JPH0460425A - Counter - Google Patents

Abstract

PURPOSE:To manage the number of charged samples to prevent an error from occurring by automatically calculating the maximum number of samples which can be counted at a time during a process of sampling. CONSTITUTION:A weight measuring unit 1 comprises a load cell 3 equipped with a mount 2, while a signal from the load cell 3 is input to a compensation circuit 6 via an amplifier 4 and a filter 5. The compensation circuit 6 compensates a creep strain of the load cell 3 taking place when a sample is mounted and a time fluctuation of a signal generated due to a drift of the amplifier 4 to output a weight signal stable in time to an analog-digital converter 7. When at least two average unit weights are obtained by first and second sample mounting, an arithmetic processor 10 obtains the maximum and minimum unit weights due to a fluctuation in samples, and based on the weights the maximum number of samples which can be counted in subsequent sampling is automatically obtained. Thus occurrence of an error in counting can be prevented.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to a counting system for determining the number of small articles manufactured under the same standard from their weight.

(Prior art) When checking the number of small items such as screws manufactured to the same standard, there are two methods: detecting each small item with an optical sensor, etc. and adding up the total using a counter; There is a method of measuring a weight of 1M and dividing this weight by the weight of one article.

Among these methods, the latter method of measuring weight allows highly accurate counting with a relatively simple device configuration.

When calculating the number of items based on weight in this way, errors occur in the number of items to be counted due to variations in the number of items. , the weight IWH of the heavy-manufactured article and the average weight Wu are determined in advance, and the maximum number N that can be counted per degree is calculated as follows: W/WL-W/WH<1 W/Wμ=N An apparatus has been proposed in which the process of loading and counting samples within this number and discharging the samples after counting is repeated several times (Japanese Patent Publication No. 60-5)
9531).

According to this method, it is possible to eliminate errors in the number of items counted, but on the other hand, it is necessary to repeat the process of removing the item from the loading platform and loading the next sample every time the number of items for each maximum N@ is counted. Therefore, there is a problem of low work efficiency.

In order to solve this problem, a single counter has been proposed in which the maximum number of samples that can be loaded on the loading platform is limited to the number N mentioned above, but it is possible to gradually increase the number of samples ( (Special Publication No. 1-21447).

According to such a system 3, there is no need for a step of removing samples from the R loading platform every time a counting operation is performed within the regulated maximum number N, making it possible to greatly improve work efficiency.

(Problem to be Solved by the Invention) However, in these devices, data for determining the maximum number N that does not include counting errors, that is, the maximum single weight WH of the sample, the minimum minimum type weight L, and the average value W
Not only is it necessary to calculate u in advance and enter it from a keyboard, etc., but also if you mistakenly insert more samples than the maximum number NM, a counting operation will be performed, causing an error in the cumulative total. There is.

The present invention was made in view of these problems, and its purpose is to automatically calculate the maximum number of samples that can be counted to -/i during the sampling process, and to An object of the present invention is to provide a new counting snowmaking system that can prevent the occurrence of errors by managing the number of samples input based on the number of samples.

(Means for Solving the Problems) In order to solve such problems, the present invention first! 12 Storage means for storing the weight of the loaded samples and the average unit weight of the samples calculated from the input number of samples, and the average unit weight and standard of the samples additionally loaded from the second time onwards. means for calculating the deviation and estimating the maximum single weight and minimum single weight of the sample based on statistical calculation processing; means for calculating the maximum number of samples without causing a counting error from the maximum single weight and minimum single weight; Each time a sample is added, the number is compared with the maximum number, and if it is less than the maximum number, the number of samples is added to the number of samples already accumulated, and if it exceeds the maximum number, a warning is issued. I prepared the means.

(Operation of the invention) When at least two average unit weights are determined by the first and second sample loading, the maximum and minimum unit weights due to sample variations are determined by statistical processing means, and based on this, the maximum and minimum unit weights due to sample variations are determined. The maximum number that can be counted in subsequent samplings is automatically determined, thereby allowing additions to be made within a range where no counting errors occur.

(Example) The details of the present invention will be described below based on illustrated examples.

FIG. 2 shows an embodiment of the present invention, in which reference numeral 1 denotes a weight measuring section, which is composed of a load cell 3 provided with a loading platform 2, and the signal from the roto cell 3 is The signal is inputted via an amplifier 4 and a filter 5 to a compensation circuit 6, which will be described later. 6 is the above-mentioned compensation circuit, which corrects temporal fluctuations in the signal caused by the creep deformation of the load cell 3 that occurs when a sample is loaded and the drift of the amplifier 4, and stabilizes the signal over time. It is configured to output the weight signal to the analog-digital converter 7.

10 is arithmetic processing snowmaking, CPU II, ROM12, R
It is composed of a microcomputer consisting of AM13, which includes a keyboard 14 for externally inputting the first sample number, a display snowmaking 15 for displaying counting data and messages, and a safe maximum number of samples to be described later. An alarm 16 is connected to issue an alarm when the majority Q is exceeded.

