation amount.

  In this embodiment, the supplied weight error from the target supply weight (Wt / M) of the supply weight to the 11th weighing hopper 6 that is the next time is predicted by the obtained error variation function e.

In Figure 3, the predicted value of the supply weight error of 11 th, indicated by E11', the predicted value of the supply weight of 11 th is represented by W ₁₁ '. That is, the predicted value W ₁₁ ′ of the _eleventh supply weight is on the linear error fluctuation function e = ax + b, and the predicted value e ₁₁ ′ of the supply weight error is the target supply weight Wt of the supply weight predicted value W ₁₁ ′. The difference from / M.

As described above, the predicted value e11 ′ of the _11th supply weight error or the predicted value W11 ′ of the supply weight for the next time is 10 times of the supply weights W _{1 to} W ₁₀ from 1 to 10 as the number of samples. The value reflects the trend of increasing fluctuations. In this embodiment, the predicted value e11 ′ of the supply weight error that reflects this gradual increase variation is canceled out so that the eleventh supply weight to the weighing hopper 6 becomes the target supply weight (Wt / M). Next, the vibration intensity of the linear feeder 4 is controlled.

  Here, as in Patent Document 1, a predetermined number of supply weights supplied to the weighing hopper 6 are added, that is, the supply weights for the number of samples are added, and the added supply weight is divided by the number of samples. When the next supply weight is controlled based on the calculated average value of the supply weight, a tendency that the tendency fluctuation cannot be accurately followed will be described.

When the number of samples is, for example, 10 from 1 to 10 as shown in FIG. 3 and the supply weights W _{1 to} W ₁₀ for these ₁₀ are averaged, the average value of the supply weights is indicated by Wa, The average value of the supply weight error is indicated by ea. In Patent Document 1, the average value ea of the supply weight error is substantially set as the predicted value of the 11th supply weight error, which is the next time, and control is performed so as to cancel it.

Average value Wa of the average value ea and supply the dispensed weight weight error, based on the feed weight W ₁ to _W-10 to one to 10 times which is the number of samples, merely by taking the mean, variance every time However, the trend of increasing fluctuations that are increasing as a whole is not considered. Therefore, the average value ea and the average value Wa of feed dispensed weight weight error is fed the number of supply times of about an intermediate, for example, four times the supply weight error of the supply count time of about 6 times and supply weight W ₄ ~ This corresponds to W ₆ , and is smaller than the supply weight error and the supply weight W ₁₀ at the time of the supply number near the latest supply number of 10 times.

  That is, the actual supply weight increases as the number of times of supply increases from one to ten times, although it fluctuates every time, and as a whole increases, the number of times of supply in the vicinity of the tenth time supplies more than in the vicinity of the first time. While the weight increases, Patent Document 1 simply calculates an average value from 1 to 10 times, so that the supply weight error is estimated to be smaller than the actual supply weight error. Control is performed based on the supplied weight error. For this reason, the control does not correspond to the gradual increase fluctuation, and the supply weight to the weighing hopper 6 cannot be correctly controlled to the target supply weight (Wt / M).

  This is not limited to a trend of increasing fluctuations, but is also the case of a trend of decreasing fluctuations. In the case of a trending decreasing fluctuation, the supply weight error is estimated to be larger than the actual supply weight error. It will be.

  Thus, in patent document 1, when there exists a gradual increase / decrease variation, it cannot fully follow.

  Next, the calculation of the error variation function e and the calculation of the operation amount for the linear feeder 4 based thereon will be described in detail.

  In this embodiment, the error of the supply weight of the latest sample number N1 of three times or more is stored in the shift register of each individual calculation unit 16 of FIG. 2, and the stored latest N1 supply weight error tendency is stored. An error variation function representing a large variation is obtained by a method such as a least square method. With this method, even if the error in supply weight varies randomly every time, the variation component can be reduced, and the trending fluctuation amount can be accurately estimated.

For example, the number of samples supplied to the weighing hopper 6 is set to the number of samples xi, and a variation function of an error appearing in the supplied weight of the number of samples xi is set to ei by the least square method as described above. (1)
Stipulate that
b = (average value of ei) −a · (average value of xi)
Therefore, if the trend variation of the supply weight error is small, that is, close to a = 0 and there is almost no trend variation, the error variation function is
b = (average value of ei).

  In this case, the predicted value of the next supply weight error is an average value of the supply weight error of the number of samples N1, and if the number of samples is not changed, the average value for the number of samples is calculated and controlled. This is equivalent to Patent Document 1.

  That is, the smoothing effect for each variation due to unpredictable random disturbances other than the tendency fluctuations is equivalent to this embodiment and Patent Document 1 if the number of samples is not changed.

However, as described above, if there is a tendency variation in the supply weight error, the trend variation amount a which is the variation rate (gradient) of the variation is
a = {Σ (xi · ei) −N1 · (average value of xi) · (average value of ei)} / {Σ (xi ² ) −N1 · (average value of xi) ² }
Therefore, as shown in FIG. 3, the trend variation amount a is a value representing the trend variation of the supply weight error of the latest number of samples N1.

  With respect to this trend variation amount a, each adjustment hopper 6 is given a certain amount of operation to each linear feeder 4 before determining control conditions such as the number of samples during adjustment operation, and each control hopper 6 in a state where no control is performed. When the trend variation amount a of the error of the supply weight for each weighing hopper 6 with respect to the target supply weight is estimated and displayed on the operation setting display 12 as the display means, the variation component is removed and only the variation tendency is extracted. Therefore, it is possible to accurately grasp the actual condition of the article supply characteristics of each linear feeder 4, and to easily determine the initial value of the number of samples at the start of the operation operation, and to adjust the article input position of the supply apparatus 2 and the supply flow rate. become able to.

  Whether the adjustment operation or the operation operation is being performed, the current control followability can be accurately grasped by displaying the error fluctuation tendency amount a for each weighing hopper 6 while performing the control. As a result, the operator can easily readjust for better control. The error trend fluctuation amount is also the fluctuation trend amount of the supply weight.

