JPS59107216A - Combinational weighing method - Google Patents

Abstract

PURPOSE:To select weighing machines of an optimum combination, by calculating the number of weighing machines with which the combinatorial operation is started on a basis of the weight of articles supplied to each weighing machine and generating a combination pattern which selects j-number of weighing machines from N-number of weighing machines. CONSTITUTION:A combination pattern generating part 12 generates the bit pattern of all ''1'' in response to a request signal RS outputted from a selecting head number control part 13 and outputs only a bit signal, where j-number of bits out of N-number of bits are set to ''1'', in accordance with the signal which instructs the part 12 to select j-number of weighing machines from N-number of weighing machines. After completing the combinatorial operation, a driving control part 6 outputs a start signal to a weighing hopper 2d and a pool hopper 2b in the same weighing section as the hopper 2d at shifted timings in accordance with the result of an optimum combination outputted from a combinatorial operation part 8 and outputs a supply enable signal to a distributing supply device 2a. Upper and lower limit setters 10 and 11 store a desirable range of the weighed value.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a combination weighing method, and relates to a combination weighing method that allows a combination in which the total value of combinations is within the set tolerance range and closest to the set value to be selected in a short time.

A weighing machine is composed of a weighing hopper and an attached weight detector, and the weighing machine is composed of a weighing hopper and an attached weight detector.
The total weight ffi value of each product is decreased by 1, and the total weight ffi value that is equal to the set value or is closest to the set value within the set value is selected. After that, a new item is put into the weighing popper from which the item was ejected, and it is used for the next combination, and the same operation is repeated thereafter to achieve high-precision automatic weighing. Combinatorial weighing methods that continue are already known.

Figures 1 and 2 are explanatory diagrams of a weighing device of a combination scale that realizes the i-+- quantity method.
The figure is a block diagram of the combination control section.

In FIG. 1, reference numeral 1 denotes a main supply device that distributes the articles radially and supplies the cough articles to a plurality of radially arranged cough articles. The main supply device a1 has a vibrating conveyance type structure and can disperse and supply articles to the receiving section by vibrating for a predetermined period of time.

2.2. . . is a total of 1 section of n units arranged radially around the main feeding device, each of which is a distributed feeding device 1iffi2a, pool hopper 2b, pool hopper day 1-2C, weighing hopper 2d, It has a weight detector 2e and a weighing hopper game) 2f. The distributed feeding device 2a is composed of a feeder-type conveying device that can be individually vibrated or a seven-layer device that can be operated individually, and each distributed feeding device 2a
The structure is such that articles can be thrown into a pool hopper 2b located below the pool. Each pool hopper 2b is provided with an f-L weighing hopper 1-2C. Each weighing machine is composed of a weighing hopper 2d and an attached weight detector 2e, and if the article is put into the weighing hopper 2d of the weighing machine, is measured and input to the combination control unit (Fig. 2).Then, as described later, the combination control unit determines the whirling weight that is equal to the set value or is within the set allowable range and is closest to the set value. The optimal combination of scales giving the value is selected.

In addition, each weighing hopper 2d is provided with a weighing hopper gate 2f, and only the weighing hopper gate 2f provided in the weighing hopper that provides the optimal combination is controlled by the drive control unit (Fig. 2). Unopened items will be ejected by the signal. Reference numeral 3 denotes a collection chute that works to collect the articles discharged from each weighing hopper 2d in the lower central part, and has a conical or polygonal funnel shape, and is operated by the weight of the articles or by a forced scraping device (not shown). Items discharged to the inner circumferential area by
is a timing hopper, which temporarily holds the M articles collected by the collecting unit and discharges the held articles out of the weighing device in response to a discharge signal output from an external device such as a packaging machine.