FIG. 1 conceptually shows the processing operation of the arithmetic processing snowmaking 1o, which detects whether or not the weight signal is stabilized based on the data from the analog-digital converter 7, and A weight signal stability detection means 30 outputs the weight signal to the subsequent stage at the time when the sample is loaded, a weight data storage means 31 stores the accumulated weight of the sample, and the weight signal of only the newly placed sample is detected.
Weight change calculating means 32 for detecting the weight change, number calculating means 33 for calculating the number of additional samples from the average type N of one sample (hereinafter referred to as average unit weight) and the weight of the additional samples; An average unit weight calculation means 34 for calculating the average unit weight of the sample, an average unit weight storage means for storing all the average unit weights, and a total number storage means 35 for storing the number of samples up to the previous time were added. A total number calculation means 37 that calculates the number of samples and the total number of pieces so far, an average value calculation means 38 that calculates the average value of all average unit weights, and a standard deviation calculation means that calculates the standard deviation of the average unit weight. 39, the standard deviation, and the reliability coefficient stored in the reliability coefficient storage means 40, calculate the interval width of the average unit weight of the sample with appropriate reliability, for example, 95% reliability, and calculate the upper and lower limit values. Maximum/minimum unit weight calculation means 4 which outputs the maximum unit weight and minimum unit weight as the maximum unit weight and minimum unit weight; and a safe maximum number calculation unit 42 which calculates the maximum number Q without causing an error based on the maximum unit weight and minimum unit weight. When the number of currently loaded samples is within the maximum number 0, the data in the weight data storage means 31 and the total number storage means 36 are updated, and when the number exceeds the maximum number Qt, the weight data storage means 31 is updated. , and a comparison means 43 that activates the alarm 16 to prompt the reduction of the number of samples without updating the total number storage means 36.

Next, the operation of the apparatus configured as described above will be explained based on the flowchart shown in FIG.

Prior to the counting operation, when a known number of samples are placed on the loading platform (Step I), a weight signal is output from the weight measuring section 1 in response to this. This weight signal is amplified to a predetermined level by an amplifier 4, then noise is removed by a filter 5, and input to a correction circuit 6. The correction circuit 6 corrects temporal fluctuations in the weight signal due to creep deformation due to load application and drift due to temperature changes in the strain gauge and amplifier, and outputs an accurate and temporally stable weight signal. do.

Load a known number of samples, once the weight signal is stable (
(Step 7), the weight data from the analog-to-digital converter 7 is taken in and stored in the weight data storage means 3 (Step C), and the average of the sample is calculated based on the number data input from the keyboard 4 (Step 2). Unit weight Ul
Calculate t (step e), these weight Wl, number n
and average unit weight U, respectively, are stored in the weight data storage means 31.
, the total number storage means 36, and the average unit weight storage means 35. The average unit weight U found now.

Based on this, the average value calculating means 38 calculates the average value X of the average unit weight (to step 2). Needless to say, although the number of data is one, X=U.

Subsequently, when a predetermined number of samples are added and placed on the loading platform (steps t, h), when the weight signal becomes stable (step ri), the weight change calculation means 32 calculates the current weight signal. The first weight is stored in the weight data storage means 31! Additional weight w2 from the difference from w+
Calculate (step nu). Next, the number calculation means 37
Based on the average unit weight U1 and weight IW2 obtained last time, the number n2 of additional samples is calculated, and the average unit weight U2Vj of the additional samples is determined by the average unit weight calculation means 34! This is stored in the average unit weight storage means 35. This number n2 is added to the number n+I by the total number calculation means 37 and displayed on the display 1.
At the same time, the data in the total number storage circuit 36 is updated (step), and the data in the weight data storage means 31 is updated to the current weight (step).

The average value calculating means 38 calculates the first average unit weight U1 and the second average unit weight U1.
Average value X = (U, +u,
) / 21F! , calculate (go to step ).

At the stage where a plurality of unit weights have been determined in this way (step 1), the standard deviation calculation means 39 compares the standard deviations, and this standard deviation S and the reliability coefficient t (n- 1) = t According to (1), the average weight of the article in the population of the entire sample estimated from the data obtained in two samplings is calculated by the maximum/minimum unit weight calculation means 41 and constitutes the sample. Statistically estimate the unit weight WL of the lightest item in the list to be X-t(n-7), /-7, and the unit weight IWH of the heaviest item (Ste Mbuwa).

Then these data WL, WH! Using the formula, find the maximum weight W of the sample amount that can be added without causing an error during coefficient calculation using the formula W/WL-W/WH<1, and calculate the maximum number that can be counted from this, that is, the safe maximum number 03 = Calculate W/WH.

Needless to say, this numerical value Q3 indicates the total number that can be accumulated by the third sampling, which is determined based on the average unit weight and standard deviation obtained by the two samplings.

This numerical value is output to the display 15 and displayed as the third possible number of sassivables (step F).

