JP2000081346A - Electromagnetic flowmeter not filled with water - Google Patents

Abstract

PROBLEM TO BE SOLVED: To suppress fluctuation or error in the measurement off low rate when the flow rate increases abruptly or a solid matter rubs an electrode. SOLUTION: Fluctuation in an electrode signal is captured by an MPU59A when the flow rate increases abruptly or a solid matter rubs an electrode and when it exceeds a specified level, a short circuit switch S2 is turned off as shown on Fig. (b). At the same time, the flowmeter is operated in a mode for not exciting a lower exciting coil but exciting only an upper exciting coil. When fluctuation in the electrode signal is settled below the specified level, operation is returned back to normal exciting mode. The short circuit switch S2 is also reset to repeat on/off operation. More specifically, function for correcting conductivity and compensating nonexcitation are interrupted temporarily when the electrode signal fluctuates significantly.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to an improvement of an electromagnetic flow meter.

[0002]

2. Description of the Related Art [First Prior Art] The applicant has a method of measuring a flow rate of a fluid flowing in a non-full state in a flow path based on the principle of an electromagnetic flowmeter, and a flow rate detection method used in the method. (Application filed January 22, 1991;
No.5631). The content of this application is international publication number WO93
And is known by reference to US Pat.

The flow detection method and the flow detector are shown in FIGS.
A description will be given based on FIG. Hereinafter, this is referred to as a first conventional technique. 6 to 8, reference numeral 1 denotes a channel having a circular cross section,
Reference numerals 2 and 2 denote a pair of electrodes provided at positions symmetrical with respect to a vertical line passing through the center of the flow path 1. Reference numerals 3A and 3B denote first and second excitation coils, which have different magnetic flux density distributions BA and BB. Occurs during the period (hours). Reference numeral 4 indicates a flow detector having such a structure.

An excitation circuit 5 alternately excites the first and second excitation coils 3A and 3B according to the signal of the timing circuit 6. 7 preamplifier for amplifying and outputting a voltage induced between the electrodes 2, 2, S ₁ is a toggle switch, and the switching operation by a signal from the timing circuit 6, and the switching of the excitation timing of the two excitation coils 3A and 3B The switching operation is performed in synchronization with the switching to the side a when the first excitation coil 3A is excited and to the side b when the second excitation coil 3B is excited.

[0005] 8A and 8B are offset compensation by inputting the signal of a contact and b-contact of the switch S _1, the sample-and-hold amplifying circuit, a CPU circuit 9, 10 amplifier 8
An A / D converter circuit for converting analog signals from A and 8B into digital signals. Reference numeral 11 denotes an arithmetic circuit, which has a program for performing arithmetic operations described later. Reference numeral 12 denotes an output terminal for outputting a flow rate signal as a calculation result.

FIG. 9 is a timing chart of the electromagnetic flow meter of FIG. 6, in which, from the top, the signal of the timing circuit 6, the exciting current of the first exciting coil 3A, the exciting current of the second exciting coil 3B, and the changeover switch S _1. Operation, output of preamplifier 7,
The input of the amplifier circuit 8A and the input of the amplifier circuit 8B are shown.

[0007] The following procedure is used to measure the flow rate of the non-full fluid flowing through the flow path 1. Procedure 1. Using a flow path having the same cross-sectional shape as the flow path 1 for which the flow rate is to be measured, changing the water level h while keeping the gradient of the flow path constant, and previously measuring the outputs OA and OB corresponding to the flow rate Q at the water level. Keep it. In the same channel, water level and flow rate are 1: 1
In a relationship. The outputs OA and OB are measured using the flow rate detector 4 based on the principle of an electromagnetic flow meter, and the same flow rate Q is measured for the first and second outputs.
10 is a flow rate signal when measured by the excitation coils 3A and 3B, respectively (FIG. 10A).

Procedure 2. The unknown flow rate Q 'in the flow path 1 through which the fluid to be measured flows is measured by the flow rate detector 4 and the output O
A 'and OB' are obtained. Procedure 3. The ratio OB '/ OA' between OB 'and OA' is determined, and the ratio OB / OA between OB and OA determined in step 1 is OB '/ O.
The flow rate Qα that becomes the same value as A ′ is obtained from the data of the procedure 1 (FIGS. 10B and 10C), and the procedure 1 for the flow rate Qα is obtained.
OAα / Qα under the condition of the flow rate Qα in the procedure 1 is calculated from the output OAα (FIG. 10D).

Procedure 4. The output OA 'measured in step 2;
From the sensitivity OAα / Qα obtained in step 3, the unknown flow rate Q ′
Is calculated by the following equation. Q ′ = OA ′ · Qα / OAα In this way, without detecting the water level, the unknown flow rate Q ′
We proposed a method of seeking.

