SU1682775A1 - Method of calibration of radio isotope devices for process monitoring - Google Patents

Abstract

The invention relates to a measurement technique, in particular, to methods for calibrating radioisotope instruments of technological control. The purpose of the invention is to improve the accuracy of the gauging of the thickness gauge by using three sets of certified measures of surface density: one full-scale and two basic sets, the materials of which are selected according to effective atomic numbers, and the check calibration characteristic is constructed taking into account the sections with the minimum error of the calibration dependencies of the two basic sets . Using the measures of the measures of the first basic set, the calibration characteristic GX1 of the output signals for the measures of this set is used, using the parametric type function, the parameters of which are determined by the measured values of the output signals from these measures, similarly to the construction of the GH2, using the measures of the measures of the second basic set. After that, a verification calibration characteristic of the GX2 transition from GX1 to GX2 is constructed, the accuracy of the transition is determined, and a combination of parameters of the GX1 function is chosen, which can be changed to achieve the desired accuracy of the calibration characteristic of the thickness gauge. Having established such combinations, the calibration characteristic of the thickness gauge is constructed as the dependences of the output signals for the existing field measures using GX1 with the numerical values of its parameters, besides the detected combination, the numerical values of the parameters of which are determined by the values of the output signals for the field measures. 1 il.

Description

The invention relates to a measurement technique, in particular, to methods for calibrating radioisotope instruments of technological control.

The purpose of the invention is to improve the accuracy of the calibration of the thickness gauge by using three sets of certified measures of surface density: one full-scale and two basic sets, the materials of which are selected according to effective atomic numbers, and the verification calibration characteristic is constructed taking into account the areas with the minimum error of the calibration dependencies of the two basic sets.

The drawing shows the calibration characteristics (GC) illustrating the method:

GH1 built on a complete set of basic samples of surface density (OBPP);

GH2 built on a complete set of basic samples of surface density (OBPP2);

The GH2 is built from the GH1 according to the complete set of basic samples of surface density (OBPP2);

GHN is constructed on the full set of full-scale surface density measures (NMP);

About 00

Yu

X |

XJ ate

GH h built of GH1 with a limited number of natural measures NMP.

The method is illustrated by an example of the specific construction of a calibration characteristic (GC) of a radioisotope absorption thickness gauge with an ionizing radiation source (SRI) based on the Cs isotope for measuring the natural material — steel (atomic number ZH 26) in the range of surface densities (PP) from 0.27 Y5 up to 4,4-Y5 g / m2 in the presence of four natural measures of surface density (NMP) with a limit of additional GC error due to the limited number of natural measures 5i ± 1%.

The method is carried out as follows.

Choose certified sets of baseline density samples OBPP1 and O BPP2 according to the formula IZH - Zi II Zi - Za I, respectively, from copper (Zi 29) and aluminum (Z2 13) in an amount that provides GC error of basic materials less than 5js, for example, 1/3 4j.

Radiation from the scientific research institute is sequentially sent to each sample and the value of the output signal of the radioisotope transducer (RIP) is recorded.

The table shows the normalized values of the measured output of the Umorm for the Cu, AI sample sets, four steel samples, and the data for the complete set of steel samples used to assess the accuracy of the proposed GC construction method. The normalization is done so that when the scientific research institute is open and the sample is not in the zone. Measurement (P 0) Unorm 1, and when the shutter is closed, III Norms 0.

The relative error of the base and full-scale PP samples is 0.1%, the measurement error of the RIP output signal

0.01% with consideration of the physical sensitivity coefficient of the Kfch 5 method,

where 3A is the relative change in the measured signal, P is the relative change in the measured RI, the measurement error of the signal transferred to the RR is

± 0.05%. Therefore, the error in the basic GC is 0.15%.

Next, using the data in the table for OBP1 and OBP2, the GH1 and GH2 are constructed using the function of the parametric type

fa (n) xit

 + Х3ГХ П +, (1)

where Xj are the parameters determined from the experimental data from the table for the corresponding samples (1 1.26). For

finding parameters X | use the method of variation-weighted quadratic approximations (VVCP). Then GH1 for copper is

facu (n) 0.85483-Ha73906 10 n +

+ 0.09340

, -0.46288 10 P

10+ 0.051777. j-1.83073 u-n

GH2 for aluminum has the form

(2)

TAL (P) 0.87384 -1-0.74694-10 P +

+ 0.06326 G0-39682 +

+ 0.062290. GM3047 10HF (3)

Then, from the basic GC1 TaCi (P) function, using the data in the tables for OBP2, build the GH2, TapA1 (P) so that its relative deviation from the GX2 faAI (n) calculated by the formula (4) does not exceed the specified the value of the limit of additional error (5, ± 1% ... PP-P., PLO, "

five

0

five

0

Fan (y) - inverse ky fan (P) function.

