JPH03137419A - Flow rate measuring device - Google Patents

Abstract

PURPOSE:To make the title device highly accurate in measuring a flow rate in a full flow zone while allowing the use of general-purpose pressure sensors, which are easily obtainable, by a method wherein said device is provided with an arithmetic and control unit which corrects a value actually measured by each sensor on the basis of an error approximate value, computes the calculated value of a flow rate from the value obtained by this correction and outputs a signal of the calculated value. CONSTITUTION:In a case where an approximate value for each of sensors 3, 6, 7, which is calculated by applying an error approximation formula, becomes larger than a predetermined value, an arithmetic and control unit 8 produces an alarm signal or corrects the value actually measured by each of said sensors on the basis of the approximate value, and the calculated value of a flow rate is computed every moment from the value obtained by this correction, whereupon a signal of the calculated value is outputted. In addition, the arithmetic and control unit 8 has a control function for controlling a flow rate, or for controlling a pressure difference or an opening degree of a variable throttle means 5 to a predetermined value, and sends an opening degree control signal to a variable throttle driving means 4 to regulate the opening degree of the variable throttle means 5.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to a flow rate measuring device that is applied to, for example, a fuel or air pipe line to be supplied to a heating furnace.

(Prior Art) Conventionally, as shown in FIG. 7, this type of flow rate measurement requires a fixed throttle mechanism 12 such as an orifice or venturi provided in a fluid pipe 11 to be measured, and a fixed throttle mechanism 12. A device that combines a differential pressure sensor 13 such as a differential pressure gauge or a differential pressure converter that measures the pressure difference before and after the differential pressure sensor 13, and an arithmetic controller 14 that calculates and outputs the flow rate based on the signal from the differential pressure sensor 13. Usually used alone.

(Problem to be Solved by the Invention) As is well known, the differential pressure of the fixed throttle mechanism 12 is proportional to the square of the flow rate and decreases rapidly as the flow rate decreases. Figure 8 (
There is a problem that the zero drift of the differential pressure sensor increases rapidly as the flow rate decreases, as in the case where the zero drift is ±1%.

On the other hand, FIG. 9 shows a flow rate measuring device designed to avoid errors in the low flow rate region, and shows a fluid pipe line 2 to be measured.
1, a differential pressure sensor 23 that detects the differential pressure before and after the variable diaphragm mechanism 22, and a variable diaphragm mechanism 22 provided in the variable diaphragm mechanism 2.
Opening sensor 24 attached to 2. It consists of a variable diaphragm drive mechanism 25 and an arithmetic controller 26 that calculates and outputs the flow rate based on the signals from each of the sensors. In addition, the 9th
In the example shown in the figure, the variable diaphragm mechanism 22 is controlled via the variable diaphragm drive mechanism 25 based on the calculated flow rate.

The flow rate measurement error in this device is mainly due to the error related to the differential pressure sensor 23 and the error related to the opening degree of the variable throttle mechanism 22. The problem is that in high flow areas, the flow rate measurement error due to the variable throttle mechanism 22 is larger than the flow rate measurement error of the flow rate measurement method using the fixed throttle mechanism 12, as in the case where the zero drift is ±0.5%. There is.

In general, errors related to differential pressure sensors and errors related to the opening degree of variable throttle mechanisms constantly fluctuate, and errors related to these sensors can cause deterioration in fuel consumption and product quality due to a decrease in flow rate measurement control accuracy. In order to avoid losses such as degradation, it is necessary to frequently correct errors related to each sensor, so-called calibration. Conventionally, these calibrations had to be performed by an operator using a reference device, etc. after stopping the operation of the equipment.Increasing the frequency of calibration would result in loss of production opportunities due to the stoppage of operation, and labor for calibration work. It will increase.

Therefore, conventionally, in many cases, they are operated at an economic compromise between the two.

The present invention has been made to solve the problems of the conventional technology, and it is possible to automatically calibrate the sensor frequently during operation, that is, to automatically correct errors related to each sensor. The present invention aims to provide a simple flow rate measuring device that uses a pressure sensor and is highly accurate over the entire flow rate range.