FIG. 2 is an example of a block diagram of the combination control section. In the figure, 5 is a combination control unit, and 5a is an n-bit counter that counts timing pulses TP of a predetermined period, and generates a combination pattern of n weighing hoppers. In other words, the combination of taking out one item from n weighing hoppers is nJth, and the combination of taking out two items from n weighing hoppers is n (n -1) / 2.
In general, the combination of r items left out of n items is n.
(n-1) (n-2)...(n-r +1)/r 1
Therefore, the total number of combinations is 2-1. Therefore,
When the binary n-bit counter 5a counts 2+L-1 timing pulses TP, it becomes 000...001 to 11.
Since a total of 27-1 types of bit patterns (1...111) are generated, the first bit is used as the first weighing machine, and the second
If the bits are made to correspond to the second total t 8 l, and the nth bit is similarly made to correspond to the nth weighing machine, the bit pattern will represent a combination pattern. 5b is a multiplexer, which selects the weight detector 2 provided in a predetermined weighing machine according to the bit pattern of the count value of the counter 5a.
The measured value of e (heavy weight value of the object to be weighed) is output to the subsequent calculation section. For example, if n=10, the count value is 100010
1011, the first. 2nd, 4th. 6th and 1st
The output w of the weight detector 2e attached to the 0th weighing machine,
, w2. w4. w, , wlo are output to the arithmetic unit. 5C is a setting value register for storing settings (Qj Wa), 5d is an upper limit setter, and 5e is a lower limit setter, which stores the preferred setting tolerance of the measured value (upper limit value Ma and lower limit value M+) f!:. Note that the lower limit value is customarily set equal to the set value.This is because if the lower limit value is set below the set value, the total weight may be less than the displayed weight, which may cause complaints71). 5f is a calculation unit that calculates and outputs the total weight value ΣW+ (=X) outputted from the multiplexers 5b to S, calculates the difference between the total weight value ΣW+ and the set value Wa, and outputs its absolute value. be. That is, the calculation unit 5f performs the calculation of so-called 1 = Total weight value ΣWI
and the set value Wa (hereinafter simply referred to as deviation value) A is output. 5g is a comparator, and 5h is a dosage counter. The comparator 5g determines whether the total weight value X is within the range of M1 to Ma, and if output≦X≦Ma・concave (5), counts up the count value of the appropriate amount counter 5h. 5j is a minimum error memory that stores the minimum deviation;
The deviation value A is automatically set only the first time, and its contents are updated in the manner described later. Note that when the lower limit value is made equal to the set value, the difference between the upper limit value and the set value may be initially set in the minimum error memory 5j. 5 is an optimum combination memory lJ that stores an optimum combination pattern, 5m and 5n are gates, and 5p is a deviation value A and a minimum error memo I which are the outputs of the calculation unit 5f when the total weight value ΣWt is within the set tolerance range.
It is a comparator that compares the magnitude of the minimum deviation value B up to the cutting edge stored in J 5 j. When A (B), the deviation value A is stored in the minimum error memory 5J via the gate 5m, and The contents (combination pattern) of 5a are stored in the optimum combination memory 5k. 6 is a drive control section, and when the contents of the appropriate amount counter 5h is 1 or more, a predetermined weighing hopper game) 2f (first
1λ) is opened to discharge the articles from the weighing hopper 2d into the collection chute 3, and the pool hopper gate 2C is opened to re-throw the articles into the weighing hopper 2d from which the articles were discharged.

The operation of this measuring device will be briefly explained. At the initial stage, articles are loaded into each pool hopper 2b and each weighing hopper 2d. A weight detector 2e provided in each weighing machine measures the weight of one article and sends the weight values W1 to W1o to the combination control section 5. Now, the n-bit counter 5a (n=
to) counts the timing pulses TP of a predetermined period, and
The barrel is configured to generate n-1 combination patterns. That is, by counting the first @th timing pulse TP, the content becomes 0000000001. As a result, the output value (weight value Ws) of the first weighing machine is outputted from the multiplexer 5b to the performance section 5f.
The calculation unit 5f executes calculations of equations (1) and (2),
Combination a-weighted leaf value ΣW1. It also outputs the deviation value A (-1W1-Wa l) between the total weight value Σwi and the set value Wa.
Ru. Since 5n and 5m are automatically opened only during the first combination calculation, the deviation value A is the minimum error memo IJ 5
At the same time, the contents of the n-bit counter 5a (m combination patterns 000...01) are stored in the optimum combination memo IJ5k. Also, the comparator 5g compares the total weight value ΣW+(=X) with the upper limit value Ma and the lower limit value M1, and if it is greater than Mt(X(Ma), it counts up the content of the appropriate amount counter 5h. If the timing pulse TP occurs, the contents of the counter 5a (combination pattern) become 00000.
It becomes 00010. The output value of the weight detector 2e provided in the second weighing hopper (weight value W-
) is output to the calculation unit 5f, and the calculation unit 5f outputs (1), (2
), the total weight value C1 (=X) and the deviation value A
(=Iwi-Wal) is output.

