JPH06103205B2 - Electromagnetic flow meter - Google Patents

Description

TECHNICAL FIELD The present invention relates to an electromagnetic flow meter for applying a magnetic field to a fluid to be measured and measuring its flow rate, and particularly to its excitation method and its accompanying signal processing method. The present invention relates to an improved electromagnetic flow meter.

<Prior Art> An industrial electromagnetic flow meter has conventionally adopted a commercial frequency excitation method in which a commercial power source is used for excitation. The commercial frequency excitation method (a) has a fast response speed and can be manufactured at low cost.
(B) It has the advantage that it is less susceptible to low-frequency flow noise (hereinafter simply referred to as noise) that increases with the flow velocity generated by a slurry fluid or low-conductivity fluid. In addition, the zero point fluctuates if left unattended for a long period of time, for example, about one day.

For this reason, the low frequency excitation method, which excites at a low frequency of 1/2 or less than the commercial frequency, has been adopted. As is well known, the low frequency excitation method has an advantage that a stable electromagnetic flowmeter having a zero point can be obtained. However, since the excitation frequency is low, it is close to the frequency of noise, so that it is easily affected by noise, and this effect becomes noticeable especially when the flow velocity becomes large. Also, it has a drawback that the response becomes slow when damping is applied to reduce the influence of noise.

Therefore, for example, as proposed in Japanese Patent Application No. 62-50809 (invention name: electromagnetic flowmeter), an exciting current component of a commercial frequency and an exciting current component of a frequency lower than this are simultaneously generated in an exciting coil. For the low frequency component of the signal voltage, low frequency demodulation is performed to detect the low frequency component, the low frequency wave operation is executed for this detection result, and the low frequency wave output is output. Of these, the high-frequency component is subjected to high-frequency demodulation, the high-frequency component is detected, the high-frequency wave calculation is executed on this detection result, and the high-frequency wave output is output. A composite excitation method has been proposed in which wave calculation results are added by an adding means to output a flow rate.

In this case, to realize each of the low-frequency wave calculation and the high-frequency wave calculation using a microcomputer, F _Li is the low-frequency wave calculation output, F _Hi is the high-frequency wave calculation output, and T _c is Time constant, ΔT _c is the calculation cycle, e _Li is the input for low frequency wave calculation, e _Hi
Is the input of the high frequency wave calculation, the output F of the low frequency wave calculation is
_Li is F _Li = F _Li-1 + ΔT _c (e _Li −F _Li-1 ) / (T _c + ΔT _c ), and the output F _Hi of the high pass filter is F _Hi = T _c (F _Hi-1 + e calculation is performed as indicated by _{Hi -e Hi-1) / (} T c + ΔT c).

<Problems to be Solved by the Invention> However, this conventional wave operation has the following drawbacks.

Now, assuming that the calculation cycle ΔT _c is 10 ms and the time constant T _c is 10000 sec, ΔT _c / (T _c + ΔT _c ) ≈10 ⁻⁶ T _c / (T _c + ΔT _c ) ≈1-10 ⁻⁶ . And since these numbers are generally not beautiful numbers, the number of digits is necessary until the rounding error can be ignored, and the coefficient in this case requires about 10 decimal digits.

If the number of digits is large like this, the floating operation is performed, but in the case of the composite excitation method, the effective digit number of the coefficient on the high-frequency wave operation side is 10 and the floating operation is not always effective. .

As can be seen from the above description, in order to realize the wave operation in the case of the composite excitation method, it is necessary to execute multiplication of many digits, which has a drawback that the operation time becomes long. In particular, since the electromagnetic flowmeter and the like require real-time processing, it is not preferable that the calculation time be long.

<Means for Solving the Problems> In order to solve the above problems, the present invention provides an exciting means for supplying a magnetic field having two different frequencies, a first frequency and a second frequency lower than the first frequency, and an exciting means for exciting the magnetic field. First demodulation means for sampling and discriminating the signal voltage generated according to the flow rate by the means based on the first frequency and outputting the first demodulation output e _Hi , and sampling the signal voltage based on the second frequency Then, the second demodulation means for discriminating and outputting the second demodulation output e _Li , and the variation coefficient ΔK such that ΔK = ΔT _c / (T _c + ΔT _c ) where the calculation cycle is ΔT _c and the time constant is T _c are used. This intermediate output X _{i is obtained} from the intermediate low-pass wave means for executing the low-frequency wave calculation of _i = X _i-1 + ΔK (e _Hi −X _i-1 ) and outputting the intermediate output X _i and the first demodulation output e _Hi. By subtracting the high-frequency wave means and the second demodulation output e _Li , F _Li = F _Li-1 + ΔK (e _Li −F _Li-1 ) Low-frequency wave means for low-frequency wave and outputting low-frequency wave output F _Li by the following calculation, and addition means for additively synthesizing respective outputs of the high-frequency wave means and the low-frequency wave means to output a flow rate output. And the time constant T _c is selected so that ΔK = 2 ⁻ⁿ where n is an integer.