(Means for Solving the Problems) In order to solve the above problems, the present invention provides a fixed throttle mechanism and a variable throttle mechanism provided in a fluid pipeline to be measured;
A variable diaphragm drive mechanism and an opening sensor attached to the variable diaphragm mechanism, a first differential pressure sensor and a second differential pressure sensor that detect the differential pressure before and after each of the fixed diaphragm mechanism and the variable diaphragm mechanism, and the fixed diaphragm mechanism and the opening sensor. A flow rate measurement value based on a signal from a first differential pressure sensor of a fixed throttle system including a throttle mechanism and a first differential pressure sensor, and an opening of a variable throttle system including the variable throttle mechanism, an opening sensor, and a second differential pressure sensor. The difference between the calculation formulas for calculating each of the flow rate measurement values based on the signals from the pressure sensor and the second differential pressure sensor, and the error related to each of the above sensors based on the "general formula for error propagation" for both calculation formulas. A simultaneous equation is created by substituting the actual measured values of each of the above sensors at the required number of flow rates at appropriate intervals into the error approximation equation assuming that the difference in the approximate equation of the flow rate measurement error is equal to the difference in the approximate equation of the flow rate measurement error. An arithmetic controller that calculates an approximate value of the error by solving for the error, corrects the actual measured value of each sensor mentioned above using the approximate value of the error, and calculates and outputs the flow rate measurement value based on the corrected value. Formed from.

(Example) Next, one embodiment of the present invention will be described with reference to the drawings.

FIG. 1 shows a flow rate measuring device according to the present invention, which includes a fixed throttle mechanism 2, an opening sensor 3, and a variable throttle mechanism in which the relationship between the flow coefficient and the opening is known, in a fluid pipe line l to be measured. A variable diaphragm mechanism 5 to which a drive mechanism 4 is attached is provided, and a first diaphragm mechanism 5 for detecting the differential pressure before and after each of the fixed diaphragm mechanism 2 and the variable diaphragm mechanism 5 is provided. Second differential pressure sensor 6.7
is the provision.

Also, 1st. An arithmetic controller 8 is provided which inputs the signal from the second differential pressure sensor 6.7 and the signal from the opening sensor 3. This arithmetic controller 8 is formed using a microprocessor or the like, and stores the flow coefficient of the fixed throttle mechanism 2 and the flow coefficient for the opening of the variable throttle mechanism 5 stored in advance in the form of a mathematical formula or a table. The relationship expressed and
Fixed diaphragm mechanism 2. based on each of the above sensor signals. Fixed diaphragm system and variable diaphragm mechanism including the first differential pressure sensor 6 5. Opening sensor 3. The difference between the arithmetic expressions for calculating the total measured values of both streams with the variable throttle system including the second differential pressure sensor 7, and the error related to each sensor based on the general formula-1 of error propagation for both arithmetic expressions. A simultaneous equation is created by substituting the actual measured values of each sensor at the required number of flow rates at appropriate intervals into the error approximation formula, which assumes that the difference in the approximate formula for flow rate measurement error is equal to the difference in the approximate formula for flow rate measurement error. By solving
Calculate its approximate value. Furthermore, this arithmetic controller 8
If these approximate values become larger than a predetermined value, an alarm signal is issued and/or the actual measured value of the sensor is corrected by the approximate value, and the flow rate measurement value is calculated from moment to moment based on the corrected value. , is what is output.

Note that this arithmetic controller 8 controls the flow rate or the variable throttle mechanism 5.
It also has a control function for setting the differential pressure or the opening degree to a predetermined value, and sends an opening degree control signal to the variable diaphragm drive mechanism 4 to control the opening degree of the variable diaphragm mechanism 5.

Therefore, first, a method for calculating approximate values of errors regarding each sensor will be described in detail.

Assuming that the flow coefficient of the fixed throttle mechanism 2 is A, and the actual measured value of the differential pressure, that is, the differential pressure value including errors related to the first differential pressure sensor 6 and the differential pressure, is Pa, the measured flow rate Qa is as follows (1 )
Expressed by the formula.

Qa=A J7'a - (1) The flow coefficient of the variable throttle mechanism 5 is the opening function f(s), the actual measured value of the opening, that is, the opening value including the error related to the opening sensor 3 and the opening. 81 If the actual measured value of the differential pressure, that is, the differential pressure value including the error related to the second differential pressure sensor 7 and the differential pressure measurement, is pb, then the flow rate measurement value Qb is expressed by the following equation (2). .