The comparator 5g determines whether formula (3) is satisfied or not, and if it is satisfied, increments the content of the appropriate amount counter 5h by one. On the other hand, the bias device 5p compares the deviation value A with the contents B (=IWt-Wal) of the minimum error memo IJ 5 j, and determines the contents of the minimum error memory 5j and the optimum combination memo IJ 5 H in A2B. Without updating A(B, so to speak, the deviation IaA is stored in the minimum error memory 5j, and W 'c f & appropriate combination memo IJ 5 in the n-bit counter 5a.
Transfer and store them respectively in k. Then, the same operation is repeated thereafter, and finally only the minimum deviation value among all the 2-1 combinations is stored in the minimum error memory 5j, and the combination pattern at that time is stored in the optimal combination memo IJ5k. The optimal combination pattern is selected. Count value of appropriate amount counter 5h and optimal combination memo IJ-5
Stored in k21. The combination pattern is the drive control unit 6.
is input. If the count value of the appropriate amount counter 5h is 1 or more, the drive control unit 6 opens the counting hopper gate 2f of the predetermined weighing machine based on the inputted matching pattern, and transfers the article to be weighed to provide the optimum combination to the weighing hopper. The collected 7 utes and 51 utes were discharged from 2d, and then the pool hopper gate 2
c is opened and the article is put into the discharged weighing hopper 2d. The dispersion supply device 2a corresponding to the empty pool hopper is vibrated for a certain period of time, and the articles are put into the empty pool hopper 2b. Thereafter, the same optimal combination control as described above is performed in line 1) with continuous weighing using the v1 combination scale. Incidentally, when the content of the appropriate amount counter 5 tL- is zero, the article is not discharged, but a predetermined weighing machine is replenished with the article and the combination calculation is performed again.

As described above, the combination weighing method is very useful in that it is possible to obtain and discharge a combination of weighing machines that gives a total IT value closest to the set value within the set allowable range. However, in such conventional methods, combinatorial calculations are performed for all 2n-1 combinations. In other words, combinatorial calculations are also performed for combinations that give a total weight that is much higher than setting 11. Therefore, a considerable amount of calculation time was required to obtain the optimal combination.

Therefore, it is an object of the present invention to provide a 2''-i) 15 combination measurement method that does not require the risk of performing a river matching operation JT for all 15 combinations.

Another object of the present invention is to provide a combination weighing method that can produce an optimal combination of weighing machines with fewer combination operations.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings, but before that, an outline of the method of the present invention will be explained. The combined dosing device has a total of N units, and each weighing machine has a weight W.
I (+=1.2...N) items are being thrown in,
If the force 1 is also set and the value ff1-U is Wa, the average weight Wm of n articles fed into one weighing machine is Wm= (, E W+ ) /N [4
1 blind = 1, and the number of weighing machines that give the set weight value Wa is (W a
/Wm) is extremely likely to be in the vicinity. Therefore,
In the present invention, (Wa/Wm
) and round off the decimal point of the calculation result (it may be rounded down to 4+) to give the optimal combination. r! Consider i, j, N
A weighing machine that continuously generates combinations of selecting j weighing machines from the set of weighing machines C, and gives a total weight value that is closest to the set value within the set tolerance range after all combinations by j weighing machines C are completed. We are trying to get rid of more items. In addition, if there is no optimal combination within the set tolerance range among all the combinations in which j weighing machines are selected from N machines totaling t1, the total weight of each combination is greater than the set tolerance range (over j k ) or below the set allowable range (light weight), the number of combinations that are overweight and the number of combinations that are light weight are counted as the number of accepted units HW and the number of lightweight units LW, respectively, and when HW) LW, (j-1) units. , also H
In the case of W (LW), (j+1) units are regarded as the number of weighing machines that provide the optimal combination, and the combination calculation is performed again to obtain the optimal combination within the setting allowable range.