[Second Prior Art] In the flow rate measuring method described above, the flow rate detector 4 to be used is designed so that the positional relationship between the electrode and the earth and the relationship between the electrode shape and the excitation magnetic flux density distribution are both the same.
Since the symmetry is easily lost, there is a disadvantage that a measurement error occurs due to the influence of the conductivity.

[0011] Therefore, the applicant has proposed an electromagnetic flow meter for non-full water which can solve such a drawback, as described in International Publication No. WO93 / 053.
67 (hereinafter referred to as a second conventional technique).
This non-filled electromagnetic flowmeter has a main body portion 24 having an upper exciting coil Cu and a lower exciting coil C1, which can be individually excited, and a pair of opposing electrodes 22, and the following requirements (a) to (g). And a transducer portion 32 having the same. (Figure 1
1, 12).

(A) An excitation timing circuit 26 is provided for notifying the excitation circuit 25 of the timing of excitation and notifying the amplifier circuit AMP of the timing. (B) generating a timing signal for controlling the opening and closing of the short-circuit switch S _2, at the same time providing the conductivity measurement timing circuit 28 to notify the timing to the arithmetic processing unit 31.

(C) An excitation circuit 25 for receiving a signal from the excitation timing circuit 26 and exciting the two excitation coils Cu and Cl is provided. A short circuit resistance Rs shorting from (d) electrode 22 to ground the conductor leading to the preamplifier 27, a signal from the conductivity measurement timing circuit 28, connect the short circuit resistance Rs, a switch S ₂ to or disconnect.

(E) A preamplifier 27 for amplifying a signal voltage between the electrodes 22 is provided. (F) An A / D conversion circuit 30 for A / D converting the output of the amplification circuit AMP and an arithmetic processing circuit 31 for processing this data and outputting a flow rate output signal are provided.

FIG. 13 shows a timing chart of one typical measurement cycle. One cycle of measurement is performed in periods a to f in the figure. First, each signal will be briefly described. (I) The upper excitation coil Cu is excited during periods a and b and periods e and f.

(Ii) The lower excitation coil C1 is excited during periods c and d. (Iii) conductivity measurement shunt switch S ₂ is closed only during the periods e, f. The difference excitation in the output of the (iv) pre-amplifier 27, the output appears in response to the position of the short-circuit switch S _2.

(V) The first amplifier circuit AMPu integrates the output of the Prian 27 while being excited by the upper excitation coil Cu, and holds and outputs this when integration is completed. Accordingly period a, the integral value of b is fixed at the time t _2, the period e, the integral value of f is determined at time t _6.

Next, the processing procedure of the obtained signal will be described. 1. The following three types of data are measured in advance and stored in the arithmetic processing unit 31. First amplifier circuit AMP at upper excitation (or lower excitation)
ratio of output Eu of u to actual flow rate Q at that time g≡Q / Eu
Becomes a constant value if the water level h and the conductivity c are determined. G (h, c) is measured in advance.

First amplifier circuit AMPu at upper excitation
And the second amplifier circuit AMPl at the time of lower excitation.
Is also constant when the water level h and the conductivity c are determined. This p (h, c) is measured in advance.

The ratio s≡ between the output Eu when the short-circuit switch S _{2 is} open due to the upper excitation and the output Eu ′ when the short-circuit switch S _{2 is} closed.
Eu '/ Eu is also determined by the water level h and the conductivity c. This s (h, c) is measured. 2. The ratio p ₀ = El / Eu is obtained from the signals Eu and El obtained during the actual measurement. If p (h, c) = p ₀ , one curve can be drawn on the hc plane as shown in FIG.
Of course, this curve is on the p = p ₀ plane,
Use the curve orthogonally projected on the c-plane.

The measuring method described in the prior art (first prior art) is based on the premise that p ₀ and water level h have a one-to-one relationship. That, p _0, as shown in dotted line in FIG. 14, it has been considered to be represented by a straight line parallel to the longitudinal axis.
In the second prior art, p ₀ is represented by a solid line in FIG.
There is no one-to-one relationship with the water level h, and the conductivity is involved there, which solves the problem newly found by the inventors, which causes a measurement error.

That is, the output E of the upper coil and the lower coil
The water level cannot be determined accurately from only u and El. 3. Therefore, the present inventors have considered that the output of the upper coil and / or the lower coil is subjected to some processing, and the obtained result is a curve on the hc plane as in FIG.
If such a curve is obtained, an intersection between the curve and the curve in FIG.
And the conductivity c can be specified. In other words, the water level h is obtained as a solution by obtaining two equations having two variables h and c. At that time, the conductivity c is determined at the same time.