Q b= f(s)F7 cho・(2
) The two flow measurements will generally differ, except by chance, due to errors associated with each sensor.

Here, regarding the difference between the right-hand side of equation (1) and the right-hand side of equation (2), an approximate equation based on the well-known "general equation for error propagation" is obtained as the following equation (3).

(θ(r (s) FTl) /θS)δS+(θ(
f(s)FTl)/θPb)δpb(θ(ALTl)/
θPa) δPa = (θf(S)/θSr lower δ5 lf(s)/(2, FTl)) δpb (A/(2fTτ)) δPa − (3) This (
Equation 3) is an approximation of the difference obtained by subtracting the right-hand side of Equation (1) from the right-hand side of Equation (2) above, so the following Equation (4), assuming that both are equal, is used to approximate the error for each sensor. This is the basic formula for calculating the value, and is tentatively called the error approximation formula.

(θf(S)/θ5)ffiδS + (f(S)/2 fl lower))δPb(A/2
(7”'a)) δPa=r(sMT'bottom-ArF"
a - (4) In this equation (4), the function f(s) of the opening degree of the flow coefficient of the variable throttle mechanism 5 and the flow coefficient A of the fixed throttle mechanism 2 are known, so the appropriate interval can be calculated in the equation (4). For example, for each 5% of the maximum flow rate, create an equation for the required number of sensors by substituting the actual values measured by each sensor at the same number of flow rates (three in this example) as the number of sensors, which is the number of unknowns. Using the equations as simultaneous linear equations, and using the errors (δS, δPb, δPa) related to each sensor as unknowns, an approximate value of the error related to each sensor is determined by a well-known method such as Gaussian elimination.

The arithmetic controller 8 issues an alarm signal when these approximate values become larger than a predetermined value, and/or corrects the actual value measured by the sensor using the approximate value, and performs flow measurement using the corrected value. Calculate the value.

Next, in accordance with FIG. 2, the logic for selecting actual measurement values by each sensor for taking in actual measurement values by each sensor, storing them, and calculating approximate values of errors related to them will be described in detail.

First, a plurality of set flow values are determined in advance at appropriate intervals over the entire required flow rate region (#2-1), and the flow rate measurement values obtained by the fixed or variable throttle system are read in (#2-1).
#2-2), and each time this matches the set flow rate value, the process advances to the next step (#2-3), and the actual measured value of each sensor is imported into the storage area corresponding to each set flow rate value (#2-4). ),
The stored data is updated (#2-5).

Each time the actual measured values are taken in by each sensor and the stored data is updated, including those that have been updated,
Select the latest measured values of the required number of sensors with different flow rate settings (#2-6), and calculate the approximate error value for each sensor using simultaneous linear equations based on the error approximation formula by substituting these values. (#2-7), this calculation is completed, and the actual values measured by each sensor are corrected based on this approximate value.

According to the calculation method detailed so far, since each sensor is calibrated in advance, the error related to each sensor is small, for example, ±1% or less, and the required flow rate range is limited to the high flow rate range. range, e.g. 60 to 80 of the maximum flow rate.
% range, it is practically sufficient to calculate approximate values of errors related to each sensor with sufficient accuracy, for example, ±0.03% or less, but as these errors and the flow rate range increase, the error approximation becomes more difficult. The approximation error of the approximation of the error for each sensor due to the approximation of the equation increases.

Next, a calculation method for sufficiently reducing this approximation error will be described in detail with reference to FIG.

Read the actual measured values from a required number of sensors, each with a different set flow rate value (#3-1), calculate the approximate error value for each sensor (#3-2), and compare this approximate value with the previous value. Determine whether the difference from the approximate value (O at first) is smaller than a predetermined value (#3-3), and if it is not smaller, correct the actual measured value of each sensor using the integrated value of the approximate value (#3-3). 35, #3-6), calculate the approximate value similar to the above for the corrected actual measured value of each sensor (#32
). The calculation of the approximate value, its integration, and the correction of the actual measured value are repeated until the calculated approximate value and/or the difference between the approximate value and the previous approximate value becomes smaller than a predetermined value. That is, at the end of the above iteration, the iterative convergence calculation is performed until the approximation error of the approximation value of the error regarding each sensor due to the approximation of the error approximation formula becomes a sufficiently small value, and the approximation value of the convergence result is Since the integrated value is an approximate value with a sufficiently small approximation error of the error related to each sensor included in the actual measured value of each sensor before correction, the actual measured value is corrected using this integrated value (#3-4).

