JPS62167416A - Electronic balance - Google Patents

Abstract

PURPOSE:To increase a response speed in the display of a balance and greatly attenuate ultralow frequency noise by using a digital filter which outputs a stable displayed value with no discontinuity. CONSTITUTION:A load-electricity transducer 1 transduces a load W on a weigh ing pan 2 to electric signals and an A/D converter 3 outputs digitally converted numerical data every prescribed time. The data are sampled every prescribed time and primary time series data Xn are obtained. then, from the data Xn, by first digital filter means (ROM 6, Y register and the like), secondary time series data Yn with a period Ty equal to a prescribed time Tx multiplied by an integer are obtained. Then, after the data Yn is judged to fall within a prescribed stable width by means of a control means (CPU) 5, from the data Yn, by second digital filter means (Z register and the like), tertiary time series data with a period Tz equal to the period Ty multiplied by an integer are obtained. Thus, final-order time series data which are the result calculated by final-order digital filter means are obtained as weighed data.

Description

[Detailed Description of the Invention] <Industrial Application Field> The present invention relates to an electronic balance, and in particular, a method for removing noise components from time series data obtained by sampling a weighing signal containing noise at intervals of time T. Regarding data processing technology.

<Prior art> Electronic balances convert the analog output of the load-to-electricity converter into A/
A D converter samples the data every predetermined time T, for example every 0.2 seconds, and converts it into a numerical value. Conventionally, this numerical data is subjected to arithmetic averaging to reduce noise components. That is, when the sampled time series data is (Xnl), the averaged time series data (Xn) is as follows.

<Problems to be Solved by the Invention> In response, the present inventor extracted the primary time series data (Xn) and data every 4fflit of this time series data, and extracted data whose period is an integer multiple, for example, 4fflit. A new data processing method in which double secondary time series data (Ynl is created, ... (2) where M is a positive integer of 2 or more + nO is a positive integer and is used as a digital filter means by executing the operation is proposed.

Electronic balances using this digital filter method exhibit excellent attenuation characteristics even for extremely low frequency noise, and can be realized using only computer software.
The structure is simple and has the advantage of not requiring a long averaging time for data processing as in the conventional example.

However, if there are large variations in the time series of the sampled signal before digital filter processing, when switching between digital filters of different orders, if the initial values at the start of digital filter processing are biased. If the digital filter is a value, its influence remains for a long time, and there is a problem that discontinuity occurs in the value when the digital filter is switched.

Therefore, the purpose of the present invention is to provide an electronic device that performs data processing using a digital filter, which outputs stable display values without discontinuities even when the dispersion of the initially sampled time series digital data is large. The aim is to provide balances.

<Means for Solving the Problems> The electronic balance of the present invention converts the load on the weighing pan into an electrical signal, and uses time-series data obtained by sampling the electrical signal at predetermined time intervals (Tx). In a device that displays a load value, based on the primary time series data {Xn} obtained by sampling at each predetermined time (Tx), the period (Ty) is an integral multiple of the predetermined time (TX). 2
The period (
second digital filter means for obtaining tertiary time series data {Zn} whose period (Tz) is an integral multiple of the period (Ty);
It is determined whether or not the above-mentioned secondary time series data {Yn} has fallen within a predetermined stable range, and after it has settled within the stable range, the above-mentioned second time-series data {Yn} is
It is characterized by having a control means for starting calculation of the digital filter means.

In the present invention, it goes without saying that integral multiples of the period of time-series data include one.

<Operation> Figure 1 shows a time chart of each time series data. Data is sampled every level 0.2 seconds for a predetermined time Tx, and primary time series data (Xn) is obtained. Based on the first time series data (Xnl), the calculation of the first digital filter means, for example, is executed to obtain the second time series data {Yn}.

After it is determined that the second digital time series {Yn} has settled within a predetermined stable range, the calculation of the second digital filter means is executed to obtain the third time series data (Zn). Similar data processing can be performed using third and subsequent digital filter means, and the final time series data, which is the calculation result of the final digital filter means, is used as the weighing data.

<Embodiment> FIG. 2 is a block diagram showing the overall configuration of an embodiment of the present invention.

The load-electricity converter 1 converts the load W on the weighing pan 2 into an electrical signal. The A/D converter 3 operates for a predetermined time T, for example 0.
．． Digitally converted numerical data Dn is output every 2 seconds. This includes noise in addition to the load W. The data processing section 4 includes a CPU 5. ROM6. It is equipped with RAM 7, which includes an X register (Xo-X3) and a Y register (Y
O~Y+6).

X register (Zo-Z+s), W register (Wo~W+
6), counter cl, C2 and flag lNTl.

2.3. The ROM 6 stores programs as described below. This program and the above registers constitute the digital filter of the present invention.

FIG. 3 shows a flowchart of a program according to an embodiment of the present invention.

At initial setting 11, the X register and Y register.

X register, W register, counters C1, C2 and flags lNTl, INT2. INT3 is cleared. Next, when the sampling signal X is input from the A/D converter, it is multiplied by, for example, -16 (12).

This 16x multiplication is performed in pure binary by shifting left by 4 bits. What is multiplied by this is the first time series (Xn).

The reason why the sampling input X is multiplied by 2 is to ensure that the displayed value completely returns to its original value when the input signal is a step signal or an impulse signal. Note that since it is divided by K after a series of digital filter processing is completed, it does not affect the displayed value.

At first, the flag INT1=O, so the process proceeds to step 13, where data X is manually entered into all digits of the X and Y registers in parallel. This is to speed up the initial response of the digital filter. At the same time, it is set to lNTl-1.

