JPH06103288B2 - Automatic moisture measurement method - Google Patents

Description

TECHNICAL FIELD The present invention relates to a method for automatically measuring the water content in a grain or granular material, such as wheat or wheat flour. In the following description, the tempering of wheat grains will be described as a representative example, but the present invention can be applied to many other grains such as grains and powders.

<Prior art and its problems> As a pre-process for improving the efficiency of crushing and separating wheat, there is a tempering (water addition or humidification) process.

In this step, it is important to uniformly homogenize the moisture content of wheat before crushing. If this is not successful, the yield of the product may decrease, the frequency of various operations may increase, and the number of failures may increase. In the past, since there was no accurate and reliable on-line measuring device, the actual process was adjusted after manually sampling from the process, measuring by the absolute dry method, determining the water addition rate.

However, this method has the following drawbacks.

(1) Water addition cannot be controlled in real time.

In the absolutely dry method, moisture measurement takes at least one and a half hours.

(2) Although the absolute drying method itself is used as a standard for measuring water content, there are many factors that cause errors such as the measurer, measuring device (constant temperature chamber), and sample particle size.

(3) Manpower is required for measurement, which is against labor saving.

(4) Since the number of samplings is small, it is difficult to grasp the true moisture value.

The absolute drying method is that wheat is roughly crushed, its weight W ₀ is measured, and water is evaporated in a constant temperature bath at 130 ° C. for 1 hour.
It is a method of cooling for 0 minutes, measuring the weight W ₁ again, and determining the water content by the following formula.

<Purpose of the invention> The object of the present invention is to eliminate the above-mentioned drawbacks of the conventional method, and in a short time, with high accuracy and without changing the shape of the sample (for example, it is necessary to grind in the absolute drying method), and the grain powder. It is possible to measure the water content of the body, which enables on-line and real-time automatic water addition control, and to provide an automatic water content measurement method that can save the manpower required for water content measurement, water content determination, and operation. .

<Brief Description of the Invention> In the present invention, the grain powder sample is sampled in a predetermined volume, the sample temperature is measured while the sample temperature is measured, and the sample sample is retained in the detection unit, and there is no sample in the detection unit. The change in the permittivity when the sample is in the detector is measured as the frequency change by an oscillator that uses an IC that is held at a constant temperature, and the frequency difference is calculated. Correction and correction by the change of the sample weight using the correction coefficient are added to the frequency difference to obtain the moisture value, and the correction coefficient of the sample temperature is corrected using this moisture value, and the corrected correction coefficient is used. again repeating the determination of the water content of the sample Te, previous water value (Moisn _-1), the difference between the moisture content water content value using the calculated correction coefficient from (Moisn) is equal to or less than a predetermined value Though no convergence in the moisture content (Mois
The present invention provides an automatic water content measuring method characterized in that n) is determined as a water content value of a sample.

<Detailed Description of the Invention> Hereinafter, the automatic moisture measuring method of the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 diagrammatically shows a moisture measuring unit of a powder or granular material such as wheat or wheat flour (hereinafter referred to as a sample). The sample is branched from the normal transportation line 1 by the branching unit 2, and the sample not used for moisture measurement is returned to the transportation line 1 via the bypass line 3. A predetermined amount of the sample to be used for water content measurement is stored in the constant volume unit 4, and a constant amount is constantly supplied to the weighing unit 5 with a weighing device like a load cell. An amount required for moisture measurement sampling is taken in the constant volume portion, and the rest is returned to the transportation line 1 via the overflow line 6.

In the present invention, in order to provide the constant volume portion, the sample weight is measured and corrected. However, when the sample weight is significantly different from the set amount, the weight correction coefficient varies (necessity of correction) and causes an error. It is because

The sample put into the weighing unit 5 is dropped to the next detection unit 7 and stored for a short time. As will be described later, the detection unit 7 captures a change in the dielectric constant between the electrode plates of the condenser on which the sample falls. Therefore, the detection unit 7 detects the sample so that the sample with a constant density is always measured. It is necessary to drop it at a constant speed to make it have a constant density and to store it. For this reason, an air rotating cylinder or the like is used for the measuring unit so that the sample is dropped instantly.

Conventionally, the sample temperature was measured by the detection unit, but there is a place where it is unclear whether the measured temperature is the sample temperature or the ambient temperature. In the present invention, the sample temperature is always measured in the constant volume part 4 in which the sample exists. Therefore, the thermometer is always placed in the sample and the sample temperature can be reliably measured.