Furthermore, when the error associated with each sensor becomes large, for example when most of the calibration immediately after manufacture is to be simplified by auto-calibration, or when the required flow rate range becomes wide, for example for combustion control with a large turndown ratio. When used, the approximation error of the approximation value of the error regarding each sensor due to the approximation of the error approximation formula may become excessive, and may diverge or not converge in the above-described iterative calculations.

A calculation method for preventing this divergence will be described in detail below with reference to FIG.

In the iterative convergence calculation as described above, the actual measured values from a required number of sensors each having a different set flow rate value are read (#4-1), and the approximate value of the error is calculated using a simultaneous linear equation based on the error approximation formula. Calculate #4-2), add a coefficient of 1 or less, for example 0, to the calculated approximate value of the error for each sensor.
．． 5 (#4-3) and/or if these approximate values are larger than a predetermined value, e.g. 10%, limit it to a predetermined value (#4-7) -8),
Perform an iterative convergence calculation that repeats correction of the measured value using the integrated value (#4-9), and if the approximation error is sufficiently small (#4-5), correct the actual measured value using the integrated value. (#4-6).

According to the calculation method detailed so far, there is no problem if the fluctuations in the actual measurement values from each sensor and the errors related to each sensor are gentle compared to the frequency of correction, but if these fluctuations are frequent (short period) In such a case, the flow rate measurement value may suddenly change due to the fluctuation of the correction value each time, and especially when the flow rate measurement value is used for control, the control accuracy may decrease due to control instability and the life span of the moving parts may be shortened. Often causes problems.

A calculation method for avoiding sudden changes in the flow rate measurement value will be described in detail below with reference to FIG.

The fluid to be measured and controlled often has extremely short-period fluctuations on a regular basis, such as pulsations caused by pumps and blowers, and pressure fluctuations due to fluctuations in usage within the system. There are often large fluctuations in the actual values measured by each differential pressure sensor.

The first method for suppressing fluctuations in correction values due to this type of fluctuation
Before incorporating the actual measured values of each sensor to calculate the approximate value of the error of each sensor, the actual measured values of each sensor are shortened by moving averaging with an appropriate time constant (filtering of high frequency components, also known as exponential smoothing). By keeping the periodic fluctuation sufficiently small (#5-1), the averaged values of each sensor are read (#5-2), and based on this, the approximate error value and correction value are calculated (#5- 3).

Since fluctuations in errors related to each sensor are fluctuations in the actual values measured by each sensor, it can be considerably suppressed by smoothing the fluctuations in the actual measurements as described above, but this is not always perfect.

If problems remain due to fluctuations in the correction values, use the second method of suppressing fluctuations in the correction values, moving averaging of the calculated correction values (#5-4), and calculate the actual values measured by each sensor. Perform the correction (#5-5).

Note that if the fluctuations in the actual measured values are slight as described above, it may be sufficient to simply move the average of the correction values described above.

According to the calculation method detailed so far, the error related to each sensor is mainly a so-called zero drift, in which a constant value is added (or subtracted) over the entire measurement range, and if other errors can be ignored, the error is sufficient. Error correction can be performed, but if the so-called proportional error that increases in proportion to the measured value or the instrumental error unique to each sensor (if no correction has been performed in advance) cannot be ignored, it is necessary to obtain a correction value that corresponds to the measured value. There is.

Regarding the calculation method to obtain the correction value corresponding to this measured value,
Next, it will be explained in detail according to FIG.

Calculating the approximate error value for each sensor requires the number of sensors, that is, a group of actual measured values of each sensor in a flow rate value group at appropriate intervals, which is the same number as the unknown number, and the error related to each sensor is determined by simultaneous equations based on the error approximation formula. The calculation of the approximate value of is based on the assumption that the errors are the same in the actual measured values of the group, so if they differ, it is assumed that the errors are the same. The assumed error is included in the calculated approximate value, and the actual measured value corresponding to the approximate value becomes uncertain.

Therefore, assuming that the calculated approximate value corresponds to the median value of the group of actually measured values, we sequentially created a formula or data table for correction values over the entire required measurement area, and performed correction using this. By sequentially repeating calculation of approximate values based on a group of actual measured values (#6-1) and correction of formulas for correction values based on the approximate values (#6-2), the actual measured values of each sensor are corrected. (#6-3), the aim is to gradually reduce the assumed error and establish the corresponding actual measured value.