FIG. 3 is a schematic block diagram of the present invention, and the same parts as in the conventional example are given the same reference numerals.

The combination pattern generating section 12 generates a bit pattern of "all pars 1" in response to the request signal R5 outputted from the selection head number (number of weighing machines) control section 15, and in accordance with the signal to select J out of N heads, N. Outputs only the bit signal of j bits i KL 1'' among the bits.

Note that the initial value of j is j = (Wa / (,ΣWi/N)) 45 if the rounded value of the calculation result is expressed as ()45.
(5) It is given by 1 (hereinafter referred to as the number of priority combinations) and is changed according to the result of the combination calculation. After the combination calculation is completed, the drive control unit 6 shifts the timing of the weighing hopper 2d participating in the combination and the pool hopper 2b in the same weighing section as the weighing hopper 2d, according to the optimal combination result output by the combination calculation unit 8. Each outputs an open signal, and also outputs a supply enable signal to the distributed supply device 2a of the same metering section.

The upper limit setter 10 and the lower limit setter 11 each store a preferred range of weight values, and are digital switches or storage devices used when determining the distribution tendency of combined weight values, which will be described later. The setting device 9 sets the set value Wa,
It is a storage device and outputs each to the combination calculation section 8.

FIG. 4 is a block diagram of the combination calculation section. In FIG. 4, the gates 8a-i, 8a-2...8a-n are connected to the weight detectors 2e, 2e...the weight direct signals (Wl, W,...
...Wn) and the multiplexer 8b, and connects the weight detector 2e corresponding to the bit '1#' to the multiplexer 8b in accordance with the bit signal output from the combination pattern generating section 12.

The multiplexer 8b sequentially sends the weight 1 direct signal that has passed through the gate to the calculation section 8C. During the combination calculation, the calculation unit 8c adds the weight values output from the multiplexer 8b, determines the deviation value A (see equation (2)) between the total weight value X and the contents of the setting device 9, and calculates the difference by using the comparator. 8f and gate 8d, and outputs the weight total value X to comparator 8g. Comparator 8
f compares the content B of the minimum error memory 8e with the deviation value A sent from the calculation unit 8C, and if the deviation value A is smaller than the content B of the minimum error memo IJ 8e, the game 1-8d
, and outputs an open signal to gate 8k. From this point, the deviation value A at that time is determined as the minimum error memo I through the gate 8d.
J8e is entered, and the contents of the minimum error memo IJ8e are updated. Further, the bit signal of the combination pattern at that time is outputted from the combination pattern generating section 12 to the game 1-8k, and the contents of the optimal combination memo +)Bl are also updated. The first time of the combinational operation is gate 8d1 (gate 8).
Both are open, and the first difference A1 combination pattern 00.
...01 are respectively stored as the minimum error memo IJQe and the optimum combination memo +)8tI. The comparator 8g receives signals from the upper limit setter 10 and the lower limit setter 11, and compares the total compensation value X with the upper and lower limit values, performs rank-based processing, and Counter 8h % steel rod counter 81° Excess amount counter 8j is counted up. Incidentally, the ranking is performed as follows. If the total weight value is x1, the upper limit set value is M a, and the lower limit setter is MI, then Mi) When X... Light cost M t
(X (When M a... Appropriate amount M a (
When X...... Follows the overdose.