The water content measured by the detecting unit 7 is intended to control the sample to an appropriate humidity by catching the change in the dielectric constant of the detecting unit 7 as a change in frequency, and the measuring circuit as illustrated in FIG. Is used.

In FIG. 2, the sample signal detected by the detecting unit 7 is oscillated by the oscillating unit 10 having the oscillator 9 via the line resistance stray capacitance unit 8, and the oscillated output is measured by the frequency counter 11. An IC (integrated circuit) is used for the oscillator 10.
Further, in the present invention, in order to keep the temperature of the IC used for the oscillator 9 constant, the oscillator 9 is arranged in the constant temperature bath 12 as shown in FIG. The circuit used for this oscillation is not limited to that shown in FIG. 2 and any suitable circuit can be used.

The water content measurement flowchart of the automatic water content measurement unit shown in FIG. 1 can be summarized as follows.

In the moisture measurement method as described above, in the moisture measurement, it is as described above that the change in the dielectric constant in the detecting unit 7 is regarded as the change in the frequency in the oscillating unit 10. The dielectric constant in the detecting unit 7 depends on the presence or absence of the sample. As a matter of course, the density, weight, and temperature of the sample in the detection unit 7 of the sample are affected, and also the temperature of the IC used in the oscillation unit 10, that is, the ambient temperature is affected. No matter how high-quality ICs are used, there are still significant effects.

By the way, the influence of the change in the weight of the sample, the change in the temperature of the sample, and the change in the ambient temperature on the frequency change is shown in FIG.
This is shown in FIGS. 4 and 5. As can be seen from FIGS. 3 and 4, changes in the weight and temperature of the sample cause large variations in frequency.

Further, FIG. 6 shows the frequency variation when the ambient temperature changes by ± 3 ° C. As can be seen from FIGS. 5 and 6,
The ambient temperature of the oscillator IC is about ± 2KH with a change of ± 3 ℃.
It can be seen that a change of z, ± 1.0 ° C has a frequency variation of about 0.5 KHz and has a great influence on the frequency variation of FIGS. 3 and 4.

The frequency difference Δf is calculated by the following equation.

Δf = femp-fin femp: frequency when there is no sample in the detection unit fin: frequency when the sample is introduced into the detection unit Therefore, in the present invention, the temperature and the sample weight of the sample are selected as factors affecting the frequency change. Then, the water content is automatically measured by performing correction using the correction coefficient for these.

The ambient temperature of the IC of the oscillator (hereinafter, simply referred to as ambient temperature) has a great influence on the measurement of the moisture value, as can be seen from the points shown in FIGS. 5 and 6.
In the present invention, automatic control is performed within a constant temperature range that does not affect the temperature.

The method of automatically measuring the water content in the sample by the frequency correction accompanying the change in the above factors is shown below. The following example uses raw wheat as a sample, but the coefficients and mathematical formulas shown below are the types of grains such as soybean, rice (including glutinous rice) corn, coffee beans, and their varieties. Is determined by Note that these various coefficients and mathematical formulas may be experimentally determined by the method described later. In the measuring method shown below, it is preferable to perform the calculation using a computer.

First, in the detection unit 7 shown in FIG. 1, the frequency femp when the sample does not exist is measured, the frequency fin when the sample is introduced into the detection unit 7 is measured, and the difference Δf is obtained by the equation (1). (See process).

Next, the weight correction shown in the process is performed by the sample weight W measured by the measuring unit 5. In this step, the weight correction value Δf ₄₀₀ is calculated by the equation (2). In this example, the barley is sampled with the reference weight of the sample being 400 g, and 4.80 is used as Cw (weight coefficient) in this example which is barley. This Cw (weight coefficient) is appropriately determined according to the type of product by the method described later.

As will be described in detail later, in the present invention in which constant volume sampling is performed in which the sample weight W does not fluctuate significantly, this Cw (weight coefficient) can be made constant without the need for correction. Steps before the step can be omitted (that is, only the sample temperature coefficient needs to be modified).

Next, the sample temperature correction shown in the process is performed by the sample temperature T measured by the constant volume unit 4. In this step, the sample temperature correction value Δft ₁ for the first time is calculated by the equation (3). In this example, the reference temperature of the sample is 20 ° C. Ctn is a sample temperature coefficient, and the first sample temperature coefficient Ct ₁ is a hypothetical value of 8.0 in raw wheat.
8 is used, and in the second and subsequent times, the value calculated by the equation (6) in the step described later is used. Note that the first-time virtual value is appropriately determined according to the type and the like by the method described later.