  In order to control the supply weight, the predicted value ei ′ of the supply weight error of the article to the next weighing hopper 6 is obtained by substituting the number of samples xi = N1 + 1 into the above equation (1).

The supply weight y of the article to the next weighing hopper 6 is set so as to cancel the predicted value ei ′ by the calculated supply weight error predicted value ei ′.
y = (Wt / M) −ei ′
Based on this next supply weight y, a feeder operation amount Yx, which is the operation amount of the next linear feeder 4, is converted into a feeder operation amount f (y) previously obtained at the time of adjustment as will be described later. Use to calculate.

  The trending variation may be expressed by a quadratic expression, and the trend variation amount a may be obtained as a coefficient of x as a result of differentiating the quadratic expression.

  Next, a method for calculating the feeder operation amount based on the next supply weight y obtained as described above will be described.

  FIG. 4 shows a model of feedback control of the supply weight for the individual weighing hoppers 6. Disturbance signals due to random variations and tendency fluctuations are added to the weight of articles supplied to the supply hopper 5 by the linear feeder 4. Since articles are supplied to the weighing hopper 6 via the supply hopper 5, a waste time for one time is generated, but this waste time element is ignored.

  The controller 13 for calculating the feeder operation amount Yx based on the deviation between the supply weight Wx of the articles supplied to the weighing hopper 6 and the target supply weight Wt / M which is a set value is supplied to the control unit 21 in FIG. Equivalent to.

  The controller 13 calculates the predicted value of the supply weight error as described above, and calculates the next supply weight y to the weighing hopper 6 so as to cancel the predicted value of the supply weight error.

  The calculated supply weight y is converted into a feeder operation amount Yx as follows, D / A converted by a D / A converter built in the controller 13, and an analog output signal is given to the feeder driver 14. The linear driver 4 is driven by the feeder driver 14.

  For example, the linear feeder 4 vibrates with a magnitude determined by an analog output signal output from the controller 13 at a predetermined frequency for a predetermined set time.

  In order to convert the supply weight y into the manipulated variable, first, at the time of adjustment of the combination weigher 1, supply of articles actually supplied to the weighing hopper 6 with respect to the input value of the D / A converter built in the controller 13 An input value p to the D / A converter is obtained at which the weight is the target supply weight Wt / M.

  When the input value p is obtained, a conversion coefficient K such that Wt / M = K · p is obtained.

  When various values y are given as the supply weight y of articles to the weighing hopper 6, the weight of articles actually supplied to the weighing hopper 6 via the linear feeder 4 becomes the value of y. Must do so.

  The operation amount for supplying an article with a supply weight y is Yx, the value of the operation amount Yx is increased or decreased so that the supply weight changes around the target supply weight Wt / M, and Yx is multiplied by the conversion coefficient K. When input to the D / A converter, the weight value Wx of the article actually supplied to the weighing hopper 6 is measured.

At this time, a relational expression for obtaining the manipulated variable input value Yx necessary for obtaining the supplied weight Wx by the least square method or the like using a plurality of sets of the actually supplied weight Wx and the input value Yx.
Yx = f (Wx)
Create During operation, since the next supply weight Wx is given by y by the following control calculation,
Yx = f (y)
And Yx is called a feeder operation amount.

  During the operation, when the next supply weight y is calculated by the control calculation, the feeder operation amount Yx is set so that the next supply weight of the article to the weighing hopper 6 becomes actual as the calculated supply weight y. Is calculated from the above equation.

  P, which is an input value to the D / A converter, is determined with Yx and the conversion coefficient K, and is input to the D / A converter. Note that the conversion gain K = 1, and the amplitude gain of the feeder driver 14 is adjusted so that the actual supply weight becomes Wt / M by the analog output voltage when Yx = Wt / M is input to the D / A converter. You may make it do.

  At the time of operation, when the next supply weight value y is obtained, the value of Yx obtained by Yx = f (y) is multiplied by the conversion coefficient K and given to the D / A converter, and the linear feeder 4 is controlled.

  As described above, in the present embodiment, the operation amount with respect to the linear feeder 4 is controlled based on the supply weight error to the weighing hopper 6 having the latest sample number N1 of 3 or more, that is, the weighing hopper 6 is supplied to the weighing hopper 6. The operation amount with respect to the linear feeder 4 is controlled using N1 pieces of the supply weight error as a sample. This sample number N1 is automatically varied during the operation operation. The setting of the sample number N1 at the beginning of the operation operation, that is, the setting of the initial value of the sample number N1 will be described.

  Factors of the supply weight error of articles to the weighing hopper 6 include variations due to random disturbances that cannot be predicted as described above, and gradual fluctuations, and the number of samples N1 is affected by variations due to random disturbances. It is preferable to set a value that can be suppressed and that can follow a gradual increase and decrease change.

  In the above-mentioned Patent Document 1, the number of samples for calculating the average value is set so as to suppress the influence of variation due to random disturbances, without special consideration for the gradual increase / decrease variation with a relatively short cycle. Has been.

  If the number of samples N1 is increased, it is possible to smooth the variation caused by random disturbances and suppress the influence thereof, but it becomes difficult to follow the trend of fluctuations with a relatively short period. Conversely, the number of samples N1 However, the influence of variation due to random disturbance cannot be sufficiently suppressed.

  In this embodiment, the influence of variation due to random disturbance is suppressed, and a guideline for setting the number of samples N1 that can follow a gradual increase / decrease variation with a relatively short cycle as an initial value at the start of operation is given. I am doing so.

  Specifically, prior to the operation operation, an adjustment operation for setting an appropriate number of samples N1 as an initial value is performed.

  In this adjustment operation, the operator operates the operation setting display 12 to set the lower limit value NL and the upper limit value NU of the number of samples N1. The control unit 21 varies the sample number N1 from the set lower limit value NL to the upper limit value NU, measures the supply weight of the articles supplied to the weighing hopper 6, and varies due to random disturbance for each sample number. A section average standard deviation and a section trend fluctuation amount, which will be described later, are calculated as evaluation index values for evaluating each of the trend fluctuations and displayed on the operation setting display 12.

