JPS63157014A - High-speed, high-accuracy mean quantity measurement system - Google Patents

Abstract

PURPOSE:To realize a high-speed, high-accuracy mean value measurement system by providing a refresh memory, a cumulative adding means, a mean measured quantity calculating means, etc. CONSTITUTION:When a continuous physical quantity or pulse train 1 to be measured is supplied to a digitizing circuit 2 first, it is converted into a digital quantity Qi at each sampling period Tsi(i=1-n) and stored in the refresh memory 3. Further, the cumulative adding means 4 integrates (n) digital values at every period T. Then, the mean measured quantity calculating means 5 adds the difference between the final measured quantity Qn of the refresh memory 3 which is obtained at each sampling period Tsi and a 1st measured quantity Q0 to the integral value found in a last period T to find a quantity proportional to a means value, and outputs the result as an output signal 6.

Description

[Detailed Description of the Invention] (Industrial Application Field) The present invention relates to an average quantity measurement method in which various physical quantities or pulse trains are digitized and measured by sampling them multiple times. , the ability to measure the average amount of physical quantities sampled multiple times in a short time and with high precision by performing minimal averaging processing. It is useful as a high-speed, high-precision average measurement method for physical quantities.

(Prior Art) When a physical quantity is digitized, sampled and measured multiple times, and the average value of the measured quantities is determined and output, a method is usually used in which the sum of the measured quantities is divided by the number of measurements and output.

That is, the average value Q is given by the following equation.

Ru.

(Problems to be Solved by the Invention) Now, a case will be described in which the physical LiQ gradually increases while changing as shown in FIG. 1 with respect to time t. In this case, T is divided into n equal parts for each measurement period T, physical quantity Q1 in each sampling period TSt, Q2 in TS2, etc.
..., TSn measures Qn, and calculates the average value oj of the j-th measurement period by calculating the following equation.

According to this method, variations in the physical quantity during the measurement period T have the effect of being averaged out. However, there is a drawback that the average value of the physical quantity during the measurement period T cannot be obtained until after the measurement period T, resulting in a measurement delay. If there is a measurement delay, it will cause instability in the control system and cause problems when used as a control detector.

Further, as shown in FIG. 1, when the physical quantity Q gradually increases while changing, the time t becomes the average value Q1 for the period o<t≦K,
There is a drawback that a considerable step occurs between the period T<t≦2 average value Q2 and the measurement error becomes large.

As is clear from the above description, the conventional technology has problems in that both measurement response and accuracy are poor.

In order to solve the above-mentioned problems, the present invention provides continuous n+1
By storing the number of digitized measurements in a refresh memory, and calculating the minimum number of times for each measurement to obtain the average value, it is possible to calculate This method is designed to measure an approximate value of the average value G shown in FIG. 1, and the purpose is to provide an average value measurement method that is n times faster and more accurate than conventional methods.

(Means for Solving the Problems) The present invention employs the following configuration in order to solve the above-mentioned conventional problems. That is, the specific invention comprises a first means that digitizes a continuous physical quantity or pulse train a plurality of times (n+1 times) and then sequentially refreshes and stores it, and a second means that intermittently accumulates the digital quantity n times. means and while sequentially refreshing and storing digital quantities;
Add the current digital amount and n+1 to the n-time accumulated value obtained earlier.
a third means for adding the difference between the previous digital quantities;
and a fourth means for outputting the result of the means as an amount proportional to the sum of n measurement values. The combined invention is a first method in which continuous physical quantities or pulse trains are digitized multiple times (n+1 times) and then sequentially refreshed and stored.
and a second means for intermittently accumulating the digital quantity n times.
and a third means for adding the difference between the current digital quantity and the n+1 previous digital quantity to the accumulated value obtained n times previously while the digital quantity is sequentially refreshed and stored; a fifth means of dividing the result of the means by the number of measurements n;
and a sixth means for outputting the result of the fifth means as an average value of n measured values. be.

(Function) According to the configuration of the present invention, instead of recording the measured quantities by sampling n times and then accumulating them to obtain the average value, the cumulative value measured up to the previous time is recorded by one sampling. By adding and subtracting the current value and the measured value n times before, an amount proportional to the average value can be obtained, and by dividing by the number of measurements n, the average value can be obtained.

