Abstract

(This is the abstract of a paper that will be presented at the 7th ECNDT in Copenhagen. The article presented here introduces the same work)
Ultrasonic arrays offer a number of advantages in NDE applications, the most important is electronic beamforming and scanning enabling generation of the desired field distributions and defect insonifica-tion from different directions resulting in improved defect detection, and characterization.
The applicability of linear array technique for inspection of copper lined canis-ters for nuclear waste fuel has been investigated recently at Uppsala University.
The objective of this project is development of ultrasonic array technique for assessing the integrity of the circumferential electron beam (EB) weld between the lid and walls of copper lined canisters developed by SKB (Swedish Nuclear Fuels and Waste Management Co.) for encapsulation of nuclear waste.
Due to the radiation emitted by the nuclear waste encapsulated in the canisters the inspection must be carried out completely automatically. In order to achieve a very high level of confidence in the NDT results of copper canisters, the inspection system should be capable of inspecting 100% of cimcurferential weld zone in thick section copper. An ultrasonic array system consisting of a linear ultrasonic array, made of piezoelectric composite material, and a computer controlled multi-channel electronics is proposed as a solution to this problem.
Modern array technique which has become recently commercially available for NDT applications has a great potential in this application. The array sys-tem chosen for this application enables electronic focusing and rapid electronic scanning eliminating the use of complicated mechanical scanner. Secondly, it is capable of reducing the influence of grain noise by using a combination of focused sound field and signal processing. Thirdly, it is characterized by inherent flexibility enabling 100% inspection of the weld zone even if the weld is not parallel to the surface.
Our present research activity in this project addresses two issues: the development of software tools for efficient beam-forming, and a laboratory verification of the performance of ultrasonic array system on real EB welded canister samples. A special attention is paid to ultrasonic grain noise scattered by the internal copper structure. A method for grain noise suppression by electronic filtering B-scan images is verified in this application. Here we present some recent results obtained during the inspection of EB welded copper test blocks containing both artificial and natural defects.

Objectives

The objective of this project is to develop a fully automated ultrasonic technique for assessing the integrity of the circumferential weld between the lid and wall of copper lined canisters for nuclear waste fuel. The aim of this research was to evaluate performance of linear ultrasonic arrays used for the inspection of the electron beam (EB) weld in copper canisters, and in particular:

• to evaluate of the ultrasonic array system Allin using samples of canisters with EB weld
• to develop software for modeling elastic fields in immersed solid
• to investigate ultrasonic noise (backscattering) resulting from copper structure

Scope of work

Our research consists of a theoretical and an experimental part. The theoretical part concerns development of modeling tools enabling efficient beam-forming of the ultrasonic arrays. The experimental part includes inspection of various canister samples using the ultrasonic array system ALLIN from NDT Systems and the linear ultrasonic array from Imasonic. The research activity in the project include the following tasks:

• Extensive laboratory tests . The experiments with the ultrasonic array in immerse mode performed using mechanical scanner on copper blocks with natural defects provided by SKB. A number of B- and C-scan images has been acquired for evaluation of overall performance of the array technique.
• Beam-forming for the linear ultrasonic array. Software package for modeling sound field propagated by an ultrasonic array into a two layer medium (liquid-solid) has been developed and tested. The cases of oblique and concave array surface have been included into the simulation program.
• Grain noise problem. The ways of grain noise estimation using ultrasonic array have been investigated using available copper samples. Electronic noise reduction methods have been adopted for the ultrasonic array and investigated experimentally.

Laboratory tests

Experimental set up


Figure 1 Figure 2 Figure 3

A number of copper blocks (canister parts) was inspected in immersion using the phase array technique. The experimental set up which has been used for the experiment consisted of Allin array system, linear phase array made of pizocomposite material and the precision mechanical scanner with water tank, see Figure 1.

All measurements have been performed using 128 channel ultrasonic array system Allin from NDT System which was connected to 3 MHz linear array from Imasonic. The array was made of piezoelectric composite material and had 64 elements with dimensions 1mm by 33.5 mm. The array was fast focused in one plane (cylindrical focus with focal distance 190 mm in water) and waselectronically focused in the other plane. The Allin enabled focusing both in the emission and reception modes and electronic scanning using an aperture up to 32 elements. Besides scanning and beam-forming the Allin performed also data acquisition and steering of the mechanical scanner used for guiding the array. The Allin enables storing all the measured data in real-time, which means that all RF signals included in the acquired B-scans can be stored. Thus when inspecting a 2D surface a sequence of B-scans is stored forming “data cube“. The stored data can then be processed and presented off-line and the desired C-scans can be calculated. This feature has been used extensively during the experiments.
The experiments were performed with the array radiating from the top of the block into the copper and the EB weld located approx. 50 mm below the top surface (see Figure 2 for details).

