SU1536215A1 - Method of photometric estimation of defect dimensions in direction of exposure to radiation - Google Patents

Abstract

The invention relates to flaw detection and can be used in the radiographic inspection of welded joints, overlays and base metal products. The purpose of the invention is to improve the accuracy of estimating the size of defects in the direction of transmission. The method implements the principle of mathematical modeling of a reference defect, which is as close as possible in its parameters (shape, transverse dimensions, depth) and image formation conditions (transmitted thickness of the metal and optical density of the background at the location of the defect) to the real defect compared with it. The process of such a simulation can be represented through the sequential introduction into the calculation formula of correction factors that take into account the difference in the corresponding parameters of the reference and real defects. A high degree of approximation of the parameters of a mathematically modeled reference defect to the corresponding parameters of a real defect compared with it is provided by using experimentally obtained dependences for estimating correction factors that reflect the real relationship between the contrast of the defect image and its specific parameters. In reality, it is impossible to model with a similar degree of approximation the required reference defect in the general case due to the limited set of reference elements and the impossibility of setting the reference simulator in the bulk of the metal (near the real defect). 9 il.

Description

The invention relates to flaw detection and can be used in the radiographic inspection of welded joints, overlays and base metal products.

The aim of the invention is to improve the accuracy of estimating the size of defects in the direction of translucence.

FIG. 1 shows a radiographing scheme of a welded joint, where the radiation source 1, the simulator reference 2, the weld 3, the defect | 4 welds, radiographic film 5; in fig. 2,3 - graphs of experimentally obtained dependencies of the values of () / del reference defects

The type of a rectangular groove is mm and mm wide and А 5 mm and mm depth, respectively, from the translucent steel thickness d (in Fig. 2 — X-ray radiation with a voltage on the X-ray tube of kV; FIG. 3; Iridium isotope gamma radiation; 192); solid line - reference defects from the source of radiation; dashed-reference defects on the side of the radiographic film adjacent to the surface of the product under test (4D is the contrast of the image of the defect, a is the coefficient of contrast of the radiographic film, taking into account the dependence of db on the optical density of image D); in fig. k, 5 - plots of experimentally obtained dependences., 4U / JD values of reference defects such as rectangular grooves of a certain width b and length of groove 1 at mm (curves 6.9), mm (curves 7, 10) and, b mm (curves 8, 25 11) for Co (FIG. k) and X-ray (FIG. 5) I in FIG. 6.7 - frequency-contrast characteristics (dependencies of the values and D / -J-.Q of reference defects on the value inverse of the transverse size of the defect) for reference defects such as cylindrical holes and rectangular grooves located on the side of the radiation source (o is the diameter of the hole, b - groove width); curve 12 - irt - 100 kV, MM, mm; curve 13 kV, MM, mm; curve 14 - Iridium-192, mm, mm; curve 15 - Cobalt-60, mm, mm; curve 16 - Iridium-192, mm, mm; curve 17 - Cobalt-60, MM, mm; curve 18 kV, MM, mm; cree30

 Rt

VA 19 - LUE-8 MeV, MM, 4d

10 mm; in fig. 8 - graphs of experimentally obtained dependencies of DTE / UF eigenvalues of reference defects such as a rectangular groove with a width of mm and a depth of mm from the depth of the defect r (r is the distance from the defect to the radiographic foam adjacent to the surface of the translucent product) for the translucent steel thickness mm : curve 20 - LUE-8 MeV; curve 21 - Cobalt-60; curve 22 - Iridium-192; curve 23 - U kV; in fig. 9 - the ratio of the magnitude of the reference defect ty35

40

45

50

55

5 „5

pas rectangular groove width mm and depth at d 100 mm and mm at d 100 mm, placed from the source (I) to the value xiD / j of this defect, placed from the radiographic film (And) adjacent to the surface of the test product, Dependences on the thickness of steel being through: curve 2k - LUE-8 MeV, curve 25 - Cobalt-60; curve 26 - Iridium-192; curve 27 - U kV; curve 28 - kV.

The essence of the method is to take into account the differences in the formation of images of the compared reference and real defects, if they have differences in shape, transverse dimensions, depth, and translucent thickness of the product at their locations. This accounting can be implemented when using in calculating the size of defects in the direction of transmission of the expression

ud

RA,. / .This. Av, Th.A /.R.A

-if-k

- eaEGA / tl). (

ZT.A

k ,, l a

v i 0 (GRDTs "(1)

where & dp-, ud

9Г.Д.