The present inventors have paid attention to the aforementioned ratio s≡Eu ′ / Eu. Then, s ₀ = Eu ′ / Eu is obtained, and s ₀ = Eu ′ / Eu is obtained.
If (h, c) = s ₀ , one curve is also shown in FIG.
It was obtained as follows. The orthographic projection to the hc plane is the same.

Assuming that the output impedance of the flowmeter main body 24 determined by the water level h in the flow tube 1 and the conductivity c of the fluid to be measured is Rw, the flow signal E generated in the flowmeter main body is as shown in FIG. Then, the voltage is divided by the output impedance Rw and the input impedance Ri of the preamplifier 27 and enters the preamplifier 27. However, since Rw ≒ Ri, Ei 通常 E.

Here, when the short-circuit switch S ₂ is closed,
Similarly, if Rs≪Ri, Ei ≒ Rs ・ E / (Rw + Rs). Therefore, according to the definition of s, s = Rs / (Rw + Rs). Rs is known and Rw (h, c), so s
(H, c).

4. In the first prior art, the water level h has a one-to-one relationship with the flow rate Q, and p ₀ = El / Eu (= OB / O
The actual flow rate was calculated on the assumption that A) had a one-to-one relationship with the flow rate Q (that is, the water level h).
_{Since 0} = p (h, c), the method described in the first prior art cannot be used for performing more accurate measurement.

Therefore, the inventors have noticed that the ratio g≡Q / Eu described above is a function of the water level h and the conductivity c. That is, since the intersection of the two curves in FIGS. 14 and 15 represents the current water level h ₀ and the conductivity c ₀ (FIG. 16), g =
_{g 0 (h 0, c 0} ) by g ₀ is obtained. According to the definition of g, the current flow rate Q ₀ is obtained as Q ₀ = g ₀ Eu.

The method uses a parameter h that affects g.
And c are determined by solving a simultaneous equation of s (h, c) = s ₀ p (h, c) = p ₀ , and the conductivity C is corrected to correct an error due to a change in conductivity. I do.

FIG. 18 shows a specific example of the second prior art. As shown in FIG. 18, when the inner diameter of the flow path 1 is φ200, the short-circuit resistance Rs is 100Ω, and the input impedance of the preamplifier 27 is 100 MΩ, the curves corresponding to FIGS. Become.

The portion where the water level h is 1.0 or more in FIG. 20 is plotted for the sake of convenience in a state where the fluid is being pumped while being full. Next, the measuring method of this specific example will be described with reference to the flowchart of FIG.

FIG. 22 shows the ratio g≡Q / Eu and the water level h.
4 is a graph showing a relationship between the temperature and the conductivity c. This FIG.
19, and the relationship between FIG. 19 and FIG.
Is stored in advance in the memory of the computer.

[0032] In step 1, the output Eu of the upper coil when the short-circuit switch S ₂ open, the switch S ₂ detects the output El of the output Eu 'and lower coil when closed,
It is stored in the register of the arithmetic processing unit 31.

In step 3, the arithmetic circuit of the arithmetic processing unit reads Eu and El stored therein from the register, calculates p ₀ = El / Eu, and stores the result in another register.

At step 5, similarly, S₀= E
Calculate l '/ Eu, save the result in another register
Good. In step 7, p obtained in step 3₀The value of
Compared to the data of FIG. 19 stored in the memory,
P obtained from 19 data₀Select data closest to
I do. On the other hand, S obtained in step 5₀The value of
Compared with the stored data of FIG.
Obtained S₀Select the data closest to. in this way
Water level h at the time of measurement from the intersection of the two data obtained₀
And conductivity c ₀Is obtained.

In step 9, the water level h ₀ and the conductivity c ₀ to g ₀ determined in step 7 are specified with reference to the data of FIG. 22 stored in the memory. Step 1
In 1, the read out output Eu of the upper coil from the register, multiplied by the ratio g ₀ obtained in the Eu and step 9 to calculate the actual flow rate Q _0.

[Third Prior Art] The applicant also has a flowmeter combining a weir and an electromagnetic flowmeter (a so-called weir-type electromagnetic flowmeter).
Japanese Patent Application No. 4-70846 discloses a weir-type electromagnetic flow meter that can measure the flow rate even when the water is full, similarly to a normal electromagnetic flow meter, and does not need to measure the water level when measuring the flow rate when the water is not full.
(JP-A-5-273015).