Next, using the specimen temperature correction value Δft ₁ , the first moisture value Mois ₁ is obtained as shown in equation (4) of the process. The equation (4) for calculating the water content (Moisn) is appropriately determined according to the type and variety of grain. The method for determining this will be described later.

After calculating the moisture value Mois ₁ for the first time, the calculation of the (new) sample temperature coefficient using the moisture value Mois ₁ by the equation (6) shown in the process without performing the process convergence check, that is, the second time Calculate the sample temperature coefficient Ct ₂ used to calculate the water content. The formula for calculating the new sample temperature coefficient is appropriately determined according to the type and type of grain by the method described later.

Using the specimen temperature coefficient Ct ₂ obtained in this way,
The process is calculated again to obtain the sample temperature correction value Δft ₂ , and the process is calculated to obtain the _second moisture value Mois ₂ .

Next, a convergence check of the process is performed between this moisture value Mois ₂ and the previous moisture value Mois ₁ using the equation (5). When this difference is less than a certain value (0.02% in this example), it is considered to have converged, and the water value Mois ₂ at this time is taken as the water value of the sample.

On the other hand, when the difference in convergence check does not become less than the predetermined value, using a water content Mois _2, calculate the sample temperature coefficient Ct ₃ performs step again by using the sample temperature coefficient Ct ₃ 3 The moisture value Mois ₃ for the second time is calculated, and a convergence check is similarly performed with the moisture value Mois ₂ for the second time.

By repeating the above operation, the difference between the previous moisture value Moisn _-1 and the current moisture value Moisn is less than the predetermined value (0.02 in this example).
%) And the moisture value Mo at this time is considered to have converged.
Let isn be the water content of the sample.

It should be noted that the previous moisture value Moisn _-1 which can be regarded as convergence
The difference between the moisture content Moisn and the moisture content Moisn of this time varies depending on the variety of the sample and the like, but may be less than 0.02% in the above-described raw barley.

In this way, the water content of a certain sample at a certain time is obtained. When it is necessary to obtain the water content over time, the water content is measured at predetermined time intervals. Use this measurement value to adjust the water supply to samples such as barley,
The water content of the sample can be controlled to fall within a predetermined range.

Equation (4) for calculating the water content (Moisn) uses a highly reliable method according to the type and variety of grain (such as the absolute drying method for wheat) to measure the water content of multiple samples. Then, the absolute moisture value and the frequency difference Δf in each sample are obtained,
From the regression curve of both (generally a quadratic curve), the relationship between the two may be used as a regression equation and this may be used.

FIG. 12 shows the relationship between the absolute dry value and Δf when raw wheat was used as a sample.

The example shown in FIG. 12 is a system as shown in FIG.
Using the measurement circuit of FIG. 14 and FIG. 14 to be described later, Δf obtained by sequentially sampling the sample with the sample reference weight of 400 g, the sample reference temperature of 20 ° C., and the reference atmosphere temperature of 20 ° C.
It shows the relationship with the moisture value obtained by the absolute drying method for the same sampling sample. The variation σ n _-1 between the regression curve and each plot when the water content by the absolute dry method is the true value is
It is 0.045, which shows a sufficient correlation.

The equation of this quadratic curve is the equation (4) for calculating the moisture value (Moisn), Moisn = ± A ± B · Δftn ± C · (Δftn) ^2, which is shown as a straight line in FIG. However, in the same example using the raw wheat as a sample, Moisn = −49.60 + 0.033 · ftn− (4.16 × 10 ⁻⁶ ) · (Δ
ftn) ² is exemplified.

The above-mentioned measurement was repeated for the same sample. The results are shown in Table 1. The average value of Δf is 3468,
σ n _-1 = 14.2, which corresponds to a water content of 0.06%, and it can be confirmed that the water content can be measured with good reproducibility.

In the above automatic moisture measuring method, the weight coefficient Cw and the sample temperature coefficient Ctn are all functions of the moisture value Moisn.

The weight coefficient Cw is used to correct the measured frequency difference Δf to the frequency difference Δf ₄₀₀ of the reference weight based on the excess or deficiency of the sample weight with respect to the reference weight (400 g in this example). As shown in FIG. 3 described above, the relationship between the sample weight and the frequency difference Δf can be expressed by a linear expression. Therefore, the weight correction can be performed by using this linear equation, and the weight correction can be performed by the above equation (2) with this inclination as the weight coefficient Cw. In the raw barley shown in FIG. 3, the weight coefficient Cw is 4.80 as described above. Further, regarding the weight coefficient Cw, a quadratic regression is obtained when the weight range and the range of the moisture value Mois are widened (see FIG. 7). However, if the moisture range is limited to some extent and the weight sampling range is kept within about ± 20 g by the constant volume sampling method, a substantially constant value should be used (see FIG. 8). Therefore, in the measuring method of the present invention, the weight coefficient Cw used for the weight correction of the process for calculating the water content does not need to be corrected (a constant value can be used), and therefore, as described above, The sample temperature coefficient Ctn is to be corrected when calculating the water content.
Only need be.