LV Shat-A

R. A ET A

Oo T in

I'm RA

.b, ka

dimensions in the direction of transmission of real and reference defects;

contrasts of images of real and reference defects; corresponding contrast coefficients of the radiographic film; density of the base metal of products; the density of the substance filling the real defect;

a correction factor that takes into account the difference in the translucent thickness of the product at the locations of compared reference and actual defects;

correction factors that take into account differences in the shape, transverse dimensions and depth of the reference, respectively515

real defects.

From the calculated expression, it follows that each given correction factor k., Kt can be determined through the ratios of values (equally filled reference and real defects (), assuming all other coefficients are equal to 1, i.e. comparing the values of / 3D / defects having the same dimensions in the direction of translucency (with the same filling) and differing only in the corresponding specific parameter: the translucent thickness of the product at the location of the defects, shape, transverse size, depth of bedding. The known translucent thickness of the product, known or previously estimated — the shape, transverse dimensions and depth of the reference and real defects; correction factors can be found from the dependencies of the values of the reference defects on their respective parameters (these dependences for the range of translucent steel thicknesses mm, energy radiation of 0.1–8 MeV and transverse dimensions of defects 0, mm were obtained experimentally and are presented in the form of graphs; some of the dependencies obtained are shown in Fig.2 - 9).

The coefficient k can be determined from the measured (known) values of the translucent thicknesses d at the locations of the reference and actual defects using the dependency graphs AD / fj, f (d) or (L1) / w) (a) - see FIG. 2.3

The shape of the defect can, to a certain extent, be characterized in terms of the ratio of the length of the defect 1 to its transverse size b. At 1 / b: 1, the real defect is characterized as local and is quite well imitated by a reference defect such as a cylindrical bore. With a real defect, it is characterized as extended and is simulated by a reference defect such as a groove. For extended defects of the same width, the difference in shape can be described by the difference in their length and, consequently, the coefficient k-f can be estimated from the measured lengths 1 of defect images using dependencies (l) - see FIG. A, 5.

g 0 5 o

0

five

five

five

The kft coefficient is estimated from the transverse dimensions (L, F) of defect images measured in the image using the dependences u0 / 7j, f (b, f) or frequency-contrast characteristics — see Fig. 6.7.

The coefficient krj is estimated using the dependences (c), see FIG. 8, according to a previously known or predetermined (e.g., double translucency method with a displacement of the radiation source) depth of a real defect.

The calculated expression is obtained for hollow (air-filled) reference defects. For real defects such as gas pores filled with incomplete air, air mines, fuses and f / () s1. For the case of defects with filling, for example, slag inclusions and for estimating the value of p / (p-rd), slag density is preliminarily determined separately for each specific type of welding and welding materials used.

The contrast ratio -vfl, determined from the characteristic curves of radiographic films, depends on the optical density of the image, therefore, introducing it to the formulas for calculating the size of defects in the direction of translucency allows you to rebuild from the effect of changes in the optical density of the image on the image area of the test section of the product (unlike 4D contrast, db / f does not depend on the optical density of the image D).

For films of type D, PT-5, D7 in the range of working densities of images, the coefficient uk kD, where k const, and the ratio of contrast coefficients in calculated terms can be replaced by the ratio of the corresponding optical densities: Uv1 2Tuj -) which simplifies the assessment, since there is no need to build characteristic curves for determining JD.

The positive properties of the method are ensured by taking into account the effect of scattered radiation on imaging defects (scatter blur, enhancing the contrasts of scattered radiation from images of defects close to the radiographic film), which leads to

The dependences of the image contrast of defects on the shape, transverse dimensions and depth of their occurrence are most noticeable in the region of large translucent thicknesses (in the known Vibix methods, the influence of scattered radiation is not taken into account).

Example. According to this method, sizes were evaluated in the direction of transmission of defects such as pipe welds of pipelines with a thickness of mm and d of 30 mm. Translucence was performed by X-rays with a voltage of 150, 300 kV on the X-ray tube and a D4 radiographic film. Lead reinforcing screens were used in accordance with GOST 7512-82. Geometric blur did not exceed 0.3 mm. The welded joint from the source of radiation was set up with standard simulators with a thickness of 1 and 2 mm, having semi-cylindrical grooves with a depth of 0.6 and 1.1 mm and a width of 2 mm. The cassette with radiographic film adjoined to the inner surface of the pipeline. Photo processing of the images was carried out on the Gevomatik automatic photo-scanning machine. Photometry of the images was carried out using the Helling-301 electronic-digital densitometer.