  In this adjustment operation, it is to grasp how the supply weight of the articles supplied to the weighing hopper 6 varies according to the size of the number of samples. Therefore, the linear feeder 4 is not controlled and the linear feeder is not controlled. The operation amount for 4 is maintained at a constant operation amount corresponding to the target supply weight Wt / M.

  This adjustment operation provides a guideline for setting an appropriate number of samples to be used during the operation operation as an initial value. Therefore, the adjustment operation is determined as much as possible except that the linear feeder 4 is not controlled. It is preferable to carry out a test on the same operating conditions, the same article, and the same combination calculation conditions as the operating operation in accordance with the specified specification conditions.

  In this adjustment operation, as described above, the number of samples is changed in order from the lower limit value NL to the upper limit value NU, the supply weight of the articles supplied to the weighing hopper 6 is measured, the data is totaled, and the random disturbance An index value for evaluation for evaluating the variation due to the above and the trending variation is calculated. As described above, the lower limit value NL and the upper limit value NU are set by the operator by operating the operation setting display 12, and the operator operates the operation setting display 12 to instruct the period during which data is aggregated. To do. In this example, a case where “5”, which is 3 or more, is set as the lower limit value NL of the number of samples and “15” is set as the upper limit value will be described.

5, during the adjustment operation is an operation sequence diagram illustrating the acquisition timing of the weight of the articles supplied to the weighing hoppers 6 shows an operation sequence for one certain weighing hoppers 6 ₁ specified .

For a period up to the count command OFF from aggregation command ON data, the weighing values are aggregated is obtained which is supplied weight of articles supplied to the weighing hoppers 6 _1. In this example, since the lower limit value NL = 5 and the upper limit value NU = 15 as described above, the number of samples 5 to 15 is set as a block as one unit, and the data calculation operation is repeated. I am doing so. Incidentally, C _A indicates a counter for counting the number of samples of up to 15 which is the upper limit NU, C _B represents a counter for counting the number of blocks. The counters C _A and C _B and the shift register are provided in the control unit 21 of FIG.

  In the total processing of this adjustment operation, the supply weight for each section according to the number of samples, for example, when the number of samples is 5, the supply weight of the section from 1 to 5 and the number of samples is 6 Is based on the supply weight of each section according to the number of samples, such as the supply weight of 1 to 6 sections, or the supply weight of 1 to 7 sections when the number of samples is 7 Perform aggregation processing.

  In this embodiment, a section standard deviation, which is a standard deviation of a section, is obtained based on the supplied weight for each section according to the number of samples, and a section average that is a standard deviation of the average value of the section is calculated based on the section standard deviation value. The value standard deviation is obtained as follows.

That is, if the number of samples is N1 and the section standard deviation is σ, the section average standard deviation is obtained as σ / (N1) ^1/2 .

  Further, the trend variation amount of the error variation function e for each section is obtained as the section trend variation amount a.

  For each section corresponding to the number of samples, the section average standard deviation and the section trend fluctuation amount are obtained. Therefore, in the adjustment operation, the section average standard over the data summing period from the summation command ON to the summation command OFF. A process of calculating and storing a maximum value, a minimum value, and an average value for the deviation and the interval trend fluctuation amount is performed.

Section trend amount a is, with the increase or decrease in the supply weight error of the articles supplied to the weighing hoppers 6 _1, but with a positive or negative polarity, here represents the magnitude of the trend amount be either positive or negative The absolute value is used as the value.

  For example, FIG. 6 shows the maximum value, minimum value, average value of the section average standard deviation and the maximum value, minimum value, and average value of the section trend fluctuation amount in the data collection period from the summation command ON to the summation command OFF. As shown, a table is created for each number of samples from the lower limit value NL = 5 to the upper limit value NU = 15 of the number of samples and displayed on the operation setting display 12 as a display means.

  In the table of FIG. 6, the section average standard deviation represents variation due to random disturbance, and the section trend fluctuation amount represents trending fluctuation.

  These are evaluation index values for evaluating the characteristics of the supply weight error related to the amount of variation and tendency variation of the combination weigher 1 in the adjustment operation without control. With reference to these index values, the operator determines an appropriate initial value of the number of samples N1 to be set at the start of operation.

  In the combination calculation, in order to control the supply weight of the articles supplied to each weighing hopper 6 to the target supply weight Wt / M so that the combination weight can be accurately selected by increasing the number of combinations to be selected, With respect to fluctuations in the supply weight, there is a need for a control that is less affected by random fluctuations and that can follow a fluctuation with a relatively short cycle.

  In FIG. 6, the interval average standard deviation of errors representing variation due to random disturbance and the interval trend fluctuation amount of errors representing trending variations increase as the number of samples decreases, and decrease as the number of samples increases. appear.

  Based on the numerical value of the section average standard deviation and the numerical value of the section trend fluctuation amount in the table of FIG. 6, the degree of variation due to random disturbance and the degree of trend fluctuation are determined, and an appropriate sample is obtained. Determine the initial value of the number.

  For example, when the interval trend fluctuation amount is small, that is, when the trend fluctuation is small, it is necessary to suppress the influence of variation due to unpredictable random disturbance, that is, to increase the number of samples in order to smooth the variation. is there.

  On the other hand, when the section trend variation amount is large, that is, when the trend variation is large, it is necessary to reduce the number of samples in order to control following the trend variation.

  When the operator initially controls the sample number to a small value, control is good for fast fluctuations in the layer thickness of the article, i.e., tendency fluctuations, but the fluctuation due to unpredictable random disturbance is large. In such a case, it is necessary to consider the adverse effect that the smoothness cannot be sufficiently smoothed and the variation in the calculated operation amount is enlarged, and the resulting variation in the supply weight error is increased.

  On the other hand, if the number of samples is initially set to a large value and controlled, the smoothing characteristics against large random disturbances are good, but the adverse effect of expanding the supply error due to the follow-up delay with respect to the trend fluctuation of the supply weight error is taken into account. There must be.

  Although the variation due to random disturbance is small, when operating on an article with a short trending fluctuation period, the sample number is initially set to a smaller value, and the tendency fluctuation period is long, but the fluctuation due to random disturbance is large. When operating on an article, it is preferable to initialize the number of samples to a larger value.

