JPH01170817A - Combined weighing instrument - Google Patents

Abstract

PURPOSE:To quickly execute a processing to find an optimum combination by exchanging plural weight data in a partial set with plural weight data in a complementary set and calculating the total value of weight data in the partial set and repeating this operation. CONSTITUTION:Weights W1, W2,...WN of articles weighed by plural weighing machines 1 are inputted to a multiplexer 2, and the multiplexer 2 successively outputs and stores weights W1, W2,...WN to and in an arithmetic control part 4 through an A/D converter 3 when the control part 4 receives the timing signal from a packaging machine 6 to output a switching signal. Weight data of (n) weighing hoppers are rearranged in accordance with data value, and a target value, an upper limit value, and a lower limit value are defined, and (r) pieces of weight data are taken out and the total value is calculated. When the arithmetic operation where a difference matrix is used to exchange r-number of or two elemential is successively advanced at this time, all combinational addition values different from one another are obtained, and the arithmetic processing time is shortened.

Description

Detailed Description of the Invention (Industrial Application Field) The present invention is a combination weighing method in which the weight of an object to be weighed obtained by a plurality of weighing machines is combined and calculated, and the optimal combination of weighing machines for a target value is selected. Regarding equipment.

(Prior technology) Objects to be weighed, such as loose confectionery and small metal fittings such as bolts and nuts, whose weight varies slightly, are distributed and fed to multiple weighing machines to optimize the target value. A combination weighing device is known in which a combination of weighing machines is selected, and an article is discharged from the weighing machines and sent to the next process such as packaging.

FIG. 2 is an explanatory diagram showing an example of such a combination weighing device. In the figure, articles to be weighed that have been transported by an incoming conveyor are supplied to a distribution table. N feeders F1 to FN%, pool hoppers P1 to PN, and weighing hoppers W1 to WN are arranged in a circle, and the weight of the objects to be weighed, which are fed from the distribution table to the weighing hopper via the feeder and pool hopper, is transferred to the load cell. Weigh with a weight detector such as A control device (not shown) performs combination calculations in a predetermined pattern based on weight data, selects the weighing hopper that provides the best combination for the target value, opens the gate of the weighing hopper, and discharges the articles to be weighed into the collection chute. .

When the number of weighing machines increases, this type of combined weighing becomes more difficult.
The number of combinations is (2N-1) when the number of weighing hoppers is N.
), which increases exponentially, resulting in a problem that the time required for arithmetic processing increases.

Therefore, in order to shorten the combinatorial calculation processing time, a combinatorial weighing device applying the difference matrix method was developed in JP-A-61
This is proposed in Publication No. 189419. This difference matrix method divides the entire set consisting of N elements into a subset PK
and a complementary set 7, and when the sum of the subsets included in PK does not converge to the target value, one element of PK is replaced with one element of 7.

■Place one element of PK in ■7.

Put one element of 0-7 into PK.

By repeating the process, the total value of the elements of px converges to the target value.

Next, a specific example of combination measurement using the above method will be explained.

(1) First, among the N weighing hoppers, M weighing hoppers are freely selected as initial settings and their total weight IW is calculated.

(2) Either the weight data of one of the selected weighing hoppers is replaced with (a) the weight data of an unused weighing hopper, (b) it is no longer used, or (C) a new unused weighing hopper is used. Among the three methods of additionally using the weight data of the weighing hopper, the method that best approaches the target value IP is selected, and the configuration of the weighing hopper used is changed.

(3) Calculate the total weight IW of the weighing hoppers used, and if the deviation from the target value is sufficiently small, the process ends there. Otherwise, the above-mentioned process (2) is repeated.

In addition, in step (1) of this algorithm, the M weighing hoppers initially selected may be set arbitrarily;
By repeating steps (2) and (3) several times, I
Since W converges to approximately the target value IP in a very small number of steps, at the point of convergence, the final combination of hopper sets is determined and the process ends. Furthermore, in the process (2), the calculation unit evaluates the degree to which IW approaches the target value I'P by searching for the following evaluation matrix A stored in the storage circuit. That is, the total number of weighing hoppers is N, and the number of initially set weighing hoppers in step (1) is M.
, the evaluation matrix A is given by the following equation.

...(1) Here, if Wl is the weight of weighing hopper i, then paddy and alJ are elements having the following values.

When i≠j, B, = Wj w+ (At this time, alj is selected without using the weighing hopper i that was used until then,
It means to use J. ) When i=j, a,,=
-W, (At this time, selecting all means that the weighing hopper i that was used up until then is not used.) In other words, the off-diagonal element alJ is the total weight ■ when switching from weighing hopper i to j. It can be used to find out how much W has changed and changed to the next IW. Furthermore, in order to add one new hopper, it is necessary to add Wl to IW in the previous step.

For this purpose, the following hopper weight vector is used with T as the transposed symbol. (w=[w+, W2,...WN]
T) If we note that M appropriate weighing hoppers have already been selected as the initial setting values as described above, (2)
In steps (3) and (3), it is not necessary to consider all N weighing hoppers.