(Example) Hereinafter, the present invention will be explained in detail based on the drawings.

FIG. 2 is a configuration diagram showing the principle of the present invention.

In the figure, 1 is the measured physical ω or pulse train, 2
is a digitization circuit, 3 is a refresh memory, and QoS
Ql, . . . , Qn are digitized measured quantities, 4 is a cumulative addition means for calculating the sum of the measured quantities from the first to the nth time for each period T, and 5 is the measured quantity of the above 4 and the first measurement. Amount Q. and 1/rl of the period T from the last measurement IQn
means for calculating an average measured quantity for each sampling period of;
6 is an output signal.

FIG. 1 is a graph illustrating the invention in detail.

First, when a continuous physical quantity or pulse train 1 to be measured is given to the digitizing circuit 2, the digital quantity Qi digitized every sampling period Tsi (i=1.2, . . . n) in FIG. and are stored in the refresh memory 3. Here, Qo is Q
This is the measured quantity one sampling period before l. The refresh memory 3 is realized, for example, by a shift register provided with the necessary number of bits. refresh memory 3
, i.e., n+1 or more measured quantities disappear. The cumulative addition means 4 calculates each period T in FIG.

The average measured quantity calculation means 5 adds the difference between the measured quantity Qn of the previous period and the first measured quantity Qo, and calculates the quantity Q proportional to the average value.
Find p. That is, this result is output as the output signal 6.

Alternatively, the average value Q obtained by dividing the above Qp by the number of sampling times n is calculated and output.

In this way, during the period o<t≦T, mQp proportional to the average value during the sampling period lsi or the average value h is output, and the measured quantity Qi is accumulated, and the value is transferred to the next period l<t≦ Used for 2 knives.

FIG. 3 is a block diagram of one embodiment of the present invention.

However, in FIG. 3, there is a method in which the function of the refresh memory 9 is performed by software processing of the microcomputer 10, but in FIG. 3, the refresh memory 9 is located outside the microcomputer 10.

In FIG. 3, 7 is a continuous physical quantity to be measured or a pulse train, and 8 is a digitizing circuit, such as an analog
Digital converter or pulse counter, 3 is refresh memory, QQlQl, ..., Qn-1, Qn
is a stored digital value, and 10 is a microcomputer. Microcomputer 10 is Q. It is composed of an input register 11, a Qn input register 12, a rewritable memory 13 for storing data, a read-only memory 14 for storing programs, a central processing unit 15, and a C output register 16.

When a continuous physical quantity or pulse train 7 is given,
The digitization circuit 8 converts this into a digital quantity, that is, a numerical value, every sampling period. This converted digital quantity is stored in a refresh memory every sampling period. The refresh memory 9 stores this numerical value while shifting it every sampling.

In the figure, Q□, Ql, ..., Qn are sampling periods T30Slsi, ..., l'-s, respectively.
This is the numerical value stored in n. Since the numerical values of each sampling period are first input one after another into the first stage of the refresh memory, these numerical values are inputted into the rewritable memory 13 of the microcomputer 10 through the Qn input register 12. These numerical values are cumulatively added every n sampling periods from i=0 to 1=n-1. That is, this cumulative value Q (T) and Qn are Qo input register 1
1 and Qn are input to the input register 12 and stored in the rewritable memory 13 of the microcomputer.

These numerical values are processed by the central processing unit 15 according to a program stored in the read-only memory 14. According to this processing method, Q is added to the value Qm obtained in the previous period T.
By adding n-Qo and dividing the total by n, an average value of 0 can be measured for each sampling period Tsi. This result is output through the output register 16.

If an amount proportional to the average value is required, n of the above
The process of dividing by can be omitted.

Hereinafter, the processing content of the microcomputer 10 for determining the average value Q will be explained in detail using FIG.

In FIG. 4, Qi is the measured value read in each sampling period, Q is the cumulative sum of these values, i is the i-th sampling number, and n-i is n-1.
Qn is the numerical value of the n-th sampling number, QT is the cumulative n value measured in the n sampling periods, and Q is the instantaneous average value measured in each sampling period.

Measured values Qi are input for each sampling period for bonds that have undergone initial value processing. The sampling period is repeated all at once from i=0 to 1=n-1.