  When both the variation due to random disturbance and the trending fluctuation speed are large, the number of samples is set to an intermediate number considering the balance of both sizes.

  Note that, for example, a plurality of tables in FIG. 6 may be created by changing the data aggregation period in the adjustment operation, and the number of samples may be set based on them.

  In this way, the operator looks at the table of FIG. 6 displayed on the operation setting display 12 to determine the variation due to random disturbance and the tendency from the section average standard deviation and the section tendency variation amount for each sample number. It is possible to determine an appropriate initial value of the number of samples at the start of the operation operation after judging the variation due to the fluctuation.

The acquisition of suitable measurement and index for the evaluation of the feed by weight of the weighing hopper 6 in the adjustment operation for selecting the number of samples initial value of, for one certain weighing hoppers 6 ₁ specified as above Alternatively, the measurement may be performed for all the weighing hoppers 6 ₁ to 6 n and the average value thereof may be displayed as a representative value.

  This adjustment operation is mainly performed by the manufacturer, and an appropriate initial value of the number of samples N1 is set and shipped as a product. Note that the user may also perform the adjustment operation to change the initial setting of the sample number N1.

  Next, based on the flowchart of FIG. 7, the aggregation process in the adjustment operation of FIG. 5 will be described.

  First, it is determined whether or not the data totaling command is turned on (step n1). If not, all the totaling registers and flags are cleared and the process ends (step n15).

When data aggregation command is ON, it reads the weight values Wx1 of a certain one of the weighing hoppers 6 ₁ designated (step n2), the weight value Wx1 read is, whether it is less than zero near the detection level Wzt1 determines (step n3), when not less than zero near the detection level Wzt1 are weighing hoppers 6 ₁ moves to step n5 as not empty, when it is less than zero near the detection level Wzt1 are weighing hoppers 6 _1, zero near the detection level A flag F0 indicating that a state less than Wzt1, ie, an empty state has passed, is set to “1”, and the process proceeds to step n5 (step n4).

At step n5, the flag F0 is judged whether or not "1", when not "1", the process returns to step n1 as the weighing hoppers 6 ₁ not through the empty, when it is "1", weighing hopper 6 ₁ is shifted to step n6 as through the empty. In step n6, it is determined whether or not the read weight value Wx1 is greater than the article placement detection level Wzt2 indicating that the article has been placed. If the weight value Wx1 is not greater than the article placement detection level Wzt2, the article is placed. If not, the process returns to step n1. In step n6, the weight values Wx1 read is, when a larger level Wzt2 out article placing置検, after weighing hoppers 6 ₁ passed through the empty, as an article has been fed, the weighing hoppers 6 ₁ through the empty The flag F0 indicating this is cleared to “0” (step n7), and the stable weight value is read again after the time of the stability waiting timer has elapsed (steps n8 and n9).

Next, at step n10, it increments the counter C _A for counting the number of samples of up to 15 which is the upper limit number of sample number, read calculates the square of the weight values Wx1 (step n11), integrating (step n12) . Further, the read weight value Wx1 is integrated (step n13), the shift register for storing the weight value is shifted to the right to store the latest weight value Wx1 at the left end (step n14), and the process proceeds to step n16 in FIG.

  Next, in step n16 of FIG. 8, it is determined whether or not the flag F5 indicating that the total calculation processing of up to five pieces of data has been completed is “0”, that is, up to five total calculation processing. Is not completed, the process proceeds to step n17 if not completed, and to n27 in FIG. 9 if completed.

In step n17, the counter C _A for counting the number of samples to determine whether it is "5", when not "5", the flow returns to step n1 as the data up to five is not being collected. In step n17, when the counter C _A for counting the number of samples becomes "5", because five of data which is the lower limit of the number of samples have been collected, it increments the counter C _B for counting the number of blocks (Step n18) Total data calculation processing for five pieces of data is performed. That is, the section standard deviation Ws5 ′ is calculated from C _A , ΣWx1, and Σ (Wx1) ² , and the section average standard deviation Ws5 is calculated from Ws5 ′ / √5 (step n19).

  Next, an error variation function is calculated from up to five pieces of data from the left end of the shift register storing the acquired weight value, and an interval trend variation amount a5 is calculated. This section tendency fluctuation amount a5 is an absolute value (step n20).

  Next, the maximum value and the minimum value are updated for each of the section average value standard deviation Ws5 and the section trend fluctuation amount a5 (step n21). Here, the maximum value and the minimum value are the maximum value and the minimum value of the section average value standard deviation Ws5 and the section trend fluctuation amount a5, respectively, in the counting period. Therefore, in the first block, the maximum value and minimum value registers for storing the maximum value and the minimum value of the section average standard deviation Ws5 and the section trend fluctuation amount a5 are stored in the steps n19 and n20, respectively. The calculated section average standard deviation Ws5 and section trend fluctuation amount a5 are stored as a maximum value and a minimum value, respectively. That is, the maximum value and the minimum value are the same value.

Next, in step n22, the section average value standard deviation Ws5 and the maximum value and the minimum value of the section trend fluctuation amount a5 calculated by the individual calculation section 16 of FIG. transmitted, and stored in the display table, section average value and the standard deviation Ws5 and the section tends variation a5 integrated respectively (step n23), each integrated value is divided by the counter C _B for counting the number of blocks aggregate An average value in the period is obtained (step n24), transmitted to the main arithmetic control unit 17 and stored in the display table (step n25). Is set and the process returns to step n1 (step n26).

  In the above step n16, when the totaling calculation process using up to five pieces of data is completed, the process proceeds to step n27 in FIG. In step n27 of FIG. 9, it is determined whether or not the flag F6 indicating that the totaling calculation process with up to six pieces of data is completed is cleared to “0”, that is, the totaling calculation process with up to six pieces of data is performed. It is determined whether or not it has been completed. If it has not been completed, the process proceeds to a totaling calculation process of up to six data in step n28, and if it has been completed, the process proceeds to step n30.

In the total calculation processing of up to six data in step n28, first, the section standard deviation Ws6 ′ is calculated from CA, ΣWx1, and Σ (Wx1) ² which are the total data for six, and by Ws6 ′ / √6 The section average value standard deviation Ws6 is calculated (step n28-1).