That is, the evaluation matrix A and the hopper weight vector W can be expressed as follows.

Here, lines 1 to M are already selected weighing hoppers. That is, when a weighing hopper with i from 1 to M has already been selected, it is possible to know how the total quantity IW changes by reading out and examining only the elements within the dashed-dotted line from the memory circuit.

(Problem to be solved by the invention) Combinatorial calculations using such a difference matrix can significantly reduce calculation time compared to the conventional case of performing (2N-1) calculations from N pieces of data. Since data is exchanged one by one between the subset and complement set, there is a problem in that the calculation process still requires a considerable amount of time.

Moreover, in the conventional method, there is a difference of 1 between the subset and the complement set.
The number of combinations of subsets obtained in this way is very small, for example,
If the number of subset data is 4 and the number of complementary set data is 10, there are only 54 combinations of subsets that can be changed, and if there is no optimal combination among them, the combination of subset and complement I had to replace a lot of things.

Even if this replacement is performed, the same subset as before may occur, and in that case, unnecessary calculations will be performed. Furthermore, in order to avoid such duplication, it is necessary to check all the calculated combination patterns, but this poses a problem in that it takes more time to check.

Therefore, the present invention provides a combination weighing device aimed at solving the problems of the prior art.

(Means for solving the problems) In order to achieve the above object, the present invention is configured as follows. That is, means for storing N pieces of weight data obtained by a plurality of weighing machines divided into a subset and its complementary set, and exchanging a plurality of weight data of a subset with a plurality of weight data of its complement. It is characterized by comprising means for calculating the total value of the weight data of the set, and means for repeating the exchange of the weight data to find a combination in which the total value of the weight data of the subset is equal to or closest to the target weight. It is something to do.

(Operation) With the above configuration, all non-overlapping combinations can be investigated, and as a result, the time required for combination calculations is shortened, and the process of finding the optimal combination for the target value can be executed at high speed.

(Example) An example of the present invention will be described below. FIG. 1 is a schematic block diagram of a combination weighing device used in the present invention. In the figure, a plurality of weighing machines 1 have the configuration described in FIG. 2, and the weight Wl of the article weighed by each weighing machine.

W2. ...WN is input to multiplexer 2.

When the timing signal from the packaging machine 6 that wraps the articles discharged from the optimal combination of weighing machines is inputted to the arithmetic and control section 4, which is composed of a microcomputer, etc., the arithmetic and control section 4 sends the switching signal to the multiplexer. The weights W1 to WN are sequentially output from the multiplexer to the A/Di converter 3, and the weight converted to a digital signal by the A/D converter is stored in the arithmetic control unit 4. It is stored in a predetermined area of memory. The target weight set by the target weight setting section 5 is stored in a predetermined area in the memory, and the arithmetic control section selects the optimal combination of weighing machines for the target weight through arithmetic processing described later. , sends a signal to discharge the article from the weighing machine.

Next, arithmetic processing will be explained.

Take out some weight data of n weighing hoppers,
An easy way to search for the combination whose added value is closest to the target value is to find all the combined added values of nCr from r=1 to n and find the value closest to the target value. The number of times n is checked is 2°-1 times, and if n becomes large, the realistic allowable calculation time will be far exceeded.

Therefore, a method for reducing the number of calculations and checks by rearranging the weight data of n weighing hoppers in ascending order by using a method of calculating a combinational addition value using a difference matrix will be described.

Define the following variables here.

target: target value upper: upper limit value (deviation from target value) lower
r: Limitation of the range to be searched by the lower limit value (deviation from the target value)> If the elements of the set are rearranged in descending order so that Wi≧W i + 1 (i-1, 2・n-1), As shown below, nC
The range of calculation of r is limited.

The range of r in nCr is 'X W + ≧lower+targetII+ If we find the minimum j such that ΣWl≧upper+target, the range of r is j≦r≦n-
becomes +1.

For example, assume that the weight data of 10 weighing machines that can participate in combination is as follows.

Weighing machine No. (1) (2) (3) (4
) (5) Weight value (L) g 29723920420
32612152142202082'63 If you rearrange this in descending order, it will look like the following.

Sort order (1) (2) (3) (4)
(5) Weight value (Wl) g 297263261239
220 ([i) (7) (8) (9) (
10) Now, if the target value is 1000g, the upper limit is 10g, and the lower limit is Og, then the range of r of nCr is ΣWl ≧1000, so j=4, that is, ΣW, -297+263+261+23-106
0ΣW, since ≧1010 = 6, i.e. ΣW l-215421442084204420
3-10441=8 From the above, the range of r in nCr is 4≦r≦10-6+1, that is, r=4.5.

Therefore, in the case of the above example, 1°C4 and 10c
It is only necessary to calculate the combinations of 5.

<Method of Calculating the Combined Added Value of nCr> A method of extracting r pieces of weight data from a set of n pieces of weight data and calculating their total value using a difference matrix will be described.

