·Home ·Table of Contents ·Computer Processing and Simulation

An Interactive CAD Environment for Ultrasonic Evaluation of Welded Joints Fairouz Bettayeb, Amar Benchaala Scientific Research Center on Welding And Non Destructive Testing, C.S.C, Route de Dely Brahim, BP: 64, Chéraga, Algiers. Tel/fax: (213-2) 361850.Email: bettayeb@excite.com, f_bettayeb@email.com Mohammed Kassel, Karim Hamzi University of Science and Technology USTHB, BP:32, El Alia, Algiers. Contact

Abstract

With the ever-increasing cost of failure in welded structures like in the petrochemical, nuclear, electrical industries, there is necessity to provide and assure ever-greater degree of structural integrity in both pre-service and in service inspection. As part of this direction, procedures are being needed, which will provide the complete quantitative non destructive evaluation of the mechanical state and term life prediction of the structures.
A better configuration of pipelines ultrasonic inspection, requires a new fundamental understanding of wave propagation and distribution in the particular geometry and defects under investigations. This can be obtained by a combination of theory, numerical modelling and laboratory measurements.
The use of ultrasonic pulse echo technique to find flaws or cracks in welded structures has become common in lot of industry. The automation of this process has facilitated collection of quantities of data, whereas the analysis is still human support.
This paper seeks to proceed with this question by the modelling of the ultrasonic wave propagation and simulation of the testing interpretation. The ultrasonic system proposed is treated as an information processing system.
Keywords: Modeling, signal analysis, simulations, mechanical state, lifetime prediction, ultrasonic testing, defect characterization, joining.

1. Introduction

The major aim of non-destructive examination (Nde) of engineering structures and systems, is assurance of their integrity by the confirmation, during the construction process and the service life of the component, of zero failure. The most non-destructive examinations are carried out at welds, and the type of defects and their possible positions into the weld are described in standards and codes [1][2]. Weld inspection is usually accompanied by the separation between geometrical indications of the structure and indications from defects. Indeed, after examination and failure detection, the request is to find appropriate method for the defect identification, position, sizing, and acceptability. The recent developments in automated inspection technology have driven emerging techniques, and the fusion between testing technology and PC. technology become an essential feature.
Conventional ultrasonic testing is theoretically a very effective testing method. It is however, severely restricted by human reliability factors of variations in interpretation skills, ability to concentrate for long periods and also operator integrity.
Indeed, the standard evaluation of manual ultrasonic NDE is based on the study of the signal drawing, designed by A scan, B scan, C scan... views, mixed with the experience and know how of the inspector. And when complex defects are present, the analysis becomes a difficult task that requires much human effort. In the purpose to deal with this issue, we present in this paper the strategy of the USNDT₂₀₀₀ system, a software package programmed and built on object oriented modeling. This system attempts to reduce the examiner influence, thanks to the support of the pattern recognition approach.

2. Ultrasonic waves characteristics

Ultrasonic non-destructive testing is the use of ultrasonic signal, in the purpose to explore materials without destroying them. The ultrasonic waves are mechanical waves that propagate in an elastic medium and consist of atomic or molecular particle oscillations of a material, about their equilibrium positions.
Generally, the propagation equation is considered as linear if the attenuation effect is neglected. Supposing x is the propagation direction, p is the sound pressure, C is the sound velocity:

(1)

The ultrasonic beams are reflected from surfaces, refracted when they cross a boundary between two materials that have different acoustic impedance, and diffracted at edges or around obstacles [4].
In Ultrasonic inspection, sound beams of high frequency are introduced into materials for the detection of surface and subsurface flaws. The sound waves travel through the material with some loss of energy and are reflected at interfaces. The reflected beam is displayed and then analyzed to define presence and location of discontinuities. The reflection depends largely on the physical state of the interface and physical properties of the material. The relation between the reflected wave amplitude and the incident wave amplitude on a perpendicular surface is defined by the following reflection coefficient:

(2)

The literature has proposed several physical formulas and numerical data dealing with the acoustic wave propagation, and numerous authors have presented many of the fundamental aspect of elastic wave distribution. Some of these data are implemented in USNDT₂₀₀₀, and will be reported in a future contribution because of the page limitation.

3. Basic ultrasonic inspection methods

The two major methods of ultrasonic inspection are the transmission method and the pulse echo method. The pulse echo, which is the most widely used method, involves the detection of echoes produced when an ultrasonic pulse is reflected from a discontinuity or an interface of the tested piece. The frequency spectrum of the received signal, is determined by transmitting a pulse and using a frequency transfer function (FFT). However, because the spectral signatures of defects are influenced by several other factors [5] such ultrasonic signal attenuation and coupling details, the flaw characterization will depends on a qualitative interpretation of the signal in the time domain.
High power ultrasonic systems used in industry are usually composed of 3 elements:
• An electrical power generator which provides the electrical energy.
• An electromechanical transducer which converts by piezoelectric effect the electrical energy into mechanical vibratory energy.
• A propagation medium in which the transducer radiates the acoustical energy.
Physical analysis of such systems usually requires numerical modeling. In standard models, the transducer and the propagation medium are described using the finite element method. The models are linear and rely upon the theory of elasticity. A simulation of a propagation system can be summarized in the following algorithm.