The calculation of the sizes of defects in the direction of transmission was carried out according to the calculation formula, specified for the case in question.

The use of defects of semi-cylindrical grooves, which imitate the shape of an armature, as a standard, allows one to assume the coefficient of differences in the form of 1f.fЈ1. The uthiae (as well as the grooves of the standard) are filled with air, the density of which is rl 0, therefore p / () g1

In this case, the difference

in the translucent metal thickness at the locations of the reference and real defects, an insignificant and coefficient kj 1. For a D4 film with a ratio of contrast ratios PDA, it is possible to replace the ratio of optical densities of defect-free areas of the image (background) adjacent to the images corresponding to

j.t. ET A / tch R.A

defects and f.

Thus, the calculation formula is simplified and takes the form

ten

15

20

25

nd 55

thirty

35

40

45

). (D T-A / up ka

. , ET.A

50

kr 4d)

The results of assessing the size of defects in the direction of transmission are presented in the table.

The table shows that the described method allows with a sufficiently high accuracy (error 2% d) to conduct a quantitative assessment of the size of defects (wets) in the direction of translucency. At the same time, in the area of large translucent steel thicknesses, the specified accuracy is ensured by taking into account differences in the transverse dimensions (width) and depth (location) of compared reference and real defects (for example, for mm, the relative error in the estimate fldp, carried out without considering differences in width and the location of the reference and real defects is k-5 times greater than in the assessment carried out taking into account these differences).

Thus, the method allows up to 5 or more times to increase the accuracy of the estimates of the sizes of defects in the direction of transmission. This reduces roughing and rejecting and, accordingly, increases the quality level of the products being tested. Improving the accuracy of defect size estimates increases the reliability of quality assessments of critical welded products, in particular, NPP equipment, and thereby reduces the likelihood of emergency situations. In addition, the proposed method reduces the complexity of radiographic monitoring of welded joints with uneven and removed weld reinforcement. It is promising to use the proposed method in radiographic testing of products using the systems currently being developed for automatic decryption of images, where calculations can be carried out using a computer.

Claims (1)

Invention Formula A method of photometric evaluation of the size of defects in the direction of transmission, consisting in the direction of electromagnetic radiation on a product, taking a radiographic image and subsequent measurements and comparing the optical density of the image the reference and actual detected defects in the radiographic image, characterized in that, in order to improve the accuracy of estimating the sizes of defects in the direction of transmission, after measuring the optical density values, the contrasts of the images of the reference and real defects are determined as the difference of optical densities of the defect images and the adjacent image the defect-free area of the image, then using the previously constructed characteristic curves of the radiographic film based on the measured values of the optical area The corresponding contrast ratio of the radiographic film is determined by the reference and real defects adjacent to the images of defect-free parts of the image, then the translucent thickness of the product is measured at the locations of the reference and real defects, the length and width of the image of the defects are measured, estimated from a previously known the location of defects or the method of double translucency with the displacement of the radiation source, the depth of the real defect is estimated from the known determined Based on the defect type or the known composition of the filling substance, or the method of welding, scanning and cutting similar welded reference samples to the object being monitored, the density of the actual defect filling substance, then using pre-built graphs of experimentally obtained image contrast dependencies of the reference defects related to radiographic film contrast ratios corresponding to the optical density of defects in defect-free areas adjacent to the images in the picture, from the translucent metal thickness, shape or length, transverse dimensions and depth of defects the values of the transmitted thickness of the product at the locations of the reference and real defects, the length and width of the images and the depth of the defects, determine the values of the correction coefficients and calculate the value dp toDM / DJT-).) " 4f (f-fA) J Wkg-V / id3 ™, 0 five 0 five 0 five RA , ET.D where is DSL, DS1 dor, / shat-A R. A D , EG.A AND RA Wkt dimensions in the direction of transmission of real and reference defects; contrasts of images of real and reference defects; Ratios of contrast of radiographic film corresponding to defect-free areas of the image near images of real and reference defects; the density of the base metal of the product; the density of the substance filling the real defect; a correction factor that takes into account the difference in the translucent thickness of the product at the locations of compared reference and actual defects; correction factors that take into account differences, respectively, in the shape or length, transverse dimensions and depth of the reference and actual defects. J V j L / r8 0.0 $ h: Iff203ff  2 1 Upr 20L  4CJ: i B 1rtf4 40 d, r sh CM vO fl Ш schf / vg)) "B" "a cs b "5i" 5i "5i t (jf / irrJ / C% / rrj fftjpf / trrJ / ezjCy / e :)