RU1783395C - Method of measuring loose material moisture content - Google Patents

Abstract

Usage: in measurement technology, in metrological support systems for neutron moisture meters and moisture density numbers. SUMMARY OF THE INVENTION: In order to reduce the reaction time during a sudden change in the measured value according to a special algorithm, the averaging intervals are optimized during the measurement process. 2 il .. 2 tablets V Ј

Description

The invention relates to measuring equipment, in particular, to methods for measuring the properties and composition of bulk solids by radioisotope methods, and can be used in the metrological support system of measuring instruments, for example, neutron hygrometer and hygrometer.

A known method of measuring the moisture content of bulk solids, which consists in irradiating a substance with fast neutrons, measuring the flow of slowed-down neutrons I and determining moisture according to the formula

W al + b,

where a. b - calibration coefficients 1 determined experimentally.

Due to the random nature of the interaction of ionizing radiation with matter, the intensity I contains random errors D h, which affect the measurement error of humidity Wi. To reduce the error in measuring moisture, calculate the average value of Wn from

the results of n measurements Wi, L / 2 Wi

Wn by the formula

  rf1 Jwi.

I 1 where Wn is the average value of humidity:

p is the number of measurements;

Wi is the current humidity value.

The result can be obtained only after taking measurements, which determines

00 Сл)

SL

about with

and rigidly fixes the performance of the moisture meter or moisture meter

In addition, for processing the measurement results, it is necessary to have a large amount of memory (memory cells), which complicates the design and reduces the noise immunity of the device for processing moisture meter information,

A known method of measuring the moisture content of bulk solids with neutron hygrometer, which consists in irradiating the substance with fast neutrons, measuring the intensities of the fluxes of thermal H and superheat g neutrons and determining the current humidity value Wi by

Wr

111 -a0) 1E2 - (121 - b0) -E2

; i2i -b0) -a1 - (Hi -a0) -b2

(1)

where Wi is the current humidity value;

Hi, (21 current intensities;

a0-a2, b0-b2 are calibration coefficients.

When averaging the humidity values in the known method for measuring humidity, an algorithm is used according to which the average humidity value is determined in accordance with the formula

Wni () Wni-i +,

(2)

where Wni -1, Wn | are successive average humidity values;

Wi - current humidity value;

u is the number of measurements.

Moreover, after each cycle of measuring the current humidity, the value of rii increases by one, but does not exceed the value

Pmax

 PI + AND Ymax.

The value of Pmax is chosen experimentally and depends both on the permissible value of the random component of the error in measuring humidity and on the dispersion of neutron fluxes recorded by the detectors

With increasing max, the random component error in the measurement of humidity decreases.

However, the measurement method has a long reaction time with a sudden change in the moisture content of the material, since when determining the average value according to formula (2), the current value of Wi has a weight coefficient - and the average value

Wn (- weight coefficient

n -1

For large values of n, the abrupt change in Wi has little effect on the average value, and only after several measurement cycles (pmax) does the average value of Wni approach the actual value of humidity.

This disadvantage leads to the fact that the accuracy of measuring moisture decreases with a sudden change in the measured value.

The aim of the invention is to increase the accuracy of measurements by eliminating the influence of random fluctuations in the measured value and optimizing the averaging interval.

This goal is achieved in that in a method for measuring the moisture content of bulk materials, including irradiating them with a fast neutron flux, measuring the intensities of thermal and epithermal neutron fluxes, from the values of which, in each measurement cycle i, the current humidity value WI is found and determined the average value of Wni, for example, by the formula

-

Wn Km- OWni-,,

where Wni - i, Wni are successive values of the average moisture content of the material;

u is the number of measurements

moreover, in each subsequent measurement cycle, nl increases by 1, but not more than

DO Pmax:

P | Py Pmax,

where pmaxЈ2 is the coefficient determined experimentally,

For each measurement cycle, the maximum permissible value is determined

the AWinaicc error values according to the previously obtained experimentally dependence of WiMaKc on the average humidity value Wni i then determine the difference between the current and average humidity values D Wi and check the conditions

IAW | l AW | Mai c, AW | -l DMmmax

(3)

If the condition is not satisfied, the measurement and averaging of humidity are continued; when the condition is met, the condition is checked

55

| AW | + DM | max,

(4

if it is not fulfilled, the cycles of measuring and averaging humidity are continued, and when executed, an abrupt change in the measured value is recorded and a new averaging cycle is started, in which the current value Wi is taken as L / n, and the value of r is taken to be 1.