(Effects of the Invention) As is clear from the above description, according to the present invention, a fixed diaphragm mechanism and a variable diaphragm mechanism provided in the fluid pipeline to be measured, and a variable diaphragm drive mechanism attached to the variable diaphragm mechanism. and an opening sensor, a first differential pressure sensor that detects the differential pressure before and after the fixed throttle mechanism and the variable throttle mechanism, a second differential pressure sensor, and a fixed throttle mechanism that includes the fixed throttle mechanism and the first differential pressure sensor. A flow rate measurement value based on a signal from a first differential pressure sensor of the throttle system, and an opening sensor of the variable throttle system including the variable throttle mechanism, an opening sensor, and a second differential pressure sensor, and a second
The flow rate measurement error due to the difference in the calculation formula for calculating each flow rate measurement value based on the signal from the differential pressure sensor, and the error related to each of the above sensors based on the "general formula for error propagation" for both calculation formulas. Set up simultaneous equations by substituting the actual measured values of each of the above sensors at the required number of flow rates at appropriate intervals into the error approximation formula, which assumes that the difference between the approximation formulas is equal.
By solving for the errors related to each of the above sensors, an approximate value of the error is calculated, the actual measured values of each of the above sensors are corrected using the approximate value of the error, the measured value of the Nada volume is calculated using the corrected value, and the output is It is formed from an arithmetic controller.

In other words, by combining and utilizing a fixed diaphragm mechanism and a variable diaphragm mechanism to automatically correct the errors of each sensor during operation, it is possible to achieve high precision using easily available general-purpose sensors. It is possible to obtain a flow rate measuring device, and the calibration work can be significantly reduced.

[Brief explanation of the drawing]

Fig. 1 is a system diagram of a flow rate measuring device according to the present invention, Fig. 2-
Fig. 6 is a flowchart showing the procedure for reading the actual measurement values of each sensor, calculating approximate values of errors of each sensor, correction, etc. Fig. 7 is a system diagram of a flow rate measuring device using a conventional fixed diaphragm mechanism; Figure 8 is a diagram showing the relationship between the flow rate measurement error and flow rate due to zero drift of the differential pressure sensor in Figure 7, Figure 9 is a system diagram of a flow rate measurement device using a variable throttle mechanism, and Figure 1
FIG. 0 is a diagram showing the relationship between the flow rate measurement error and the flow rate due to the zero drift of the differential pressure sensor and the zero drift of the opening sensor in FIG. 9. l... Fluid pipe line, 2... Fixed throttle mechanism, 3... Opening degree sensor, 5... Variable throttle mechanism, 6, 7... First.
Second differential pressure sensor, 8... Arithmetic controller. Figure 7: 3 Figure 9: Figure 4: Figure 6: Figure 5: Figure 8

Claims (1)

[Claims] (1) A fixed throttle mechanism and a variable throttle mechanism installed in the fluid pipeline to be measured, a variable throttle drive mechanism and an opening sensor attached to the variable throttle mechanism, and each of the fixed throttle mechanism and variable throttle mechanism A first differential pressure sensor, a second differential pressure sensor that detects the differential pressure between the front and rear, and the fixed throttle mechanism,
a flow rate measurement value based on a signal from a first differential pressure sensor of a fixed throttle system including a first differential pressure sensor, and the variable throttle mechanism;
The difference between the calculation formulas for calculating each of the flow rate measurement values based on the signals from the opening sensor, the opening sensor of the variable throttle system including the second differential pressure sensor, and the second differential pressure sensor, and the difference between the calculation formulas for both calculation formulas. The error approximation formula assumes that the difference in the approximation formula for the flow rate measurement error due to the error related to each sensor is equal to the difference in the approximate formula for the flow rate measurement error due to the error related to each sensor based on the "general formula for error propagation". By creating simultaneous equations in which the actual measured values are substituted and solving for the errors related to each of the above sensors, an approximate value of the error is calculated.The actual measured values of each of the above sensors are corrected using this approximate value of the error, and after this correction, 1. A flow rate measuring device comprising: a calculation controller that calculates and outputs a flow rate measurement value based on the value of .