The sample temperature coefficient Ctn is used to correct the weight-corrected frequency difference Δf ₄₀₀ to the frequency of the sample at the reference temperature based on the difference between the reference temperature (20 ° C. in this example) and the sample temperature. As shown in FIG. 4 described above, the relationship between the sample temperature and the frequency difference Δf can be expressed by a linear expression, and basically, similar to the weight coefficient Cw described above, this slope is used as the sample temperature coefficient Ctn, The sample temperature can be corrected by the equation (3).

However, as shown in FIG. 4, the sample temperature coefficient Ct
The gradient of n varies not only with different water addition rates but also with the same water addition rate. Therefore, in the present invention, in order to perform accurate measurement, the sample temperature coefficient Ctn is corrected and repeated calculation is performed until the moisture value (Moisn) converges.

Here, although the water addition rate is used as a parameter in FIG. 4, in the same measurement, FIG. 9 shows the sample temperature coefficient Ctn (slope) with the water value corresponding to Δf as the horizontal axis. As shown in FIG. 9, when the moisture range is wide, the sample temperature coefficient Ctn becomes a quadratic regression. However, if the moisture range is limited to some extent, linear regression will be performed as shown in FIG. The new sample temperature coefficient Ctn ₊₁ may be calculated from the previously calculated water content (Moisn) using this linear equation. In the above example, Ctn ₊₁ = -1.38 + 0.89 · Moisn is used. Good.

As the initial value (Ct ₁ ) of the sample temperature coefficient Ctn, for example, from the experimental values of FIG. 4 and the like, the average value of the gradient may be adopted according to the corresponding water addition rate. In the above example, 8.08 is a virtual value. Used as.

Further, regarding the ambient temperature correction, when the ambient temperature coefficient is calculated so as to correspond to the change in the ambient temperature of the IC, it becomes a quadratic regression (see Fig. 11). However, in the present invention, since the ambient temperature of the oscillator using the IC is controlled to be constant, it is not necessary to make this correction.

The above example is an example of measuring the water content of the raw barley, but the calculation formula can be obtained by the same method even for various kinds of grains other than this. The following is an example when the sample is soybean.

Weight correction Δf ₃₂₀ = Δf-8.08 (W-320) Reference weight is 320 g Sample temperature correction Δftn = Δf ₃₂₀ −42.6 (T-20) Reference temperature is 20 ° C Moisture value Moisn = 40.839 + 0.025 · Δftn − (5.51 × 10 ^-6 ) ・ (Δftn) ² New temperature correction coefficient Ctn ₊₁ = -18.323 +5.08 Moisn In the present invention, the oscillator 9 using an IC is put in the constant temperature chamber 12 and the temperature of the constant temperature chamber 12 is measured. A temperature sensor 13 and a heater 14 for heating the constant temperature bath 12 are provided in the constant temperature bath 12, and these are connected to a temperature control circuit 15 as shown in FIG.
The atmosphere temperature in 12 is controlled to be substantially constant so that the oscillator 9 using an IC is not affected by the atmosphere temperature.

FIG. 14 shows an example of the temperature control circuit. In the figure,
16 is a temperature signal amplifier, 17 is an upper limit temperature T ₁ comparator, 18 is a lower limit temperature T ₂ comparator, 19 is a temperature-heater current converter, 20 is a first relay, 21 is a second relay, and 22 is a transistor. .

In this circuit, when the constant temperature bath temperature T becomes higher than the upper limit temperature T ₁ , the first relay 20 is turned on and the heater current becomes zero. When the constant temperature bath temperature T becomes lower than the lower limit temperature T ₂ , the second relay 21 is turned on and the heater current becomes maximum. When the constant temperature bath temperature T is between the lower limit temperature T ₂ and the upper limit temperature T ₁ , the heater current has a value proportional to the temperature T. In this way, the constant temperature bath temperature T is controlled within a narrow temperature range between the lower limit temperature T ₂ and the upper limit temperature T ₁ . By the way, T ₁ −T ₂
Is preferably about 2 ° C.