4. General algorithm

1. After performing FFT on the A-scan data, frequency domain feature analysis is done.
2. The domain is scattered into a number of finite elements.
3. The displacement values interior the element are expressed in terms of their nodal values.
4. The nodal values are determined by minimizing the potential energy.
5. A system of matrix equations of second order in time is then obtained:

	(3)
	(4)

[F] : applied load vector
[M] : global mass matrix determined by the density distribution of the medium
[K] : global rigidity matrix determined by the elastic properties of the medium.
[¶u/¶t] : velocity vector
[D] : displacements vector

5. Usndt₂₀₀₀: a procedure for ultrasonic inspection modeling.

5.1 Signal acquisition
Once the condition of testing are defined, and the probe which characteristics have been taken from a krautkramer database[6] is truly positioned on the tested piece, the system performs an acquisition and a numerical representation of the received signal thanks to the simulation of a virtual oscilloscope.
Pulse echo waveforms are obtained from the weld area. This area includes the weld shape, surface and the weld root. The ultrasonic pulse is directed into the weld section at a specific angle, and a software is used to monitor and control the transducer pulsing, digitization of acquired signals and data storage.

Then the program generates from digitized A-scan data an image with high resolution. A computation of the path of the central beam is performed, and the expected response in amplitude is given. The limits of the area to be scanned by the probe, are computed and designed by A and B in fig.1. The path tracing is estimated considering the geometric reflections of the component as well as of the defects. And a 2D visualization of the beam travel inside the component is displayed. If a defect is detected, the system displays its depth P and its position by computing the distance S, displayed in fig.1.
Signal analysis

The implementation of the philosophy of the ultrasonic inspection requires: to register, during the transducer scanning, all signals reflected from the defects, to determine linear co-ordinates of the beginning of each defect and its extension and to have parameters of the inspected joint into the memory [7].

Fundamental and applied aspects of the mathematical background of ultrasonic modeling, computer technologies and analytical methods developed for non-destructive evaluations have produced new advanced techniques for flaw identification. The simulation models of ultrasonic wave propagation, in solid media contribute to reduce the number of laboratory testing and to learn more about inspection results [8]. These models can be classified into 3 categories: exacts solutions, approximate solutions and numerical solutions .

USNDT₂₀₀₀ performs a computerized model of the signal propagation, a qualitative agreement between theory and experiment ultrasonic interpretation methods. So, in addition of the numerical modeling formulated in (3) & (4) of the general algorithm below, experimental methods[9] such as the triangulation, the DGS diagram, -6dB, have been implemented and drawn in fig.1 and fig.2.

Fig 1: "USNDT2000 software package"

Fig 2: "DGS diagram calculation

The oscillations are scattered from defect in the direction of the transducer, the echo signal obtained is selected in time and the transfer function of the defect is analyzed. Once the A-scan traces are selected by USNDT₂₀₀₀ system, the parameterization is automatically performed and displayed in fig.1, both in time and frequency domain (pulse width, run time, echo, and energy...). The combination of detected signal data, with all informations on the welded joint, component geometry, material properties and the welding process, allows extracting relevant data for better defect characterization.

5.2 Displaying results
When the flaw has been sized, a pattern recognition software [10] introduced in USNDT₂₀₀₀ performs its identification by the use of a set of normalized databases about planar and volumetric defects. Then the reference, the severity and the kind of the defect are displayed. And a short definition and comments on the site and orientation of the flaw are given. At last, USNDT₂₀₀₀ performs a graphical representation of the discontinuity.

6. Conclusion

At this time, most non destructive evaluation computer-based decision making applications, have been developed only to act as an aid to the human operator. These systems can resolve some problems associated with human influence as fatigue, inattention, variability and subjectivity. However, future goals must be directed towards replacing the operator entirely. This raises the question of how efficiently the computer based decision maker, can be developed to provide "error free" decisions. The future possibility of USNDT₂₀₀₀ system is a project to implement a computer automatic scanning and signal processing, with extensive imaging capabilities.

References.

1. Iso (1982). Classification des défauts dans les soudures par fusion des métaux, avec commentaires explicatifs. 1^ere édition. Paris.
2. Eurotest, (1985). Collection of references radiographs of but welds in steel. International institute of welding (Ed.). Brussels.
3. Nockeman, C., Fortunko, C. (1997). In: Proceedings of conclusions of the workshop European-American workshop determination of reliability and validation method of NDE. Germany proceedings (Ed.). Berlin.
4. You, C.P., Knott, J.F. (1986), Engineering fracture mechanics, Vol.24, N°2, pp.291-305, Elsevier (ed.), New York.
5. Nichols, R.W. (1982). Advances in non destructive examination for structural integrity. Elsevier science publishing co., Inc. New York.
6. Krautkramer (1998). A database CD: The new world of krautkramer sounds round the globe. Krautkramer GmbH & CO (Ed.). Germany.
7. Bettayeb, F., Alem, K, Maamache, B. (1998). In: Proceedings of the 7^th European conference on non destructive testing: Ndt at work. Vol. 2, pp.1258-1265, force institute (Ed.). Copenhagen.
8. Paradis, L. (1983). Doctor Engineer Thesis , ENPG, Grenoble, France.
9. Pelletier, J.L (1984). La pratique du contrôle industriel par ultrasons. Tomes I & II, ENSAM (Ed.), Paris.
10. Bettayeb, F. & al. (1997). In: Proceeding of the 2^nd Convention for non destructive testing: Ndt to improve quality, reliability and productivity. pp. 179-185, Saint & Gerstel graphics (Eds), South Africa.

© AIPnD , created by NDT.net

|Home|    |Top|