To this end, the values of the parameters of the GX1 function are sequentially changed, starting with one and gradually increasing the number of variable parameters. With each change of the parameter, the value of 5 is checked using the formula (4). The variation of the parameters is performed until a combination of parameters is found, at which the value of & will be less than the specified limit of additional error ± 1%. In this case, the parameters Xi and Xs of expression (1) are most sensitive to changes in the material. The value of these parameters is determined by the VVPP method according to the table for OBP 2.

Found in this way GH2 d | has the appearance

AI /

fan (P) 0.87882 + 0.06934

-0.46288 10 P

| -0,73906-Ю П +, -fr

,-one. 83073-10TI

(five)

+ 0,05177

Then, using the data given in the table for the NMP, the basic function of the GH1si is similarly used to construct the GC St for the natural material steel. Wherein

using the VVPP method, the numerical values of only those parameters (Xi and Xs) of the GHCHS function, by changing which the construction of GC2 was achieved d | with a specified error value d ± 1% for aluminum, more different in Z from the first base material than natural material, which corresponds to the ratio | 2st - Zcul IZcu - ZAI. The GC st found in this way, which is the desired calibration characteristic of a thickness gauge for measuring a given material, has

view

fanCT (n) 0.87194-1

-0,73906-10 P

+ 0.07629

-0.46288-10 5P

+ 0.0517777-G1 8307310 p. (6)

In order to estimate the GHT of the GHT with respect to the GHst, we construct the GHst using the function (1) and the data given in the table for the complete set of NMPP steel

Taste (L) 0.86573-G0 75239 10 +

+ 0.09233

| -0.45884 -10 P + -1,89041-10 P

+ 0.04194- | -. (7)

Thus, the task constructed by GH of a full-scale material with a given accuracy is reduced to determining a smaller number of unknown parameters.

In this example, it is enough to have even one full-scale measure of appropriate accuracy to determine the coefficients Xi and Xs, taking into account their linear relationship from the normalization condition Xs 1 - Xi -0.05177.

Using a greater number of natural measures, it is possible to improve the accuracy of determining the specified parameters, and therefore the GC art. Therefore, the method makes it possible to achieve a given additional error of graduation according to natural measures with their limited number and provides an increase in the accuracy of the graduation of the thickness gauge. In addition, it provides for the creation of metrological assurance for radioisotope thickness gauges, intended for measuring materials, the manufacture of field measures of which is difficult.

Claims (1)

The method of calibration of a radioisotope thickness gauge, which consists in choosing certified full-scale surface density measures (NMP) of a material with an effective atomic number ZH, for which measurement the thickness gauge is intended, with successively changing density (PP) in the measurement range, consistently directed to the NMP radiation from the working ionizing radiation source (III) and for each of them 5 register the output signal value, select certified samples of the first base set (OBP1) with successively varying in the measuring range values of PP, the number of which 0 is determined from the condition of providing the specified error limit of the calibration characteristic (GC) & thickness gauge, consistently directed to the OBP 1 radiation from the working III, for each of 5, they record the value of the output signal and build the calibration characteristic GX1 of the dependence of the output signals for OBP1. Using a parametric-type function, the parameters of which are determined by the measured values of the output signals for OBP1, characterized in that, in order to improve the accuracy of the GC thickness, the certified samples of the second basic are additionally selected 5 sets (OBPP2) with sequentially varying values of PP in the measurement range, the number of which is determined from the conditions for ensuring a given limit of GC dT error of the thickness gauge, and the effective atomic numbers Zi and Z2 of the materials of the basic sets must correspond to the relation (ZH - Zi) (Zi - Z2), the radiation from the working III is sequentially directed at the OPP2, for each of them 5, the output signal is recorded, the calibration characteristic GX2 is plotted for the output signals for OBP2, also using a parametric type function, the parameters of which are determined from the measured output signals for OBP2, the calibration characteristic GX2 is used, using the output signal value for OBP2 and the GX1 function with its parameters found 5, the difference between GX2 and GX2, compares it with the specified value of the additional error dj, if the difference exceeds this value, then the numerical value of one of the combinations of the parameters GH1 is changed and GH2 is built every time and compared with GX2 until a combination of parameters will not be obtained, at which their difference will not exceed dj, construct GC dependences of the output signals for the NMP using GX1 with the numerical values of its parameters, except for the detected combination, the numerical values of which parameters values for m are output signals for NPM. 3 as9 hhg