The third prior art weir-type electromagnetic flow meter has:
A weir, a pair of electrodes arranged in a horizontal direction facing the flow tube on or upstream of the weir, and an excitation coil provided on at least one of the upper and lower portions of the flow tube to generate a vertical magnetic field. Was what it was.

At the critical point on the weir and on the upstream side of the weir, a certain relationship is established between the water depth and the flow velocity, and this relationship is a functional relationship uniquely determined by the shapes of the flow pipe and the weir. Therefore, the flow rate at the time of non-full water is calculated from the signal voltage of the electrode which is proportional to the flow velocity, using the above-described functional relationship and the like.

[0039] Fourth prior art] Applicant, in the second prior art, precisely to remove the adverse effect of exponential variation of the dc offset voltage is short-circuited switch S ₂ generated with the opening and closing Japanese Patent Application No. 8-37849 (Japanese Unexamined Patent Publication No. 9-229733) has proposed an electromagnetic flowmeter capable of performing an accurate flow measurement.

In the fourth prior art, the short-circuit switch S
_An electromagnetic flowmeter that calculates and calculates the conductivity of the fluid based on the flow signal when the switch ₂ is turned off and the flow signal when the switch ₂ is turned on. flow rate at the sought after flow rate signal, the excitation state as well as short-circuit switch S ₂ is determined the flow rate signal at the time and on the off de-energized, de-energized from the flow signal at the excitation state by subtracting signals, shorting switch S ₂ is performed with the aim to improve the DC offset voltage removed to measurement accuracy the adverse effects of fluctuations in the generated along with the turning off-on, so to speak a non-excitation compensation.

[Fifth Prior Art] The inventors of the present application are conducting an improvement study on a weir-type non-filled electromagnetic flowmeter based on the measurement principle of Japanese Patent Application No. 4-70846. Then, the following two points (a) and (b) were found in the development subject.

(A) Second prior art (international publication number WO
93/05367), the measurement accuracy is better when the conductivity is corrected. (B) Measurement accuracy is better when non-excitation compensation is performed based on Japanese Patent Application No. 8-37849.

Therefore, the improved overall block configuration of the weir type non-full electromagnetic flowmeter is as shown in FIG. A weir 52 is provided in the flow tube 51, and the weir 52 has electrodes 53 and 5.
3. An earth electrode 54 is provided. The upper excitation coil 55 and the lower excitation coil 56 are separately excited by an excitation circuit 57. Short-circuit switches S ₂ , S ₂ and short-circuit resistors Rs, Rs are connected between the electrode 53 and the ground.

Reference numeral 58 denotes an amplifier comprising a differential amplifier, and a microprocessor (hereinafter abbreviated as MPU) 59 generates an excitation timing signal to the excitation circuit 57, a sampling timing signal to the A / D converter 60, and outputs Control and calculation. 61 is a correction table, and 62 is an output circuit.

FIG. 24 shows a typical example of the excitation and short-circuit switch on / off (ON / OFF) procedures of the weir-type non-full electromagnetic flowmeter of FIG. 23 and a change in the input voltage (signal voltage) ein of the amplifier.

The signals generated at the electrodes 53, 53 are
8, the input voltage (signal voltage) ein of the amplifier is hardly affected by the impedance due to the conductivity of the fluid to be measured (hereinafter simply referred to as water).
The input impedance Rin of No. 8 is normally set to a high value of several tens to several hundreds MΩ (see FIG. 25).

Short-circuit switch S between electrode 53 and ground potential
The resistance value of the short-circuit resistor Rs connected via ₂ is set to a value similar to the assumed impedance of water, that is, as low as several hundred Ω, so as to obtain the conductivity information of water. By doing so, the dynamic range of the obtained conductivity information is widened, which is convenient for correcting an error due to a change in conductivity (conductivity correction).

Short-circuit switch (hereinafter abbreviated as short-circuit SW)
A period e1 to S ₂ is excited in and under the excitation coil 56 OFF, the lower and the short-circuit SWS ₂ is ON excitation coil 56
The period e3 during which excitation is performed is a period for obtaining conductivity information for correcting a measurement error caused by a change in the conductivity of water.

In the period e1, since the short-circuit SW is OFF, the signal sampling value becomes e. In period e3,
Assuming that the impedance of water is Rw, the signal voltage is e / 2, and the input impedance of the amplifier 58 is sufficiently larger than the impedance Rw of water as shown in FIG. 26, the + input voltage E ₊ of the amplifier 58 as a differential amplifier is provided. Is represented by the following [Equation 1].

[0050]

(Equation 1) Amplifier 58 - side input voltage E _- is similarly represented by the following expression (2).