Next, a third sampling is performed (step
y), When the weight signal becomes stable (step ta), the weight signal is taken in and the difference with the weight data stored in the weight data storage means 31 up to the second time is calculated (step shi). , the weight signal from the heavy jl branching means 1 is corrected by a correction circuit for changes over time due to creep and softness, and the weight signal is adjusted from the second stable point to the third sample as shown in FIG. Since the value is maintained at a constant value until immediately before addition, the additional weight can be calculated by calculating the difference W3 W2 from the weight data w2 stored in the weight data storage means 31 when the weight signal was stable last time. It is possible to measure accurately.

The weight jlW3 of this added sample W2 is divided by the average unit weight U2 that has already been calculated to find the number n3 of added samples, then the average unit weight U3 is calculated, and the number n3 is calculated by dividing the weight of the added sample by the average unit weight U2 that has already been calculated. Add the number nl to +n2 and display the total number n, +n2 +n3 (Step So)
, Now compare the total number of pieces calculated with the safe maximum number of pieces 03 determined by step (force), @Total number of pieces on the loading platform n, +n
2. If +n3 is within the safe maximum number of samples 03, the additional sample number n3 is added to the number n and +n2 stored in the total number storage means 36 and displayed. , stores it in the total number storage means 37, stores it in the average unit weight U3M average unit weight storage means 35, and updates the weight data in the weight data storage means 31 (
step ne).

Until the next sample is added, the fourth safe maximum number 04 is calculated by steps (to) to (force).

When samples of bacteria have been added four times (step Y) and the total number of samples due to addition is calculated from step (ri) to step (n), the comparison circuit 43 calculates this number n4 and the safe maximum number. Compare 04. In this case, since the total number exceeds the maximum safe number Q, G (step 2), the comparison means 43 causes the display 15 to display the message "Please reduce the number of samples to a countable number" or Based on this display and the alarm sound, the total number of samples is reduced to n4° (step ra), and when the weight signal becomes stable again (step ta), the weight The signal is acquired and the weight of the additional sample is determined (Step 1).This reduction brings the number of samples within the safe maximum number of 04, so through the process of Steps (S) to (T), the weight data is updated and the total weight is calculated. The number of samples is updated (step ne), and the number of samples that can be loaded next, that is, the safe maximum number Qst, is calculated by roughly stepping (step) or step (force).
Prepare for the next counting operation.

By the way, in addition to calculating the average unit weight of only the newly added sample, we also calculate the interval of the sample average value, that is, the maximum unit weight and minimum unit weight of the next sample to be counted, from the variance S that takes this into account. Since it is calculated based on the calculation, it is possible to calculate the number of pieces and determine the safe maximum number of pieces by taking into account the time changes in the manufacturing process, and it is possible to perform counting operations continuously with high reliability.

In this embodiment, the total number of samples to be reduced is displayed, but the relative number to be reduced may be displayed.

Note that in this example, the confidence interval is estimated using a normal distribution to estimate the maximum and minimum unit weight of one sample, but the same result can be obtained using other methods such as one distribution. It is clear that this effect is achieved.

(Effects of the Invention) As described above, the present invention includes a storage means for storing the weight of the first loaded sample and the average unit weight of the sample obtained from the input number of samples; Means for determining the average unit weight and standard deviation of samples additionally loaded after the third loading, and estimating the maximum unit weight and minimum unit weight of the sample based on statistical calculation processing, and from the said maximum unit weight and minimum unit weight. A method for determining the maximum number of samples without a counting error, and each time a sample is added, the number is compared with the maximum number, and if the number is less than the maximum number, the number of samples is calculated as the number of samples already accumulated. Since we have a means to add the total number of samples and to issue an alarm if the maximum number is exceeded, the sample can be divided into at least two separate samples. ! By using this method, the maximum number of samples per run without errors is automatically managed using error statistical methods, so the number of data to be input from the outside can be reduced to the initial number of samples, improving usability. Not only is it possible to calculate the number of pieces using the new average unit weight each time a sample is added, but also to calculate the distribution of the samples to be counted from now on, so it is highly accurate even for samples whose degree of dispersion changes over time. It is possible to obtain a count value with reliability.

[Brief explanation of drawings]

Fig. 1 is a block diagram showing the concept of arithmetic processing of the device of the present invention, Fig. 2 is a block diagram of the device showing an embodiment of the present invention, and Figs. 3 and 4 respectively illustrate the operation of snowmaking. 2 is a flowchart and a line diagram shown in FIG. Applicant: Ishida Kouki Seisakusho Co., Ltd. Agent: Patent attorney: Katsuhiko Kimura, Nishikawa Kajika+l*IA

Claims (1)

[Claims] A storage means for storing the weight of the first loaded sample and the average unit weight of the sample calculated from the input number of samples, and the average unit weight and standard of the additional sample loaded from the second time onwards. means for determining the deviation and estimating the maximum single weight and minimum single weight of the sample based on statistical calculation processing; means for calculating the maximum number of samples without a counting error from the maximum single weight and minimum single weight;
Each time a sample is added, the number is compared with the maximum number, and if it is less than the maximum number, the number of samples is added to the cumulative number of samples, and if it exceeds the maximum number, a warning is issued. A counting device comprising: means for emitting.