The claimed method is characterized in that for each measurement cycle the value of the maximum permissible error AWiMaKc is determined. the difference between the current and average humidity values is determined and conditions (3), (4) are checked sequentially, while condition (4) is fulfilled, the presence of an abrupt change in the measured value is recorded and a new averaging cycle is started.

The inventive method of measuring humidity allows to increase the accuracy of measurement in case of a sudden change in the measured value by optimizing the reaction time by taking measurements in the case of a slow change in the measured value with a large time constant (reaction time is g pmax. Where t is the measurement cycle time ), revealing the moments of stepwise measurement of the measured value and a sharp decrease in this constant time (the reaction time is equal to the cycle time) and, with the next stable process, icheni time constant again to a maximum value

Figure 1.2 shows the dependences of the readings of the hygrometer Wi on the number of measurement cycles i with an abrupt change in humidity from 0.8 to 12.1% (Fig. 1) and from 8.4 to 7.1 (Fig. 2).

On i plots, position 1 shows the current values of humidity Wi measured by a moisture meter without averaging information; position 2 is the average humidity values Wni determined by the known measurement method: position 3 is the average humidity values of the claimed measurement method; Fig. 1.2 ti is the reaction time when performing the claimed measurement method; T.2 - a known measurement method.

The inventive method for measuring humidity is implemented as follows.

Bulk material, e.g. blast furnace coke, in a weight funnel is irradiated with a fast neutron flux. The intensities of thermal h and epithermal l2i neutrons are measured over time t. corresponding to one measurement cycle i. The current humidity value Wi is determined by the formula (1).

The average humidity value is determined, for example, by the formula (2) Moreover, after each measurement cycle, the value ni

increases by one, but not more than

DO d) max

ni ni-i + 1 pMax.

After each measurement cycle, the difference between the average Wni and the current Wi moisture values is determined by the formulas

10

AWi Wm -Wi. . AWi-i Wni-i -WM.

(5)

At the same time, taking advantage of the dependence obtained earlier experimentally, they determine for Wni the value of the maximum value of the permissible error of the device in this cycle

WiMaKC

Verify that condition (3) is met. If this condition is not fulfilled, then the averaging of humidity is continued according to formula (2) If condition (3) is satisfied, i.e. If two consecutive current humidity values Wi and Mi exceed the pre-set value, 5 times A WiMaKc, then condition (4) is satisfied.

If condition (4) is not satisfied, then the averaging of the humidity value by formula (2) is continued. Under the condition (4)

Q, the value Wi is taken as Wni; further, the averaging of the humidity value by formulas (2) is continued with ni 1.

Thus, the invention lies in the fact that the calculation of the average

5, the values of Wn are produced according to the formula (2), the value being determined depending on the difference of two consecutive current values of Wi and Wi-i and the average value of Wn | , Wn | - 1 .

0 If each of two consecutive values of the difference A M and AWi-i exceeds the specified limit DMyaks, and the sum of these two differences, taking into account the sign AW IA Wi + AWi-i | exceeds given

5 is the limit k AW | MaKc, then this most likely means that an abrupt change in the measured quantity occurred. In this case, the value of n is assigned a unit value (ni 1) and averaging continues from the initial condition (2)

Since for a random process, the probability that two consecutive values of AW and AWi-1 go beyond the given limit with the same sign is small, then the analysis of two consecutive error values Ai, A1-1 (taking into account their sign) allows a large the probability to determine the jump in the measured quantity and at the same

time to not respond to a single exit of an error beyond the set limits.

The value of the coefficient k is chosen experimentally so that a change in the value of n occurs with a change in humidity and does not occur with random fluctuations in humidity due to statistics.

Since the dispersion of the readings of the hygrometer due to statistics is in the range from + D L / | max to -, k must not be less than 2, since D W 2 Д D L / Max.

Since the process of interaction of neutrons with matter is random, the random component of the error in measuring moisture increases with increasing humidity.

Therefore, the value of A L / s should be a variable value depending on the average value of Win.

For example, for a moisture densitometer experimentally found dependence

FOR D L / | max.

Awi

Max

A + B-Wni,

(6)

where A, b are empirical coefficients, which are found experimentally as follows.