[0051]

(Equation 2) Therefore, the input voltage (signal voltage) ein of the amplifier 58 becomes the following [Equation 3].

[0052]

(Equation 3) Since the ratio between the sampling value e in the signal voltage period el and the sampling value in the period e3 is a function of the impedance of water as in the following [Equation 4], [Equation 4] is modified to obtain the conductivity of water. The impedance Rw relating to the information can be obtained [Equation 5].

[0053]

(Equation 4)

[0054]

(Equation 5) However, when the operation of turning on the short-circuit SW in the period e3 is performed, the electrode 53 and the ground electrode are connected with a low resistance Rs, so that a so-called corrosion current flows between the electrodes, and the potential of the electrode and the potential of the ground electrode are reduced. As shown in FIG. 27, it changes with time to approach each other.

In these descriptions, the expression “electrode potential” or “electrode potential” as shown in FIG. 27 indicates a so-called offset potential obtained by subtracting a signal voltage (see FIG. 28).

In the periods e1 and e2, the apparent input impedance of the amplifier 58 as viewed from the electrodes 53 and 53 becomes higher because the short-circuit SW is turned off, so that the electrode potential and the earth potential try to return to their natural potentials. Since both of these potential changes occur exponentially with respect to time, after all, e1 and e2
In each of the period e3 and the period e3, the electrode potential always changes exponentially. This exponential change is taken in by sampling and treated as a flow signal, resulting in a measurement error (see FIG. 29).

Therefore, immediately after this period of e1, e2, and e3, a non-excitation period e1 ', e2', and e3 'is provided, in which only the operation of the short-circuit SW is performed at the same timing without performing excitation. , There is no flow signal at all, but a signal containing the same exponential potential change of the electrode as during the period e1 is obtained. Therefore, a value obtained by subtracting the sampling value during the period e1 ′ from the sampling value during the period e1 is newly sampled. By setting E1, a sampling value excluding the effect of the exponential potential change of the electrode is obtained (see FIG. 30). Hereinafter, E2 and E3 are obtained in the same manner.
This is called non-excitation compensation.

Since a one-to-one relationship is established between the signal E2 and the flow rate when the conductivity is constant, the relationship between E2 and the flow rate is measured in advance using the conductivity as a parameter at several points.
It is stored as a correction table A. On the other hand, the relationship between the value of the ratio E3 / E1 and the flow rate is measured using the conductivity as a parameter, and stored as the correction table B.

The contents of the correction tables A and B are shown in [Table 1] and [Table 2], respectively. These two correction tables A and B are stored in the memory constituting the correction table 61 in FIG. It is something to keep.

[0060]

[Table 1]

[0061]

[Table 2] It should be noted that the correction table A in Table 1 and the correction table B in Table 2 are shown in a graph, and are shown in FIGS. 31 and 32, respectively.
become that way.

Then, the flow rate Q is calculated by the flow rate calculation algorithm shown in FIG. The calculation is performed by the MPU 59. FIG.
4 is a flow rate calculation algorithm shown in FIG.
FIG. 6 is an explanatory diagram for obtaining the relationship between the flow rate and E2 when the conductivity is C using FIG.

FIG. 35 is an explanatory diagram for obtaining the flow rate Q from the value of E2 measured and obtained by the flow rate calculation algorithm of FIG. FIGS. 36 (a) and 36 (b) are explanatory diagrams showing the relationship between E3 / E1 and the conductivity when the flow rate is Q using the correction table B of [Table 2] in the flow rate calculation algorithm of FIG. is there.

FIG. 37 is an explanatory diagram for obtaining the conductivity C 'by illuminating E3 / E1 measured by the relational expression between the obtained E3 / E1 and the conductivity by the flow rate calculation algorithm of FIG. In this way, it is possible to accurately measure even a very low flow rate as compared with the third conventional technique. FIG. 38 illustrates the accuracy of such a weir type non-full electromagnetic flowmeter.

[0065]

In the above-mentioned non-filled electromagnetic flowmeter, the natural potential of the electrode is determined by the electrochemical equilibrium of a portion of the electrode surface in contact with water.
Thus, when the flow rate is small, the portion above the electrode is above the water surface, and is not in an electrochemical equilibrium state.

If the flow rate suddenly increases, the electrode portion that has not been soaked in water comes into contact with water. Therefore, in order to stabilize the potential of the electrode, the electrode portion including this portion is newly added. An electrochemical equilibrium must be achieved over the entire contact surface between water and water. This can take from a few seconds to a few minutes at long times. During this time, the potential of the electrode is not stable and changes irregularly with a voltage amplitude much larger than the signal voltage.