<Operation> According to the present invention, the coefficient of variation ΔK in low-frequency and high-frequency wave calculation is set to 2 ⁻ⁿ , the time constant is selected so that n becomes an integer, and the actual calculation is performed n times to the right in binary. Achieving high-speed operation without performing troublesome multiplication just by performing shift operation.

<Examples> Examples of the present invention will be described below with reference to the drawings.
FIG. 1 is a block diagram showing an embodiment of the present invention.

Reference numeral 10 is a detector conduit of the electromagnetic flow meter, which has an insulating lining on its inner surface. 11a and 11b are electrodes for detecting a signal voltage. Reference numeral 12 is an exciting coil, and the magnetic field generated by this is applied to the fluid to be measured. An exciting current _If is supplied from the exciting circuit 13 to the exciting coil 12.

The exciting circuit 13 is configured as follows. The reference voltage E ₁ is applied to the non-inverting input terminal (+) of the amplifier Q ₁ and its output terminal is connected to the base of the transistor Q ₂ . The emitter of the transistor Q ₂ is connected to the common COM via the resistor R _f and is also connected to the inverting input terminal (−) of the amplifier Q ₁ . The excitation voltage E _s is applied between the common COM and the collector of the transistor Q ₂ via the series circuit of the switches SW ₂ and SW _{3 and} the series circuit of the switches SW ₄ and SW ₅ connected in parallel with this. . The exciting coil 12 is connected to the connection points of the switches SW ₂ and SW _{3 and} the connection points of the switches SW ₄ and SW ₅ , respectively. Timing signals S ₂ , S ₃ , S ₄ , and S ₅ are switches SW ₂ , SW ₃ , and
Controls the opening and closing of SW ₄ and SW ₅ .

On the other hand, the signal voltage is detected by the electrodes 11a and 11b, and the preamplifier 1
Output to 4. The preamplifier 14 removes the common mode voltage and converts the impedance, and outputs the result to the output terminal 15.

The output of the preamplifier 14 at the output terminal 15 is the analog / digital converter (A / D _L ) 16 and the analog / digital converter (A / D _L ).
Each is converted into a digital signal by D _H ) 17 and stored in a random access memory (RAM) 19 via a bus 18.
The read-only memory (ROM) 20 stores a predetermined calculation program and initial data, and the calculation is performed according to the calculation procedure stored in the ROM 20 under the control of the processor (CPU) 21, and the result is stored in the RAM 19. To be done.

Reference numeral 22 is a clock generator, and the clock generated here is divided into 1 / M by the frequency divider 23 to _obtain C as the system clock S _h.
It is supplied to the PU 21 and the analog / digital converter 17.

CPU21 uses the bus 18 according to the arithmetic program stored in ROM20.
Excitation current to timing signal output port (TO) 24 via
Outputs the timing that determines the waveform of I _f . The timing signal output port 24 outputs timing signals S ₂ , S ₃ , S ₄ , S ₅ for switching the exciting current according to this timing.

Further, the timing signal output port 24 outputs the timing signal S ₁ to the analog / digital converter 16 according to the timing designated by the CPU 21, and samples the output of the preamplifier 14.

On the other hand, a predetermined calculation is executed by the CPU 21 using the data stored in the RAM 19 by the calculation program stored in the ROM 20, and the result of the calculation is stored in the RAM 19 and the digital / analog converter 25 via the bus 18. Is output to the output end 26 as a flow rate output.

Next, the operation of the embodiment shown in FIG. 1 will be described with reference to the timing chart shown in FIG. 2, the flow chart shown in FIG. 3, and the operation chart shown in FIG.

System clock S _h obtained at the output of the divider 23 shown in FIG. 1 is a waveform shown in FIG. 2 (a), which is supplied to the CPU 21.

In Step 1 of Figure 3, CPU 21 in synchronization with the interrupt timing of the system clock S _h (Fig. 2 (g))
The bus 18 is operated by a predetermined arithmetic program stored in the ROM 20.
A timing signal indicating the excitation waveform switching timing is output to the timing signal output port 24 via.