The measured value Qi is accumulated in the next step.

1=n−i, i.e. every n sampling period,
The cumulative numerical value QT for that period is recorded.

During 0≦i<n, for each sampling period, Qo, Q
i is read and the result is instantaneously output by calculating the following equation.

In the above explanation, the calculation of QT = ΣQi is, in principle, necessary only the first time this program is started, and if there is no malfunction, it is sufficient to calculate QT = ΣQi only once. Therefore, by calculating this for each period T, it can be used to check rationality when an abnormal difference from the previous time occurs.

This measurement method can obtain the average value n times for each sampling, so it can output the average value n times faster than the conventional measurement speed, and with a high accuracy of about 1/n of the conventional measurement speed. It can be measured. In addition, since the calculation steps only need to be performed twice, which is a ratio of 2/n compared to the conventional n calculation steps, the probability of calculation errors is reduced accordingly.

Further, if an amount proportional to the average value is sufficient, there is no need to divide by n, so it is not necessary to use a microcomputer, and it can be configured with a refresh memory, an accumulator, an adder, etc.

(Effects of the Invention) As explained above, according to the present invention, after recording the measured quantities by sampling n times, they are accumulated,
Instead of finding the average value, by adding and subtracting the accumulated value measured up to the previous time by one sampling and the measured value this time and n times ago, the amount proportional to the average value is calculated by 1.
q, and the average value can be obtained by dividing by the number of measurements n. Therefore, there is an effect that the average value of a changing physical quantity can be measured with n times faster measurement speed than the conventional method and with a measurement error of about 1/n. Furthermore, since the calculation steps are 2/n of the conventional method, there is an effect that the probability of a calculation error is reduced to 2/n. Moreover, this averaging method is based on the continuously changing length 1. It is effective as a high-speed, high-precision average measurement method for all physical quantities such as pressure, flow rate, voltage, current, and power, as well as pulse trains.

[Brief explanation of the drawing]

FIG. 1 is a graph explaining the present invention in detail, FIG. 2 is a configuration diagram showing the principle of the present invention, FIG. 3 is a block diagram of an embodiment of the present invention, and FIG. 4 is a graph explaining the present invention. This is a flow chart showing the operation. (Explanation of symbols) 1... Physical quantity to be measured or pulse train, 2... Digitization circuit, 3... Refresh memory, 4... Cumulative tightening means, 5... Mean h1 survey fraud means, 6... Output signal, 7... Physical quantity to be measured or pulse train, 8... Digitization circuit, 9... Refresh memory, 10... Microcomputer, 11... QO manual register, 12... - Qn input register, 13... rewritable memory, 14... read-only memory, 15... central processing unit, 16... Q output register. (Explanation of symbols) 1...Wave meter 211I# J! l! ! Alternatively, pulse train 2...digitization circuit 3...refresh memory 4...cumulative amplification means 5...average measured quantity calculation means 6...output signal, 7...physical quantity to be measured or pulse train 8. ... Digitization circuit 9 ... Refresh memory 10 ... Microcomputer 11 ... QO manual register 12 ... Qn input register 13 ... Replaceable memory 14 ... Read-only memory 15 ...・Central processing unit 16...Hundred output registers Fig. 1, Fig. 2, Fig. 4 (explanation of symbols) 13...Writable memory 14...Read-only memory 15...Central processing unit 16... times output register Figure 3

Claims (2)

[Claims] (1) Repeating a continuous physical quantity or pulse train multiple times (n+
a first means for sequentially refreshing and storing the digital quantity after digitizing it; a second means for intermittently accumulating the digital quantity n times; a third means of adding the difference between the current digital quantity and the n+1 previous digital quantity to the cumulative value obtained n times; A high-speed, high-precision average value measurement method characterized by having a fourth means for outputting as follows. (2) Continuous physical quantities or pulse trains multiple times (n+
a first means for sequentially refreshing and storing the digital quantity after digitizing it; a second means for intermittently accumulating the digital quantity n times; a third means for adding the difference between the current digital quantity and the n+1 previous digital quantity to the cumulative value obtained n times; and a fifth means for dividing the result of the third means by the number of measurements n; and a sixth means for outputting the result of the fifth means as an average value of n measured values.