In both the non-excitation compensation and the conductivity correction, it is assumed that the flow rate, the conductivity, and the electrode potential are constant over time during one cycle of the excitation / sampling procedure.
In this state, the premise does not hold.

For example, the sampling periods e1 and e1 '
In general, the change in the electrode potential can be regarded as the same, so that E1 =
This can be e1-e1 ', but if the electrode potential fluctuates unstablely, this operation will instead introduce an irregular error into E1. Periods e2 and e
The same applies to 2 ′, periods e3 and e3 ′.

The conductivity correction is performed by using the fact that E1 and E3 have a certain value in relation to each other. In this situation, when E1 and E3 both have large variations, Instead, an error is introduced by the correction.

As described above, when the flow rate changes abruptly, particularly when the flow rate suddenly increases, the natural potential of the electrode changes abruptly and tends to be temporarily unstable for a while. As a result, there is a problem that the non-excitation compensation and the conductivity correction may cause a variation and an error in the measured flow rate.

The so-called slurry noise generated when a solid rubs the electrode is, in essence, noise due to instability of the electrode potential. There was also a problem that the ratio correction caused a strong output.

Accordingly, an object of the present invention is to provide a non-filled electromagnetic flowmeter which can solve these problems.

[0073]

In order to achieve the above object, the invention according to claim 1 includes an upper excitation coil disposed above a flow path, a lower excitation coil disposed below the flow path, and a side face of the flow path. In a non-filled electromagnetic flowmeter having a pair of electrodes disposed opposite to each other and having a conductivity correction function and a non-excitation compensation function, the flow rate suddenly increases, and the fluctuation of the electrode signal when a solid rubs the electrode is described. This non-full electromagnetic flowmeter is characterized in that the conductivity correction and the non-excitation compensation are temporarily stopped, and the flow rate is calculated using the conductivity just before the variation of the electrode signal is captured.

According to a second aspect of the present invention, there is provided the non-filled electromagnetic flow meter according to the first aspect, wherein a weir is provided in the flow path, and a pair of electrodes are disposed on an inner surface of the weir. Therefore, when temporarily stopping the conductivity correction and the non-excitation compensation,
It is characterized in that the excitation mode is set to the excitation only in the upper direction.

[0075]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, preferred embodiments of the present invention will be described with reference to examples. [Embodiment 1] The weir-type non-full electromagnetic flow meter shown in FIGS. 1 (a), 1 (b), 1 (c) and 2 is the fifth type described in FIGS.
The prior art weir type non-full electromagnetic flowmeter can be implemented by changing only the software as described later without changing the block configuration shown in FIG. 23 in terms of hardware. In comparison with FIG.
Only the PU code is changed from 59 to 59A, and the functions of the other components are the same and the same codes are shown.

In the first embodiment, the period e1 shown in FIG.
通常 E2−E2 moving average value >> 60 mV ... Judgment condition 1 and normal excitation continuation frequency> 20 times ... Judgment condition At 2, the procedure of the excitation and the short-circuit switch ON / OFF is changed from the normal excitation state in FIG. 24 to the excitation state only in the upper part of FIG.

FIG. 1 (c) shows the algorithm for calculating the flow rate when only the upper part is excited. In the flow rate calculation algorithm of FIG. 1 (c), the last normal excitation means the normal excitation immediately before the change to the excitation only above.

Also, when the state of excitation is repeated only in the upper direction, | ΔE2 | exponential average <3.75 mV ... Judgment condition 3 or the number of times of continuous excitation> 120 times in judgment condition 4 Next, the procedure of the excitation and short-circuit switch ON / OFF is returned from the state of excitation only to the upper part of FIG. 1B to the state of normal excitation of FIG. The normal excitation state in FIG. 2 is the same as that in FIG.

The meanings of the terms used in the conditions 1 to 4 are as follows. (1) | ΔE2 | exponential average A value obtained by exponentially averaging the absolute value of the difference between the most recently measured E2 and the immediately preceding E2 a plurality of times. For example, in this embodiment, the number is set to 16 times.

The exponential average of n times is a time series of measured values.
_{· A n-2, a n} -1, a n, with respect to a _{n +} 1 ···,

[0081]

(Equation 6) Is the sequence value {An} obtained by This is an index indicating the degree of variation of E2. When the value of | ΔE2 | exponential average is large, it is estimated that instability of the electrode potential due to slurry noise or a sudden change in the flow rate still remains, so it is determined that excitation is continued only above.

If this | ΔE2 | exponential average becomes smaller to some extent, it is estimated that the instability has been eliminated, and it is determined to return to normal excitation. Numerical values such as the voltage and the number of continuations, which are the criteria for determination in this embodiment, are optimal values in the embodiment, but may vary depending on the characteristics of the entire flow meter, the amplification degree of the amplifier, and the assumed measurement symmetry. There is no intention to limit to this value.