(1) Calculation of initial value Calculate the total value tr of the weight data from the first weight data to the rth weight data.

tr= ΣW+ (2) How to use the matrix Of the difference matrix, what is needed for this calculation is 1=
=1 to r and j=r+1 to n, so only this part is calculated here.

Here, using the difference matrix, we can calculate the combined addition value (
Determining T) can be expressed mathematically as follows.

T=tr+a-t,-j, (1 element exchange) T=tr+a-jl・j+"a42・j2 (2 element exchange) T-tr"a'i+'jl"a-i2・j2÷・
・"a'fk'jk (exchange of r elements), where 1≦ik≦r, r+1≦jk≦1).

By performing such element exchange for all of +jk, it is possible to obtain all combinational addition values of ncr.

However, when adding two or more values of the difference matrix (
If you discard two or more initially selected elements and add as many initially unselected elements as you discarded)
It is sufficient to add only specific elements.

This is because, for example, adding a2.8 and 83.6 at the same time is equivalent to adding 82.8 and 83.8 at the same time. That is, both elements 2.3
This is because elements 6 and 8 are added by discarding .

Therefore, if the number of matrix elements to be added simultaneously is expressed by a variable rk, then when rk-1 to r are rkl, the calculation is T-tr+a-i-j (1≦i≦r, i+1
≦j≦n)rk2, the calculation is T-jr”a'f+'jl”a'f2'J2(1≦11
≦r−1, i, +1≦12≦r.

r+1≦jI≦n-1, J, +1≦j2≦n) rkr
When , the calculation is T-tr+alj++a2j2 +...-+arJr(
r+1≦j1≦n'-r+1, r+2≦j2≦n-
r+2.

...r+r≦Jr≦n) can be limited to.

As an example, N=10. The case of IOc 4 where r=4 will be explained with reference to Table 1.

Table 1 However, in Table 1, the vertical axis item i indicates data to be removed from the combination, and the horizontal axis item j indicates data to be included in the combination. For example, al, s are data 1 (
This means removing data 5 (N=1) and allowing data 5 (N-5) to participate.

In this example, the combination calculation will be performed according to the following procedure.

(1) From the 1st machine (N=1) to the rth machine (N=4 in this example)
) is added. That is, tw=Σ wi is determined.

(2) Next, there are the following 10 combinations (scs=10) in which the other four data are exchanged using al and s.

■tw+al, 5 +a26+a3.7 +a4. a■
tW+al, 5 +a2,5 +a3.7 +84,9
■tw+a,,5 +a2. e +83,7 +84.
+.

■tw+ a H5+ a2. a + a3. a
+ a4.9 ■tw+a+, s +a26 +83.
6 +84. +.

■tw+a1. s+a2. e +as9+a4. IQ
■tw+a',,5 +a2. ．． +a3,6 +a4
．． g■'tw+a, 5 +a2.7 +a3. a
+a4. .

■tw+a+', s +a2,7 +a3. g+8
4.1°(lj) tw+aI,s+a2. a +a
39+ 84. t.

In the above example, in the direction of the solid line arrow in Table 1,
By proceeding with the calculation of [phase], all non-overlapping combined addition values can be obtained.

(3) Similarly, when exchanging 3 pieces of data using al and s, there are 30 ways, 3C2+502-3×10=30 (4) When exchanging 2 pieces of data using al and S, sc+ +5C1=3X5=15 ways, al,
(1) When exchanging 4 pieces of data, 4C3=4 ways, ■tw+a16 +a2,7 +a, , a +a4.
g■tw+a, 6+a2,7 +a3,6 +a41゜■twtena, 6+a2,7 +a3. g +a4,10
■tw+a,,, +a2,6 +a3. g+a4.
,0(2)' When exchanging 3 pieces of data, 3C
2X 4C2=3X6=18 ways.

In other words, the feature of this method is that by sequentially increasing ij of element a, all combinational addition values can be obtained without waste.

Although the gist of the present invention has been described above with reference to specific embodiments thereof, it is clear that various modifications or modifications of the present invention based on what has already been described are possible.

(Effects of the Invention) As described above, according to the present invention, it is possible to provide a combinational weighing device that can perform high-speed processing with a shorter combination calculation processing time than conventional combinational weighing devices.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing a schematic configuration of a combination weighing device to which the present invention is applied, and FIG. 2 is an explanatory diagram of the combination weighing device. 1... Weighing machine, 2... Multiplexer, 3... A
/D converter, 4... Arithmetic control section, 5... Target weight setting section, 6... Packaging machine. Patent applicant Calbee Co., Ltd. and two other representatives Patent attorney Minoru Tsuji Figure 1 Figure 2

Claims (1)

[Claims] Means for storing N pieces of weight data obtained by a plurality of weighing machines divided into a subset and its complement, exchanging a plurality of weight data of a subset with a plurality of weight data of its complement, The method is characterized by comprising means for calculating a total value of weight data, and means for repeatedly exchanging the weight data to find a combination in which the total value of the weight data of the subset is equal to or closest to the target weight. Combination weighing device.