  An error variation function is calculated from up to six pieces of data from the left end of the shift register, and an interval trend variation amount a6 is calculated (step n28-2). The section trend fluctuation amount a6 is an absolute value.

Next, the maximum value and the minimum value are updated for each of the section average value standard deviation Ws6 and the section trend fluctuation amount a6 (step n28-3), transmitted to the main calculation control unit 17, and stored in the display table. (step N28-4), section average value and the standard deviation Ws6 and the section tends variation a6 integrated respectively (step N28-5), each integrated value is divided by the counter C _B for counting the number of blocks aggregate An average value in the period is obtained (step n28-6), transmitted to the main calculation control unit 17 and stored in the display table (step n28-7), indicating that the total calculation processing with up to six data has been completed. “1” is set in the flag F6 (step n29), and the process returns to step n1.

  In the above step n27, when the aggregation calculation process with up to 6 data is completed, the process proceeds to step n30, and the flag F7 indicating that the aggregation calculation process with up to 7 data is completed is set to “0”. It is determined whether or not it has been cleared, that is, whether or not the aggregation calculation process with up to 7 data has been completed. If completed, the process proceeds to step n33 in FIG. The process proceeds to a totaling calculation process with up to seven pieces of data in step n31. The aggregation calculation process with up to 7 data is basically the same as the aggregation calculation process with up to 6 data. When the aggregation calculation process with up to 7 data ends, “1” is set in the flag F7. Set (step n32) and return to step n1.

  In step n30, when the total calculation processing with up to seven pieces of data has been completed, the process proceeds to step n33 in FIG. 10 to determine whether the total calculation processing with up to eight pieces of data has been completed. When the total calculation processing with the data up to is completed, the process proceeds to step n36, and when not complete, the total calculation processing with up to eight data at step n34 is performed.

In the same manner, aggregation calculation processing is performed with up to 9, 10,..., 15 data, and in step n54, aggregation calculation processing with up to 15 data, which is the upper limit value NU, is completed. when the minute aggregate is determined to be completed, it clears the counter C _a to "0" (step n57), the integrated value ΣWx1 weight value, the integrated value of the square of the weight values Σ a (Wx1) ² "0" clears (step N58), the counter C _B count value, i.e., the number of aggregation operation processed block, and displays the transmitted to the main arithmetic control section 17 (step n59), each indicating the completion of the aggregation processing The flags F5, F6,..., F15 are cleared to “0” and the process returns to step n16 in FIG. 8 (step n60).

  As described above, in the adjustment operation, the maximum value, the minimum value, the average value of the section average standard deviation and the maximum value, the minimum value, and the average value of the section trend fluctuation amount are calculated and operated for each number of samples. It is displayed on the setting display 12.

  Based on the section average standard deviation and section trend fluctuation amount obtained by this adjustment operation, an appropriate initial value of the sample number N1 is set, and an upper limit value NU and a lower limit value NL of the sample number are also set. Threshold values α · am <N1>, γ · σm <N1>, and β · am <N1>, which will be described later, are also set and shipped as products.

  In the operation operation by the user, the initial value of the initially set number of samples N1 is automatically varied according to the fluctuation of the error in the supply weight of the article.

  That is, it automatically varies between the upper limit value NU and the lower limit value NL according to the number of samples N1 and the occurrence of supply weight error.

  In this embodiment, the control cannot follow due to an increase in the rate of change of the layer thickness of the article on the linear feeder trough 4a, and the error fluctuation function is calculated when the fluctuation tendency of the error in the supply weight of the article to be generated increases. The number N1 of error samples to be obtained is reduced, for example, by one retroactively from the present time, and a new fluctuation function for estimating the next supply weight is estimated by making it easy to follow the fluctuation speed.

  On the contrary, when the layer thickness of the article on the linear feeder trough 4a is stabilized and the fluctuation tendency of the error in the supply weight of the generated article becomes small, the sample number N1 of the supply weight error for obtaining the error fluctuation function is determined, for example, Go back to the past and increase by one to improve the smoothness characteristics of random disturbances.

  Thus, the next supply weight error is predicted by varying the sample number N1 of the supply weight error according to the change speed of the weight of the article on the linear feeder trough 4a, and the next supply weight is controlled.

  During the operation, the control unit 21 as the sample number varying means automatically varies the number of samples according to the magnitude of the error fluctuation tendency. For example, it is determined as follows.

  The above-described section trend variation amount in FIG. 6 is obtained by determining the variation trend amount a in the supply weight error corresponding to the number of various samples that appeared in the adjustment operation without controlling the operation amount of the linear feeder 4. , A is the value when the weight of the article on the linear feeder trough 4a is substantially constant, and therefore the supply weight error hardly fluctuates, and the maximum value of a is on the linear feeder trough 4a. This is the value when the variation in the weight of the article is the largest, so that the supply weight error varies to the maximum.

Therefore, using the settable coefficients α and β (for example, α = 1.5, β = 0.3) for the average value of a that is a value when the fluctuation is in an intermediate state, the sample in FIG. Assuming that the average value of the interval trend variation amount a in the number N1 is am <N1>, the absolute value of the trend variation amount a obtained in the sample number Ni of the supply weight error during the operation operation appears large, and the upper threshold value α · am <N1>a> α · am <N1>
If the number of samples Ni is not good, the control followability is not good, and the number of samples is changed to Ni-1, and an error variation function is obtained again to predict the next supply error.

However, when the section average standard deviation in the sample number N1 in FIG. 6 is σm <N1>, the standard deviation σ calculated in the section of the sample number Ni is relative to the threshold γ · σm <N1>.
σ> γ · σm <N1>
In other words, that is, when the random variation of the supply weight is too large, for example, when γ = 2 to 3 and the above formula is satisfied, the number of samples is not reduced.

In addition, the absolute value of the trend fluctuation amount a obtained in the sample number Ni of the supply weight error during the operation operation appears small, and the lower threshold β · am <N1>
a <β · am <N1>
If the number of samples Ni is sufficient, the control followability is sufficiently good. Therefore, in order to further improve the smoothness of the random disturbance, the number of samples is changed to Ni + 1, the error variation function is obtained again, and the next supply error is predicted. The feeder operation amount of the feeder 4 is calculated.