At various humidity values in the range from Umin to L / max, the error D L / | Max is determined. The dependence is built by formula (6) and empirical coefficients are determined by the formulas

A L / max

AV4, aKc2 -LMmax

AT

Wn2-wm

where Д L / Max, D L / Max2 is the maximum value of the error at the minimum value of the average humidity L / n and the maximum value of the average humidity L / P2 It should be noted that the success of the application of this method depends on the degree of reliability of revealing the moments of a jump-like change in the average values of the measured value, and precisely with the aim of increasing this reliability and preventing false alarms, a double check is used, which is reflected in condition (3). Assuming, for example, that the random variable Wi obeys the normal distribution law, and sets, as is customary in metrology (see, for example, GOST 11.001-73). the value of the confidence interval equal to 2 about, get the probability of random occurrence of the shutdown AWi in excess of 1 a,

5

0

5

0

5

0

5

45 55

equal to 0.05, and the probability of coincidence of such events in two consecutive cycles will be 0.05 g 0.0025, i.e. less than 3 cases out of thousands, and can be further reduced if necessary at the next stage of verification (4) by selecting the coefficient The presence of two inequalities in condition (3) instead of one stronger one prevents false alarms of the system under the influence of random emissions and pulsed interference .

As for verification by inequality (4), it is necessary to prevent the system from tripping when WI oscillates in different directions around a constant or slowly varying average value of Wn |.

Example. The implementation of the method is carried out using a moisture meter of a neutron stationary VPN-7208, containing a neutron sensor and a specialized computing device (processing device). The measurement is carried out in a weight funnel into which blast furnace coke is loaded with different humidity (0.8% and then 12%). Using the sensor, coke is irradiated with fast neutrons and the intensities of the fluxes of thermal and epithermal neutrons H and 12 are measured. Humidity Wi is calculated in the processing device by the formula (1). The average moisture value is calculated by the formula (2). The difference between AWi and AWi-i is determined by formulas (5). They find the value and determine whether both differences go beyond this limit. If two successive values of the difference go beyond, for example, 0.6%, then the total value of the two differences is found by the formula (4) and determines whether the sum of the two differences 1AW1 beyond the limit 1oD L / | max-1.2%, k-2.

If it is outside the limit of 1.2%, then a value of n is assigned a unit value (r1). Next, the moisture measurement process is repeated. The pmax value was chosen equal to 15. The measurement results are shown in Table 1.

Similarly, the measurement of humidity by a known method.

The measurement results are shown in Table 2.

The results are also shown in Fig. 1, which shows the dependences of the current humidity value Wi (1), the average humidity value Wni according to the known measurement method (2) and the average humidity value Wni according to the claimed method (3) on the number of measurement cycles.

As can be seen from the above results (Table 1.2 and Figure 1.2), the claimed method for measuring humidity has a shorter reaction time with a stepwise change in the measured value in comparison with the known methods.

Figure 2 shows the dependence of the readings of the radioisotope moisture meter with a change in humidity from 8.4 to 7.1%.

The measurement is carried out on a conveyor belt.

Thus, the claimed method in comparison with the known method allows to reduce the reaction time upon a sudden change in the measured quantity, thereby increasing the accuracy of measuring systems, measured physical quantities in a dynamic process, for example, humidity, density of bulk solids in a hopper or conveyor belt.

Claims (1)

SUMMARY OF THE INVENTION A method for measuring the moisture content of bulk materials, including irradiating them with a fast neutron flux, measuring the intensities of thermal and suprathermal neutron fluxes, by the values of which in each I measurement cycle the current humidity value Wi is found and the average humidity value L is limited / w according to the formula JWnr (ni-JJWni - 1 + W, / m, where Wni - 1, Wm are consecutive nor the average humidity of the material; nl is the number of measurements and in each subsequent measurement cycle P | P | -1 + KPmzks. where pMax 2 is a coefficient determined experimentally, characterized in that, in order to increase the accuracy of the measurement by eliminating the influence of random fluctuations in the measured quantity and optimizing the averaging interval, the value of the maximum permissible error A L / is determined for each measurement cycle | Max according to previously obtained experimentally the dependence of D L / gmax on the average humidity Wni, determine the difference D Wi between the current and average humidity values taking into account the difference D Wi-i and check so that the conditions - | AW | l DMax,  D L / | -1max, when conditions are met, a condition is checked iAW | + kD L / | Max, , where is the coefficient established experimentally, when the last condition is fulfilled, a change in the measured quantity is abruptly recorded and a new averaging interval begins, in which the current value Wi7a is taken as Wni and taken equal to 1. Table 1 The results of the measurement of moisture according to the claimed method Table 2 The results of measuring the inference by a known method I §   j    l., .1 i11a. " tpue.f measured i. tt i ... I