From the second time onward, the flag INT=1, so step 1
Proceed to step 4, where the sampling force X is input into the lowest digit of the X register.

In step 15, a first-order digital filter calculation is performed. This calculation is executed every time the sampling input XO arrives, and the second time series {Yn} is created. At the same time, the contents of the X register and Y register are shifted by one digit, and X3. The data of Y2O is discarded, and at the same time, the counter C1 is incremented by 1.

This counter C1 is 1 for every 4 of the secondary time series (Yn).
This is to create a third-order time series {Zn} by sampling the number of times. In step 16, INT2=O
When , the contents of cl are cleared.

Judgment step 17 determines whether the second time series {Yn} is stable. Specifically, data Y1 of each digit of the Y register. Y2. ..., the minimum value and maximum value of Y2O are extracted, and when the difference between them is smaller than a predetermined amount A, it is determined that the stability range has been reached. The determination as to whether or not the system is stable may be made by determining that the system is stable even if all 16 digits of data are not present, for example, if all three data are within a predetermined width. If it is determined that the data is not stable yet, it is determined that there is no stable data, and the fluctuating data 4×↓ is displayed in step 18.

On the other hand, if it is determined in judgment step 17 that the time series data {Yn} is stable, the process proceeds to step 19, where the average value Y of the stable data Yl-Y16 is calculated. When INT2=1 in step 20, stable data Y/16 is displayed in step 22, and when INT2=0, this stable average value Y is transferred to Zl, z2, and z3 of the X register, and is sent to INT2=1. Ru. -Manban [INT in M step 16
When 2=1, when C1-4 is determined in step 23, the calculation of the second digital filter (6) is executed in step 24, and is stored in Zo. At the same time, the contents of the Z register are shifted by one digit, and the counter C2 is incremented by 1.
Counter C1 is cleared.

What should be noted here is that Yl is included in the second term of equation (6).

Ys , Ys , Yl3 To4 (1a Tobi notator is sampled. In this way, when performing the next digital filter calculation by sampling 41 consecutive data of the previous time series data, The maximum point or minimum point of the frequency characteristic of the first digital filter coincides with the minimum point or maximum point of the frequency characteristic of the subsequent digital filter, which has the effect of improving the overall frequency characteristic.This is because: This is because at the minimum point of the frequency of the digital filter, the attenuation amount is -■dB and the output becomes completely 0, and the overall frequency characteristic also has an attenuation amount of -ood3 at that point.

It is also determined in step 26 whether or not the third time series data (Z n) obtained by the second digital filter has become stable, and only after it has become stable does the process proceed to step 27. A stable average value Z in Znl is calculated.

Then, when INT3=1 in step 28, stable data Z/16 is displayed in step 30, and (NT
When 3=O, the stable average value is Wl of the W register,
W2. It is transferred to W3 and INT3=1 is set.

On the other hand, if INT3=1 in judgment step 25, if C2=4 in step 31, step 3
At step 2, the third-order digital filter calculation (8) is executed, and at the same time, the contents of the W register are shifted by one digit, and the counter C2 is cleared. This digital filter also samples every four pieces of data from the previous stage time series.

If the time series (Wn) is stable, judgment step 3
3 to 34.35, and set the total average value of the W register to 1.
/ is displayed multiplied by 1/16. Time series (Zn
), but if the time series (Wn) is still not stable, the process proceeds from judgment step 26 to step 27.30, where a value based on the total average value of the Z register is displayed, and the time series (Yn) If only the Y register is stable, the process similarly proceeds from decision step 17 to step 19.22, where a value based on the total average value of the Y register is displayed.

Fig. 4 shows the frequency characteristics of the time series (noise attenuation rate (dB) in Znl) in this example.What should be noted in this characteristic diagram is, first, the frequency O and f
The amount of attenuation in the intermediate band excluding the vicinity of is extremely large at around 150 dB. Second, there is a frequency point at which the output drops to 0 (attenuation 1-oodB) every 0/16, and therefore the amount of attenuation near frequency O is much larger than in the conventional case.

<Effects of the Invention> According to the present invention, the configuration is simpler than that of conventional averaging processing, the display response speed is fast, and ultra-low frequency noise can be greatly attenuated.

Therefore, it is possible to easily eliminate display flickering caused by wind and slow display undulations, which have been extremely difficult to eliminate in the past.

Furthermore, since the processing of the subsequent digital filter is started after confirming that the output data of the previous digital filter is stable, display discontinuity does not occur, and the characteristics of removing noise components are further improved.

[Brief explanation of drawings]

FIG. 1 is a time chart of time series data according to the present invention. FIG. 2 is a block diagram showing the overall configuration of the embodiment of the present invention, and FIG. 3 (shown in two pages a and b) is a flowchart of the program of the embodiment of the present invention. FIG. 4 is a frequency characteristic diagram of the attenuation rate in the above embodiment.

Claims (2)

[Claims] (1) In a device that converts the load on a weighing pan into an electrical signal and displays the load value using time series data obtained by sampling the electrical signal at predetermined time intervals (Tx),
From the first time series data {Xn} obtained by sampling at each predetermined time (Tx), the second time series data {Xn} whose period (Ty) is an integral multiple of the predetermined time (Tx) is obtained.
a first digital filter means for obtaining Yn};
a second digital filter means for obtaining tertiary time series data {Zn} whose period (Tz) is an integral multiple of the period (Ty) from the tertiary time series data {Yn}; An electronic balance comprising a control means for determining whether or not {Yn} has fallen within a predetermined stable range, and for starting calculation of the second digital filter means after it has fallen within the stable range. (2) An electronic balance according to claim 1, which has a plurality of orders of digital filter means, and is configured to preferentially display processing results by digital filters of large orders.