  The threshold values α · am <N1>, γ · σm <N1>, β · am <N1> are set together with the initial value of the number of samples N1 as described above, and are shipped as products. The user may perform adjustment operation to change the settings of the threshold values α · am <N1>, γ · σm <N1>, β · am <N1>.

  Next, the operation at the time of the operation operation by the user who automatically varies the number of samples will be described based on the flowcharts of FIGS. 11 and 12. This process is performed for each weighing hopper 6 as described above.

  First, in step n101, it is determined whether or not the operation operation is turned on. If not, the shift register is cleared and the process ends (step n113).

  When the operation operation is ON, the weight value of the weighing hopper 6 is read (step n102), and it is determined whether or not the read weight value Wx is less than the near-zero detection level Wzt1 (step n103). When it is not less than the detection level Wzt1, the process proceeds to step n105. When it is less than the near-zero detection level Wzt1, the flag F0 indicating that the weighing hopper 6 has passed through the empty state is set to “1”, and the process proceeds to step n105 (step n104).

  In step n105, it is determined whether or not the flag F0 is “1”. If it is not “1”, the process returns to step n101, and if it is “1”, the process proceeds to step n106, and the read weight value Wx becomes the article. Is determined to be greater than the article placement detection level Wzt2, which indicates that is placed, and if not greater than the article placement detection level Wzt2, the process returns to step n101. In step n106, when the read weight value Wx1 is larger than the article placement detection level Wzt2, it is determined that the weighing hopper 6 has passed through the empty state, assuming that the article has been supplied after the weighing hopper 6 has passed through the empty state. The flag F0 shown is cleared to “0” (step n107), the stable weight value is read again after the time of the stabilization wait timer has elapsed (steps n108 and n109), and in step n110, the shift register is shifted right and updated Is stored at the left end.

  Next, whether the control can be started or not, specifically, the number obtained by adding one to the sample number Ni in the shift register, that is, (Ni + 1) pieces, so as to cope with the case where the number of samples is increased, that is, the operation operation is started. At the time of initial initialization, it is determined whether or not there are (N1 + 1) data (step n111). If there are no (Ni + 1) or more data, the process returns to step n101, and (Ni + 1) or more data is stored. In some cases, data that can be subjected to control calculation is available, so the standard deviation of the Ni weight value data is obtained, and the error variation function is obtained from the latest Ni data. Further, the tendency variation amount a is obtained (step n112), and the process proceeds to step n114 in FIG. This tendency variation amount a has polarity.

  Next, in step n114 of FIG. 12, it is determined whether or not the trend variation amount a obtained in step n112 is larger than the upper threshold value α · am <N1>. Then, the process proceeds to step n119, and when it is large, it is determined that the tendency variation is large, and the process proceeds to step n115. In step n115, it is determined whether or not the standard deviation σ obtained in step n112 is larger than the threshold value 2 · σm <N1>. Since it is large, the random number variation is prioritized over the tendency variation, and the number of samples Ni is increased by one and the process proceeds to Step n125 (Step n118).

  In step n115, when the standard deviation σ is not larger than the threshold value 2 · σm <N1>, the random variation is not large and the trending fluctuation is large. Therefore, the number of samples Ni is decreased by one and the process proceeds to step n117 (step n117). n116). In step n117, it is determined whether or not the sample number Ni is not less than the lower limit value NL. If the sample number Ni is not not less than the lower limit value NL, the process proceeds to step n118, the sample number Ni is incremented by 1 and returned to the original step. Move to n125. In step n117, when the number of samples Ni is not less than the lower limit NL, the process proceeds to step n123.

  In step n114, when the trend variation amount a is not larger than the upper threshold value α · am <N1>, the process proceeds to step n119, and whether or not the trend variation amount a is smaller than the lower threshold value β · am <N1>. If it is not smaller than the lower threshold β · am <N1>, the process proceeds to step n125.

  In step n119, when the trend variation amount a is smaller than the lower threshold value β · am <N1>, the trend variation is small, so that the number of samples Ni is increased by one in order to smooth the random variation. To step n121 (step n120). In step n121, it is determined whether or not the sample number Ni is less than or equal to the upper limit value NU. If it is not less than or equal to the upper limit value NU, the sample number Ni is reduced by 1 and returned to the original state (step n122).

  In step n121, when the number of samples Ni is less than or equal to the upper limit value NU, the process proceeds to step n123, where the standard deviation of the data for the new number of samples Ni is obtained, and the error variation function is obtained for the latest Ni pieces of data. The amount a is obtained and the process proceeds to step n124. In step n124, the value of the new sample number Ni is transmitted to the main calculation control unit 17, and the section average standard deviation and the trend variation amount a are transmitted to the main calculation control unit 17 for display, and the process proceeds to step n125.

  In step n125, the trend variation amount a calculated from the data of the number of samples Ni up to that time, without changing the number of samples in step n112 of FIG. 11, or the new variation changed in step n123 of FIG. The next predicted supply weight error e ′ is obtained based on the trend variation amount a calculated from the data of the number of samples Ni. The next supply weight value y is determined based on the calculated predicted supply weight error e ′, the feeder operation amount Yx of the linear feeder 4 is calculated, the feeder operation amount Yx is transmitted to the main calculation control unit 17, and the process returns to step n101. (Step n126). The main arithmetic control unit 17 controls the supply amount of articles by the linear feeder 4 by the feeder operation amount Yx.

  As described above, during operation, the sample number Ni is varied as necessary based on the trend variation amount a, and the next predicted supply weight error e ′ is determined by the trend variation amount a calculated from the sample number Ni. The next supply weight value y is calculated so as to cancel the predicted supply weight error e ′, and the feeder operation amount Yx of the linear feeder 4 is calculated and controlled. An appropriate number of samples can be selected and properly tracked and controlled.

  In the above-described embodiment, the error variation function indicating the gradual variation in the weight error of the articles supplied to the weighing hopper 6 is obtained using the least square method, but may be obtained using another method. .