<Examples> [Example 1] Comparison of automatic moisture measurement value online and absolute dry value Comparison of moisture value measured by the method of the present invention and moisture value determined by the absolute dry method for various raw barley samples Is shown in Table 2. From this Table 2, it can be seen that the moisture value obtained by the method of the present invention is very close to the absolute dry value, indicating that the method of the present invention is reliable.

The calculation of the water content of the method of the present invention in this example is
This was performed according to the flowchart on page 6 above. The weight coefficient Cw was 4.80, and the first sample temperature coefficient Ct ₁ was 8.08. Further, the moisture value Moisn in the process was calculated using the following formula (4), and the new sample temperature coefficient in the process was calculated using the following formula (6). Furthermore, in the process convergence check, the moisture value (Moisn _-1 ) and moisture value (Moisn)
If the difference between and is less than 0.02%, it is considered as convergence.

Moisn = -49.60 + 0.033 ・ Δfin− (4.16 × 10 ^-6 ) ・ (Δftn) ² (4) Ctn _{+ 1} = -1.38 + 0.89 ・ Moisn (6) <Effects of the Invention> In the absolutely dry method, which is a conventional method for measuring water content, water control cannot be performed in real time, and it takes at least one and a half hours to measure water content. Although there were many problems that the number of times was small and it was difficult to grasp the true water content value, the method of the present invention, in a short time, accurately and without changing the shape of the sample (the pulverization is required in the absolute drying method). ) It is possible to measure, and therefore, online and real-time automatic hydration control can be performed, and manpower required for water content determination, hydration rate determination, and operation can be saved.

[Brief description of drawings]

FIG. 1 is a flowchart of a method for measuring the water content of a sample according to the present invention. FIG. 2 is a part of the measurement circuit diagram used for the detection unit of FIG. FIGS. 3, 4, and 5 are graphs showing the influence of changes in sample weight, sample temperature, and ambient temperature on Δf and f ₀ , respectively, and FIG. 6 shows that the ambient temperature change is ± 3 ° C. is a graph showing changes in f ₀ when. FIG. 7 and FIG. 8 are graphs showing the relationship between the weight coefficient Cw and the water content. 9 and 10 are graphs showing the relationship between the sample temperature coefficient Ctn and the water content. FIG. 11 is a graph showing the relationship between the atmospheric temperature correction coefficient and moisture. FIG. 12 is a graph showing the relationship between the absolute dry value and Δf. FIG. 13 is a measurement circuit diagram used in the present invention. FIG. 14 is a temperature control circuit diagram for keeping the oscillator of the measurement circuit diagrams shown in FIGS. 2 and 13 at a constant temperature. Explanation of symbols 1 ... Transport line, 2 ... Branching part, 3 ... Bypass line, 4 ... Constant volume part, 5 ... Measuring part, 6 ... Overflow line, 7 ... Detecting part, 8 ... Line Resistance stray capacitance part, 9 ... Oscillator, 10 ... Oscillation part, 11 ... Frequency counter, 12 ... Temperature chamber, 13 ... Temperature sensor, 14 ... Heater, 15 ... Temperature control circuit, 16 ... Temperature Signal amplifier, 17 ... upper limit temperature T ₁ comparator, 18 ... lower limit temperature T ₂ comparator, 19 ... temperature-heater current converter, 20 ... first relay,
21 …… Second relay, 22 …… Transistor

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Kazunori Suda 1906, Kamikubo, Oi-cho, Iruma-gun, Saitama 139 No. 139 (72) Inventor Ryuichi Haginoya 3-4, Suehiro-cho, Kawagoe-shi, Saitama Prefecture (72) Invention Fumio Hara 947, Mojiba, Kawagoe City, Saitama Prefecture 3 (56) References JP-A-55-2939 (JP, A) JP-A-50-123498 (JP, A) JP-A-56-43545 (JP, A) )

Claims (1)

[Claims] 1. A sample of a grain or granular material sample is sampled in a predetermined volume, the sample temperature is measured and the sample temperature is measured, and the sample sample is retained in the detection unit. The change in the permittivity when there is is measured as a frequency change by an oscillator using an IC held at a constant temperature, and the frequency difference is calculated. The moisture value is obtained by adding the correction due to the change of the used sample weight to the frequency difference, and the moisture temperature value of the sample is corrected using the moisture value, and the moisture value of the sample is corrected again using the corrected sample temperature coefficient. Is repeated to converge when the difference between the previous moisture value (Moisn _-1 ) and the moisture value (Moisn) using the correction coefficient corrected using this moisture value is less than the specified value. Na Though, an automatic moisture measuring method characterized by determining the water content value (Moisn) as a moisture value of the sample.