Selective head? The ati control section 15 instructs the combination pattern generation section 12 how many of the N weighing types to select. As shown in FIG.
31, a priority combination number update unit 152, and a control unit 153, and the priority combination number calculation unit 131 calculates the number of weighing machines R (R is normally the total number of weighing machines N, but it decreases if there is a malfunctioning head. register 131a in which the value (not constant) is set
and the weight value WI(I
"" 1j 2#...N) to detect whether an article is loaded into the weighing machine, and output an empty signal ES when no article is loaded into the weighing machine due to zero point correction or other reasons. 131b, a counter 131C that counts the number of occurrences of the empty signal ES and outputs the number M of weighing machines in which no articles are loaded, and N' = R-M.
+61 calculation'X is performed, and the weighing machine in which the article is loaded (output of calculation unit 8C) and N' are set (iWa). ”, and the gate 131f center, t
! It has a register 131g that stores the number j of previous combinations. On the other hand, the invader combination number update unit 132 updates the number of preferred combinations when there is no optimal combination within the set tolerance range among all combinations for selecting j machines in total from N' weighing machines. If an optimal combination does not exist, the overload number HW and the light weight number LW, which are the contents of the overload counter 8j and the light weight counter 8h, are compared in magnitude, and HW<LW
A comparator 132a outputs an up signal US when HW) and a down signal DS when HW)LW, and an up signal US="
a gate circuit 152b that outputs a numerical value +1 when the value is 1”;
Gate circuit 132C outputs -1 when down signal DS="1", register 131g, previous priority combination number j stored in IC and gate circuits 132b, 13
It has an adder/subtractor 32d that performs an operation of j+1 or j-1 using the output of 2C, and stores and updates the result of the operation in a register 131g via a game controller 131f.

Next, the present invention will be explained in detail. Prior to one combination performance, the number of selected heads IHIJ is controlled by the control unit 135 (fifth
From the figure), all bits are “1” in the combination pattern generation unit 12.
The request signal RS is applied to generate a pattern (111...concave 11). As a result, the combination pattern generating section 12 generates a bit pattern with all pits "1".
111...11". The multiplexer 8b sequentially outputs weight value signals w, , w, , W3...Wn through gates 8a-1,... (3a-n).
Play it! N, part 8C and selection head numeral (i11f'(S
15 and the sky detection unit 131b. The calculation unit 8C calculates the total weight value x (= ”I W+
) is calculated as 1;1 and inputted to the calculation section 131e, and the subtracter 131 is calculated and inputted to the calculation section 151e in the same manner. The calculation unit 131e uses X, N', and Wa to calculate the formula (5), sets the calculation result in the J% register 131g, and outputs the calculation result j to the 1 combination pattern generation unit 12 as the number of priority combinations. . After that, the combination pattern generator 1
2 is a measurement of N' units to a measurement of j units 4! ! Select N
``J bit patterns of all combinations are sequentially output, and the combination calculation section 8 starts the above-mentioned combination calculation. When the combination pattern flower arrangement section 12 finishes generating N/CJ total patterns, the conversion is performed. 1. When the combination calculation section 8 finishes the combination calculation for all N'Cj weighing machines from N' weighing machines 1 to j weighing machines, the combination pattern generation section 12 selects the appropriate quantity counter 8I. Take your time to read the content,
If it is larger than 1, the drive tttlJrm section 6ζ outputs a discharge enable signal DC5 to perform discharge and supply control.

That is, when the discharge enable signal DC5 is input, the drive control unit 6 outputs an open signal to the weighing hopper gate corresponding to the logic "1" bit of the optimal combination pattern stored in the optimal combination memo IJ8t. After loading the selected items into the collection chute from the selected weighing machine and loading them, an open signal is sent to the pool hopper gate corresponding to logic "1" in the #L suitable pattern, and the items are stored in the pool hopper. The article is put into the weighing machine, and the distributed supply device corresponding to the optimum pattern '1'' is further controlled to supply the article to the empty pool hopper.

On the other hand, as a result of the combination calculation for the priority combination number j, if the content of the appropriate amount counter 81 is zero (3L), the combination pattern generation unit 12 sends a new j output request signal NIR to the selected head number control unit 13. issue. The comparator 132a of the selected head number control section 13 reads the contents HW of the light weight counter 81], the excess amount counter 8j, according to an instruction from the control section 133.
Swallow the LW, compare the two, and if the content of the overload counter is greater, decrease j by 1, otherwise increase j by 1. That is, if HW>LW, the adder/subtractor 132d performs the operation (j-1), and if HW is less than LW, the adder/subtractor 132d performs the operation (j+1), stores the result of the operation in the register 131g, and generates a combination pattern. output to section 12. As a result, the combination pattern generator 1
2 selects (j+1) or 1-1) weighing machines from N' weighing machines N'Cj+1 or NICj-
The combination calculation unit 8 sequentially outputs the bit patterns of all combinations of , and performs the same combination calculation as described above.