  For example, in the section of the number of N1 samples, the average value of the supply weight error in the first half section from 1 to (N1 / 2) and the average value of the supply weight error in the second half section from (N1 / 2) to N1 A straight line connecting the two may be an error variation function.

  Further, as shown in FIG. 13, for example, when N1 = 9 supply weight errors e1 to e9 occur, an average value of nine fluctuation amounts (slopes) a1 to a9 between adjacent errors is expressed as a. The average value of the supply weight errors e0 to e9 is E, and the error variation function e has a straight line of the average variation amount a through (4.5, E) by the average value 4.5 of the nine error acquisition timings. You may ask for.

  As another embodiment of the present invention, a memory hopper to which articles weighed by the weighing hopper 6 are supplied may be provided, and the combination calculation may be performed based on the weight of the weighing hopper 6 and the articles of the memory hopper.

(Embodiment 2)
In the above control, the articles supplied from the linear feeder 4 to the supply hopper 5 are supplied to the weighing hopper 6, and the number of samples is changed based on the measured value measured by the weighing hopper 6. The supply amount by the linear feeder 4 was controlled so that the weight of the article to be supplied became the target supply weight.

  On the other hand, as another embodiment of the present invention, each supply hopper 5 in FIG. 1 is also provided with a weight sensor 11 so that the measured value of the article supplied to each supply hopper 5 can be measured. Instead of the weighing value of the weighing hopper 6, the weighing value by the supply hopper 5 may be handled in the same manner as the weighing value by the weighing hopper 6 to control the supply amount of articles by the linear feeder 4. In this case, the target supply weight of the articles to the supply hopper 5 is the same as the target supply weight of the weighing hopper 6.

  When the measurement value of the weighing hopper 6 is used for the control of the linear feeder 4, the article supplied from the linear feeder 4 is temporarily retained in the supply hopper 5 at a certain timing, and the article is supplied to the weighing hopper 6. When the article is newly supplied from the linear feeder 4 to the supply hopper 5 at the next timing, the measurement value of the supply amount by the linear feeder 4 is measured by the measurement value of the article supplied to the weighing hopper 6, and this measurement is performed. Since the control operation amount calculated based on the value is further reflected in the supply amount of the article supplied from the linear feeder 4 to the supply hopper 5 next, in the feedback control, as shown in FIG. An extra control response delay (dead time) occurs for one supply.

  Therefore, as in the present embodiment, the supply hopper 5 is also provided with the weight sensor 11 so that weighing can be performed, and the operation amount of the linear feeder 4 is calculated based on the measured value of the article in the supply hopper 5, and the calculated operation is calculated. If the supply amount by the linear feeder 4 immediately after that is controlled according to the amount, there will be no extra delay in the feedback control response once compared to the method using the measurement value of the weighing hopper 6, so that the supply amount can be controlled more accurately. Become.

(Embodiment 3)
In the above embodiment, the linear feeder 4 is controlled based on the tendency variation amount, but the present invention is based on the average value of the supply weight of the articles supplied to the weighing hopper 6 as in the above-mentioned Patent Document 1. The linear feeder 4 may be controlled. That is, instead of the trend variation amount a of the above embodiment, an average value of the supply weight error based on the number of samples at that time may be used.

The table of FIG. 6 also calculates and displays the maximum value, the minimum value, and the average value for the average value of the supply weight error for each number of samples (section average supply weight error) instead of the trend variation amount during the adjustment operation. .

  The error interval average supply weight error appears larger as the number of samples decreases, and decreases as the number of samples increases.

  If the variation in the supply weight error is small, it is necessary to increase the number of samples in order to minimize the random variation.

  On the other hand, if the variation in the supply weight error is large, it is necessary to reduce the number of samples and speed up the control response unless the random variation is large.

  Based on the table in which the trend variation amount is replaced with the average value of the supply weight error, the initial value of the number of samples is set as in the above embodiment.

  Further, based on the table, an upper threshold value and a lower threshold value for the average value of the supply weight error are set in advance.

  During operation, the average value of the supply weight error is controlled so as to be canceled, and when the average value of the supply weight error appears large and exceeds the upper threshold, the random variation amount is not large. Reduce the number of samples to speed up the control response and allow the control to follow the fluctuations, while if the average value of the supply weight error appears small and falls below the lower threshold, to reduce the effects of random variations Increase the number of samples.

  This makes it possible to perform optimal control according to the state of occurrence of supply weight error in the system, and to improve the combination accuracy by making the supply weight of each weighing hopper 6 close to the target supply weight.

  Others are the same as that of the said embodiment.

(Other embodiments)
In each of the above embodiments, the articles supplied from the linear feeder 4 to the supply hopper 5 are supplied to the weighing hopper 6, and the number of samples is changed based on the measured value measured by the weighing hopper 6. The supply amount by the linear feeder 4 is controlled so that the weight of the article supplied to the target supply weight becomes the target supply weight, or a weight sensor 11 is added to the supply hopper 5 so that the supply hopper 5 can be weighed. The article supplied from the linear feeder 4 to the supply hopper 5 is measured by the supply hopper 5, and the number of samples is changed based on the measured value measured by the supply hopper 5. The supply amount by the linear feeder 4 is controlled so that the weight becomes the target supply weight. However, as another embodiment of the present invention, the measured value and meter of the article measured by the supply hopper 5 are used. By the linear feeder 4 so that the weight of the article supplied to the supply hopper 5 or the weighing hopper 6 becomes the target supply weight by changing the number of samples based on the measurement value of the article measured by the hopper 6. The supply amount may be controlled. In this case, for example, the number of samples may be changed based on the measured value measured by the supply hopper 5 and the average value of the measured value measured by the weighing hopper 6.