(2) Number of times of continuous excitation only The number of times excitation has been performed after switching from the normal excitation to the upward only excitation. For example, only the number of excitations shown in FIG. 1B is six.

The reason for setting a limit on the maximum number of times is as follows.
This is to avoid a state where the flowmeter is not normally excited when the flowmeter is installed in an environment where there is much variation. Since the conductivity cannot be measured by excitation only above, it is necessary to return to normal excitation at least at a certain interval and measure data for conductivity correction.

(3) | E2-E2 Moving Average Value | The absolute value of the difference between the most recently measured E2 and the moving average value of E2 obtained from the measured value of E2 immediately before E2. The number of collected moving averages is eight in this embodiment.

The n moving averages are

[0087]

(Equation 7) Is the sequence {Ai} obtained by | E2−E2 moving average value | below a certain value means that the variation of E2 stays within a certain range around the average value, and the electrode potential due to slurry noise or sudden change of flow rate. Since it is estimated that there is no instability, the normal excitation continuation is determined.

On the other hand, if the electrode potential becomes unstable due to slurry noise or a sudden change in the flow rate, an extremely large change occurs in E2, so that | E2-E2 moving average value |
The condition that is below a certain value is violated. Therefore, by taking this into account, it can be determined that only the upper side is excited.

(4) Normal excitation continuation number The number of times excitation has been performed since the switching from the excitation to the normal excitation. When the variation is large, the excitation is continued only on the upper part, but if the excitation is continued only on the upper part for a long time, new data for conductivity correction cannot be obtained, so it corresponds to the change in water conductivity during that time Can not do. Therefore, as described in the section (2), it is necessary to return to the normal excitation at least at certain intervals.

However, when the excitation is returned from the upper excitation to the normal excitation, the data E1 and E3 for accurate conductivity correction may not be obtained due to the dispersion of signals in one measurement operation. It is usually necessary to continue excitation and obtain an average E1 · E3. Therefore, once the excitation is returned to the normal excitation from the upper excitation, the normal excitation is continued at least for a specified number of times.

In the normal excitation procedure shown in FIG.
The flow rate can be calculated only once during 1
As shown in (b), only the excitation procedure can be performed three times, so that the effect of effectively increasing the excitation frequency by six times can be obtained, and the number of measurements per unit time increases, thereby further reducing variations. Can be done.

FIG. 3 shows the state of E2 when the non-filled electromagnetic flowmeter having a diameter of 150 mm of Example 1 was installed in a toilet water pipe and tested. In response to the first change in E2 caused by the drainage generated around the time 1080 seconds, the flow meter has shifted to the excitation mode only at the top. Symbol A indicates the period of the excitation mode only at the top.

FIG. 4 shows the value of E2 when the conventional non-full electromagnetic flowmeter is incorporated in the water distribution pipe of the same toilet, at time 461.
Drainage occurs around 0 seconds, and the signal fluctuates greatly. It is clear that the variation is effectively suppressed in FIG. 3 as compared with FIG.

In this embodiment, since various excitation and sampling timing signals are supplied from the MPU by a software command, the block configuration shown in FIG. Only by such a configuration.

When the timing circuit is separately constituted by hardware, for example, a timing circuit for generating a timing signal for each of the normal excitation and the upper excitation is prepared, and is switched by a switching command signal from the MPU. It can also be implemented by doing.

[Embodiment 2] The present invention is not limited to the weir-type non-full electromagnetic flowmeter, but is also applicable to the second prior art type non-full electromagnetic flowmeter proposed in International Publication No. WO93 / 05367. Can be.

In this case, there is no weir 52 and the electrodes 53, 53
Are disposed to face the inner wall side surface of the flow tube 51. In addition, since the calculation of the flow rate requires both the signal when excited by the lower excitation coil and the signal when excited by the upper excitation coil, the excitation in the first embodiment when only the upper excitation corresponds to the excitation. The procedure of ON / OFF of the short circuit switch is changed as shown in FIG. 5 instead of FIG.

The procedure for calculating the flow rate in this method is disclosed in detail in International Publication No. WO93 / 05367, and the description is omitted here.

[0099]

Since the non-full electromagnetic flow meter of the present invention is constructed as described above, when the flow rate changes suddenly, especially when the flow rate increases rapidly, the conductivity correction and the non-excitation compensation are temporarily performed. In this case, the variations and errors that were introduced by doing these do not enter. As a result, variations and errors in the flow measurement were reduced.