In step 2, the timing signal output port 24 receives this switching timing, and timing signals S ₅ (FIG. 2 (b)), S ₄ (FIG. 2 (c)), S ₃ (FIG. 2 (d)) ,
S ₂ (Fig. 2 (e)) is the switch SW of the excitation circuit 13
Output to ₅ , SW ₄ , SW ₃ , and SW ₂ . Or timing signal S _4
May be simultaneously output to the switches SW ₃ and SW ₄ , and the timing signal S ₂ may be simultaneously output to the switches SW ₂ and SW ₅ . The exciting circuit 13 receives these timing signals and outputs an exciting current _If having a waveform shown in FIG. 2 (f) to the exciting coil 12. As shown in FIG. 2 (h) (i), this excitation waveform is a waveform which forms one cycle with timing number i of 0 to 15 and repeats this. In FIG. 2, the n cycle portion is centered. It is shown. This excitation waveform is a multiplication type waveform obtained by multiplying a low frequency waveform and a high frequency waveform.

Then, the process proceeds to step 3. Step 3 to Step 6
Up to the above, the procedure for reading data from the analog / digital converters 16 and 17 is shown.

In step 3, the data input from the analog / digital converter 17 in each cycle in synchronization with the system clock _Sh (Fig. 2 (a)) is transferred to the bus 18 as shown in Fig. 2 (j).
Via the control of the CPU 21 via a predetermined data area H _i of the RAM 19
To store.

Next, in Step 4, the read timing number i
Is 0, the process proceeds to step 6 if it is not 0, and the process proceeds to step 8 if it is 0.

In step 6, it is judged whether or not the read timing number i is 8, and if not 8, the process proceeds to step 8, and if it is 8, the process proceeds to step 7.

In step 7, the data input from the analog / digital converter 16 is shown in FIG. 2 (l) at the sampling timing based on the timing signal S _l (FIG. 2 (k)) output from the timing signal output port 24. A predetermined data area of the RAM 19 under the control of the CPU 21 via the bus 18,
Store in L _o (N−1), L _o (N), L _o (N + 1), ...
Go to step 5.

Next, in step 6, the analog / digital converter is operated by the sample timing based on the timing signal S _l (FIG. 2 (k)) output from the timing signal output port 24.
As shown in FIG. 2 (l), the data input from 16 is controlled by the CPU 21 via the bus 18, and the predetermined data area of the RAM 19 is ... L ₁ (N-1), L ₁ (N), Store in L ₁ (N + 1), ... And move to step 8.

In step 8, it is determined whether or not the timing number i is an odd number. If it is an odd number, the process proceeds to step 9, and if it is not an odd number, the process proceeds to step 11.

In step 9, high frequency demodulation calculation is performed. In the demodulation calculation, the data H _i stored in the RAM 19 is used, and in the column of the high frequency demodulation calculation e _Hi shown in FIG. 4 stored in the ROM 20 under the control of the CPU 21 at the timing shown in FIG. 2 (m). The calculation is performed using the calculation formula shown and the result is stored in the RAM 19. By this demodulation calculation, the electrochemical DC voltage generated in the electrodes 11a and 11b is removed, and the differential noise is held at a constant value and does not cause an error.

Next, it moves to step 10. Here, the high frequency wave high frequency operation F _Hi is executed.

The high frequency wave operation F _Hi is composed of an intermediate low frequency wave operation step 10A and a subtraction step 10B.

First, the intermediate low frequency wave calculation in step 10A will be described.

In this wave calculation, the calculation cycle ΔT _c and the calculation time constant T _c , ΔK = ΔT _c / (T _c + ΔT _c ) ... (1)
Intermediate low of X _i = X _i-1 + ΔK (e _Hi −X _i-1 ) ... (2) stored in ROM20 based on the control of CPU21 using _Hi and the result of the previous intermediate low-frequency wave calculation. Intermediate wave output X _i
Is stored in a predetermined area of the RAM 19. The calculation result stored in this way is the intermediate low-frequency wave calculation X _{i in} FIG.
It has the form shown in the column. The calculation time constant T _c is set so that the coefficient of variation ΔK used in this calculation is ΔK = 2- ⁿ (3) where n is an integer.

That is, the necessary operation time constant T _c is set by setting n to an arbitrary integer. As a result, the value of the calculation time constant T _c does not necessarily become the round number (round number), but it is not necessary.

Next, the process proceeds to step 10B in FIG. In this step, data e _Hi stored in RAM19 and intermediate low-frequency wave calculation
The subtraction shown in the following equation is executed using the result of X _i , and finally the high frequency wave calculation result F _Hi is calculated and stored in a predetermined area of the RAM 19.

F _Hi = e _Hi −X _i (4) This result has the form shown in the subtraction F _Hi column in FIG.

After all, the above-mentioned intermediate low-frequency wave operation executes the operation of X _i = X _i-1 +2 ^-n (e _Hi −X _i-1 ), which is a binary number and is shifted to the right n times. Since this is equivalent to executing an operation, there is no need for multiplication as in the conventional case, and high-speed operation becomes possible.

Then go to step 11. In step 11, it is judged whether the timing number i is 0 or 8, and if it is 0 or 8, it moves to step 12, and if it is not 0 or 8, it moves to step 13.