1 Combination Weigher 2 Feeder 3 Main Feeder (Distributed Feeder)
4 Linear feeder (straight forward feeder)
4a Linear feeder trough 4b Exciter 5 Supply hopper 6 Weighing hopper 9 Control device 11 Weight sensor 12 Operation setting indicator 16 Individual calculation unit 17 Main calculation control unit 18 Storage unit 19 Vibration control circuit unit 21 Control unit

Claims (8)

A plurality of feeders each supplying articles, and a plurality of weighing hoppers each weighing the weight of the supplied articles while the articles are respectively supplied via the plurality of feeders,
In a combination weigher that performs a combination calculation based on the weight of the article and a target combination weight that are respectively weighed by a plurality of weighing hoppers,
Predicted value calculation means for estimating and calculating a predicted value related to the supply weight of the weighing hopper based on the supply weight of the article for two or more samples supplied to the weighing hopper;
A sample number varying means for varying the number of samples based on a supply weight of the article supplied to the weighing hopper;
Based on the predicted value of the weighing hopper estimated and calculated by the predicted value calculating means, the supply weight of the article supplied to the weighing hopper is defined based on the target combination weight. so that the target supply weight, and control means for controlling the supply amount by the feeder,
The number of samples is 3 or more,
The predicted value calculation means estimates and calculates a change amount of the supply weight per sample number as a trend variation amount based on the supply weight of the sample supplied to the weighing hopper. In addition, based on the trend variation amount, the predicted value related to the supply weight reflecting the trend variation amount is estimated and calculated.
Combination scale. A plurality of feeders that respectively supply articles, a plurality of supply hoppers to which the articles are supplied via the plurality of feeders, and the articles are supplied and supplied from the plurality of supply hoppers, respectively. A plurality of weighing hoppers for weighing each of the articles,
In a combination weigher that performs a combination calculation based on the weight of the article and a target combination weight that are respectively weighed by a plurality of weighing hoppers,
Each of the plurality of supply hoppers can respectively weigh the supplied article.
Predicted value calculation means for estimating and calculating a predicted value related to the supply weight of the supply hopper based on the supply weight of the article for two or more samples supplied to the supply hopper;
A sample number varying means for varying the number of samples based on a supply weight of the article supplied to the supply hopper;
Based on the predicted value of the supply hopper estimated and calculated by the predicted value calculation means, the supply weight of the article supplied to the supply hopper is defined based on the target combination weight. so that the target supply weight, and control means for controlling the supply amount by the feeder,
The number of samples is 3 or more,
The predicted value calculation means estimates and calculates a change amount of the supply weight per sample number as a trend variation amount based on the supply weight of the number of samples of the supply weight of the article supplied to the supply hopper. In addition, based on the trend variation amount, the predicted value related to the supply weight reflecting the trend variation amount is estimated and calculated.
Combination scale. The sample number varying means varies the sample number based on a supply weight of the article supplied to at least one of the weighing hopper and the supply hopper and the target supply weight.
The combination weigher according to claim 1 or 2. The predicted value calculation means is a predicted value of an error from the target supply weight of the supply weight of the article supplied to at least one of the weighing hopper and the supply hopper as the prediction value related to the supply weight. Estimate
The combination weigher according to any one of claims 1 to 3. A plurality of feeders each supplying articles, and a plurality of weighing hoppers each weighing the weight of the supplied articles while the articles are respectively supplied via the plurality of feeders,
In a combination weigher that performs a combination calculation based on the weight of the article and a target combination weight that are respectively weighed by a plurality of weighing hoppers,
Predicted value calculation means for estimating and calculating a predicted value related to the supply weight of the weighing hopper based on the supply weight of the article for two or more samples supplied to the weighing hopper;
A sample number varying means for varying the number of samples based on a supply weight of the article supplied to the weighing hopper;
Based on the predicted value of the weighing hopper estimated and calculated by the predicted value calculating means, the supply weight of the article supplied to the weighing hopper is defined based on the target combination weight. Control means for controlling the supply amount by the feeder so as to achieve a target supply weight ,
The predicted value calculation means is configured to estimate and calculate, as the predicted value, an average value related to the supply weight of the weighing hopper based on the supply weight of the article for the number of samples supplied to the weighing hopper,
A standard deviation of the supply weight of the article supplied to the weighing hopper is calculated for each different number of samples, and the supply is a difference between the supply weight of the article supplied to the weighing hopper and the target supply weight A standard deviation and supply weight error calculating means for calculating an average value of weight errors;
Display means for displaying an average value of the standard deviation and the supply weight error together with the number of samples;
A combination weigher. A plurality of feeders that respectively supply articles, a plurality of supply hoppers to which the articles are supplied via the plurality of feeders, and the articles are supplied and supplied from the plurality of supply hoppers, respectively. A plurality of weighing hoppers for weighing each of the articles,
In a combination weigher that performs a combination calculation based on the weight of the article and a target combination weight that are respectively weighed by a plurality of weighing hoppers,
Each of the plurality of supply hoppers can respectively weigh the supplied article.
Predicted value calculation means for estimating and calculating a predicted value related to the supply weight of the supply hopper based on the supply weight of the article for two or more samples supplied to the supply hopper;
A sample number varying means for varying the number of samples based on a supply weight of the article supplied to the supply hopper;
Based on the predicted value of the supply hopper estimated and calculated by the predicted value calculation means, the supply weight of the article supplied to the supply hopper is defined based on the target combination weight. Control means for controlling the supply amount by the feeder so as to achieve a target supply weight,
The predicted value calculation unit, before SL based on the supplied weight of the article of the number of samples to be supplied to the supply hot path, the average value for the supply weight before Symbol supply hot path is estimated calculated as the predicted value Is,
The standard deviation of the supply weight of the article supplied to the supply hopper is calculated for each different number of samples, and the supply is the difference between the supply weight of the article supplied to the supply hopper and the target supply weight A standard deviation and supply weight error calculating means for calculating an average value of weight errors;
Display means for displaying an average value of the standard deviation and the supply weight error together with the number of samples;
A combination weigher. A display means for displaying the trend variation amount estimated and calculated by the predicted value calculation means together with the number of samples for each different number of samples;
The combination weigher according to claim 1 or 2 . Standard deviation calculating means for calculating a standard deviation of a supply weight of the article supplied to at least one of the weighing hopper and the supply hopper for each different number of samples,
The display means displays the standard deviation calculated by the standard deviation calculation means together with the number of samples.
The combination weigher according to claim 7.

Images