In this way, the combination calculation is continued for each j, and all combinations from 1 to N' are exhausted until the content of the appropriate quantity counter becomes n1'' or more.As a result, the total weight value of the combination that falls within the quantity range is found among all the combinations. If the combination does not exist, it is determined that the combination is not established, and resupply and recombination calculations are performed.In addition, in the combination pattern generation unit 12, the combination pattern generator 12 selects j, (j+1), or 1- from the N′ weighing machines.
1) Regarding the method of generating all combination patterns for selecting weighing machines, please refer to Japanese Patent Application No. 1983, which has already been filed by the applicant.
See specification No.-107790.

As described above, according to the present invention, a combination of articles within a desired range can be obtained without performing all 2n-1 combination calculations, in other words, by a small number of calculations, and the calculation time can be significantly shortened.
It has become possible to speed up the weighing operation. In addition, in the present invention, each time a certain combination of total weight values is obtained, the light ht1. Ia-in-law, if there is no optimal combination among all the combinations in which j weighing machines are selected from N lead weighing machines by rank of weight, j is selected according to the size of the overweight and underweight numbers. Since it is updated to f J + 1 or j-1, even if the optimal combination does not exist among the combinations of selecting j units, ink: 111 + fil can be updated in a short time.
You can select one match.

Although described in detail above, the present invention is not limited to the embodiments and can be modified in various ways. For example, although the explanation has been made considering the case where no article is loaded into the weighing machine, if the article is loaded into all the needles tfa, the empty detector of the selected head number control section 15, the empty counter, Subtractors etc. can be omitted. Father, the present invention is configured by single-function hardware.
Although an embodiment has been shown, it may also be constructed using a microcomputer or the like.

[Brief explanation of drawings]

1 and 2 are explanatory diagrams of the conventional combination weighing method, FIG. 3 is a diagram for realizing the combination weighing method according to the present invention, and FIG. 5 is a block diagram of the selection head control unit according to the present invention. be. 1... Main supply device, 2... Measuring section, 2e.
...Weight detector, 5... Combination control unit, 6... Drive control unit, 8... Combination pattern generation unit, 9... Setting device, 10... Upper limit setting device, 11... Lower setting device, 1
2... Combination pattern generation unit, 13... Selected head number control unit, 131... Priority combination number calculation unit, 132
...Priority combination number update department Patent applicant: Ishidasha Ichiki Seisakusho Co., Ltd. Representative: Patent attorney: Minoru Tsuji (2 others)

Claims (3)

[Claims] (1) Combining the weighing data of N weighing machines and finding a combination where the combined total value is equal to the set value Wa or the value closest to the set value within the set tolerance range, and the set of weighing machines found In the combination weighing method in which articles are discharged from
1, 2, . . . N), a step of further calculating the number j of weighing machines to start the combination calculation g, and continuously generating combinations to select one weighing machine from N weighing machines. step, determining whether the total value of each combination by j weighing machines is within a preset tolerance range, storing the optimal combination within the set tolerance range, one weighing machine A combination weighing method comprising the step of ejecting an article based on the stored combination when there is an optimal combination existing in a set permissible @concave circle after all combination calculations have been completed. (2) A step in which the total value of each combination is within the set range or below the set range by rank of ηΔ, and for each rank -, excess ita, ili quantity, and sato 1 quantity 2. The combination weighing method according to claim 1, further comprising the step of counting. (3) When the number of suitable views is zero, the step of comparing the size of overexcited flattery and the light number LW, the step of decreasing j by 1 when history > LW in frequency and increasing j by 1 when V is LW , the step of performing a combinational operation based on the operation number t'I:jl.
2) Combination weighing method described in section 2).