In step 12, low frequency demodulation calculation is performed. At the time of demodulation calculation, the data stored in RAM 19 ..., _Lo (N-
1), L _o (N), L _o (N + 1), ... L ₁ (N-1), L
₁ (N), L ₁ (N + 1), ..., In the column of low frequency demodulation calculation e _Li shown in FIG. 4 stored in the ROM 20 under the control of the CPU 21 at the timing shown in FIG. 2 (n). The calculation is performed using the calculation formula shown and the result is stored in the RAM 19.

In step 13, it is determined whether the timing number i is an odd number. If it is an odd number, the process proceeds to step 14, and if it is not an odd number, the process proceeds to step 16.

In step 14, the low frequency wave operation F _Li on the low frequency side is executed.

For wave calculation, data e _Lo , e stored in RAM19
_{Using L8} and the previous wave calculation result, the low-frequency wave calculation program is executed by executing the low-frequency wave calculation program of F _Li = F _Li-1 +2 ^-n (e _Li −F _Li-1 ) ... (5)
_Li is calculated and stored in a predetermined area of RAM 19. The calculation result stored in this way is the low frequency wave calculation F _{Li of} FIG.
It has the form shown in the column.

Under the control of the CPU 21, the low-frequency wave calculation F _Li stored in the ROM 20 shown in FIG.
Stored in a predetermined area of AM19.

The calculation in this case also uses the same coefficient of variation as the coefficient of variation of the high frequency wave operation F _{Hi with} the coefficient of variation ΔK = 2 ⁻ⁿ .
Since a rounding error due to the coefficient does not occur, an accurate calculation can be performed.

Step 15 executes an addition operation. Using the high-frequency wave calculation result F _Hi and the low-frequency wave calculation result F _Li stored in the RAM 19, the addition calculation e _A shown in FIG. 4 and stored in the ROM 20 under the control of the CPU 21 is shown. Perform the operation with the operation expression and RA the result.
Store in M19, move to step A, and finish flow rate calculation.

As described above, the signal voltage including the low frequency and the high frequency detected by the electrodes 11a and 11b is read by being divided into the low frequency side and the high frequency side by using the microcomputer, and the low frequency side is low. It demodulates with a high frequency and the output is passed through the low-pass wave means, and the high frequency side is demodulated with a high frequency and the output is output through the high-pass wave means. By adding and synthesizing and outputting, a flow rate output with stable zero point, strong against noise, and good response can be obtained.

<Effect of the Invention> According to the present invention as specifically described above with reference to the embodiments, the value of the operation time constant is set so that the coefficient of variation ΔK becomes 2 ^−n. It is possible to perform high-speed calculation without shifting, which requires a large number of digits for calculation of the high-frequency wave, and only requires a simple shift operation, and high-frequency wave calculation and low-frequency wave calculation. Since the same variation coefficient ΔK is used for the rounding error, the rounding error is reduced and the accuracy is improved.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention, FIG. 2 is a waveform diagram explaining the operation of the embodiment shown in FIG. 1, and FIG. 3 is a signal processing procedure of the embodiment shown in FIG. FIG. 4 is an operation diagram showing an operation procedure in the flow of FIG. 10 ... Conduit, 12 ... Excitation coil, 13 ... Excitation circuit, 16, 17 ... Analog / digital converter, 18 ... Bus, 19 ... Random access memory, 20 ... Read only memory, 21 ... Microprocessor, 22 ... Clock generator , 24… Timing signal output port.

Claims (1)

[Claims] 1. A first frequency and a second frequency lower than the first frequency.
Exciting means for supplying magnetic fields having three different frequencies, and signal voltage generated by the exciting means and corresponding to the flow rate is sampled and discriminated based on the first frequency to output a first demodulation output e _Hi . First demodulating means for
Second demodulation means for sampling and discriminating the signal voltage based on the second frequency and outputting a second demodulation output e _Li ; ΔK = ΔT _c / (T with a calculation cycle of ΔT _c and a time constant of T _c. an intermediate low frequency wave means for outputting _{X i = X i-1 +} ΔK (e Hi -X i-1 intermediate output X _i running low frequency wave operation) with _c + [Delta] T _c) becomes the coefficient of variation [Delta] K From the first demodulation output e _Hi to this intermediate output X _i
The subtraction to the high-frequency wave means comprising a subtracting means for outputting a high-frequency wave output F _Hi, the second with a demodulated output _{e Li F Li = F Li-} 1 + ΔK (e Li -F Li-1 ) Addition of low-frequency wave means for low-frequency wave and outputting low-frequency wave output F _Li by the following operation, and output of flow rate by additively combining the respective outputs of the high-frequency wave means and the low-frequency wave means And a time constant T _c is selected so that ΔK = 2 ⁻ⁿ where n is an integer.