In addition, variations and errors due to so-called slurry noise in which the solids rub the electrodes were reduced. According to the second aspect of the present invention, the variation is further reduced as a result of the increase in the effective excitation frequency.

[Brief description of the drawings]

FIG. 1A is a block diagram of an embodiment of the present invention, and FIG.
7A is a timing chart when only the upper portion is excited, and FIG. 9C is a flow rate calculation algorithm when only the upper portion is excited.

FIG. 2 is a timing chart at the time of normal excitation in the embodiment of FIG. 1;

FIG. 3 is a diagram showing a variation of E2 in the embodiment of FIG. 1;

FIG. 4 is a diagram showing a variation of E2 in the related art.

FIG. 5 is a timing chart of another embodiment of the present invention.

FIG. 6 is a block diagram of a first related art.

FIG. 7 is a front view of a first prior art flow rate detector.

FIG. 8 is a sectional view taken along the line AA ′ of the flow detector of FIG. 7;

FIG. 9 is a timing chart of the prior art of FIG. 6;

FIGS. 10A and 10B are diagrams for explaining the procedure of a flow measurement method according to the first prior art, in which FIG. 10A is a flow meter output diagram, FIG. 10B is a diagram showing a flow meter output ratio, and FIG. The figure explaining the procedure which calculates | requires the flow rate Q (alpha) of the same water level from the output ratio obtained from the flow rate,
(D) is a diagram illustrating a procedure for obtaining the sensitivity from the output curve OA.

FIG. 11 is a block diagram of a second prior art.

FIG. 12 is a diagram showing details of a part of the block diagram of FIG. 11;

FIG. 13 is a timing chart of the prior art of FIG. 11;

FIG. 14 is a diagram of a p-curve on the hc plane.

FIG. 15 is a diagram of an s curve on the hc plane.

FIG. 16 is a diagram of a p-curve and an s-curve on the hc plane.

FIG. 17 is a main part of the electric circuit of the prior art of FIG. 11;

FIG. 18 is a specific example of the electric circuit of FIG.

19 is a diagram of a p-curve on the hc plane according to the specific example of FIG. 18;

20 is a diagram of an s-curve on the hc plane according to the specific example of FIG. 18;

FIG. 21 is a flowchart illustrating a flow measurement method according to a second conventional technique.

FIG. 22 is a graph showing a relationship among g, h, and c.

FIG. 23 is a block diagram of a fifth related art.

FIG. 24 is a timing chart of FIG.

FIG. 25 is a diagram illustrating the input impedance of the amplifier.

FIG. 26 is an electric circuit diagram illustrating the relationship between the impedance of water and the input signal of the amplifier.

FIG. 27 is a diagram showing changes in an electrode potential and a ground electrode potential.

FIG. 28 is a diagram illustrating potentials of electrodes.

FIG. 29 is a diagram illustrating generation of an error.

FIG. 30 is a diagram illustrating non-excitation compensation.

FIG. 31 is a graph of the correction table A of [Table 1].

FIG. 32 is a graph of the correction table B of [Table 2].

FIG. 33 is a flow rate calculation algorithm.

FIG. 34 is a diagram showing the relationship between the flow rate and E2 when the conductivity is C from the correction table A.

FIG. 35 is a diagram for obtaining a flow rate Q from a measured value of E2.

36A and 36B are diagrams for obtaining the relationship between the conductivity and E3 / E1 when the flow rate is Q. FIG. 36A shows the relationship between the flow rate and E3 / E1, and FIG. 36B shows the relationship between the conductivity and E3 / E1. Show.

FIG. 37 is a diagram for obtaining conductivity C ′ from measured E3 / E1.

FIG. 38 is a diagram showing an example of the accuracy of the fifth related art.

Claims (2)

[Claims] 1. An upper excitation coil disposed above a flow path,
In a non-filled electromagnetic flowmeter having a lower excitation coil disposed on the lower side and a pair of electrodes disposed opposite to the side surface of the flow path and having a conductivity correction function and a non-excitation compensation function, the flow rate may suddenly increase, Capturing the fluctuation of the electrode signal when a solid rubs the electrode, temporarily stopping the conductivity correction and de-energization compensation, and calculating the flow rate using the conductivity just before the fluctuation of the electrode signal is captured And non-full water electromagnetic flowmeter. 2. A weir-type non-filled electromagnetic flowmeter having a weir in a flow path and a pair of electrodes disposed on an inner surface of the weir, wherein the conductivity correction and the non-excitation compensation are temporarily stopped. The non-filled electromagnetic flow meter according to claim 1, wherein the excitation mode is set to the excitation only in the upper direction.