Detecting artificial and natural defects

Sensitivity to the artificial defects has been investigated using the blocks provided with various side drilled and flat bottom holes in the weld zone as well as natural defects.
Based on the inspection of artificial holes, we can say that the ultrasonic array technique is capable of detecting artificial defects (drilled holes) in the weld zone of the copper canister in the diameter range of 1 mm to 3 mm depending on the type of defect. Based on this we can estimate that it would be capable of detecting single pores with diameter >2 mm. Extensive evaluation of the sensitivity and resolution is performed using other blocks made of various grades of copper.

Grain noise suppression

Background
Metals with distinct coarse grains are difficult to inspect ultrasonically due to the attenuation and backscattering from the grain boundaries. This effect becomes very distinct for the ultrasonic waves with lengths in the same range as the grain size. In pulse echo inspection small reflections scattered from the grain boundaries appear in the received signal as a kind of noise often referred to as grain noise. Grain noise is different from thermal noise since it does not vary with time. This means that the for each transducer position one particular noise realization is obtained, which is observed as a constant signal on the screen. Due to this fact the grain noise cannot be reduced by averaging in time but it can be reduced by averaging over different transducer positions, the procedure called spatial diversity. Another way to reduce the grain noise is to apply frequency diversity algorithms, referred to as split spectrum processing (SSP). The SSP is based on the analysis of narrow band signals obtained by splitting the transducer frequency band in a number of small frequency bands by means of FFT (fast Fourier transform). Several SSP algorithms have been proposed with different schemes of analysis of split signals [3]. Their common disadvantage is sensitivity to parameters set by the operator and degraded temporal resolution of the output signal. A number of more robust schemes has been proposed such as cut spectrum or consecutive polarity spectrum [3], and more recently new algorithms based on a concept of noncoherent detection were developed [4]. The latter algorithm was tested on ultrasonic data obtained from copper block with coarse grain structure.

Signal Processing Procedure
The noncoherent detector (NCD) is a nonlinear filter used for detecting transients embedded in a background noise in radar and communication applications. It has been adopted to ultrasonic noise suppression by M. Gustafsson, the details have been presented in [4], [5].

Figure 4

The idea is to construct a filter processing the original ultrasonic signal to yield a signal in which the echoes are more likely to be detected. The filter suppresses the ultrasonic clutter (grain noise) and maximizes the probability of detecting certain family of transients. The noise suppression is based on the estimated noise covariance matrix and the detector is designed given a prototype transient. The NCD is more advanced than an ordinary matched filter since it aims at detecting a member of a predefined transient family. In our case the family has been defined as a set of transients with given envelope and center frequency but unknown phase.


The NCD algorithm has been implemented using Matlab^(r) (from MathWorks) so that it can be used for processing B-scan images in data cube acquired by our array system. When all B-scans in the data cube have been processed C-scans can be extracted from the processed data, example is shown in Fig. 4.

Beam-forming of ultrasonic arrays in immersed solids

Background
Ultrasonic beam-forming is usually performed by means of ultrasonic arrays [6,7]. Arrays can be differently shaped and of different dimensions. Most commonly used are the one-dimensional arrays, like linear arrays consisting of equally spaced elemental radiators (or elements for abbreviation) laid out in a straight line, concave and convex arrays made up of elements aligned around a curve, usually a cylinder. A linear array can be used as a phased array provided that the hardware is furnished with a complicated time delay device to steer beams. Two-dimensional arrays have also been put into use, such as planar arrays composed of elements oriented on a geometric grid in a plane. The arrays of our interest are linear, phased and concave arrays.

To make best use of the ultrasonic array system ALLIN for NDT of materials, it is necessary to get knowledge of how waves from an array propagate in the materials inspected. With this purpose, we started the theoretical study of wave propagation in solids, and up to now, we have developed a software package for modeling the wave propagation. The model is implemented on personal computers, and can simulate acoustic fields in fluids, and longitudinal wave (LW) and shear wave (SW) in immersed solids (with planar interfaces) from linear, phased, and concave arrays. The fields simulated can be time-harmonic or transient, can be in far or near regions [1,6,12-14].

The theories for ultrasonic array beam-forming used in NDT [1,6,7] are, in the case of homogeneous fluid medium and in far-field range, well established, and similar to those well established in phased array antennas [8] and microwave phased arrays [9]. They treat the array elements as point-like radiators. However in NDT applications, the problems of interest are often in the near-field range, and the physical size of the array elements must be taken into account on calculating sound fields because it is not small compared to the wave length. As presented in previous sections, the linear array is used to the immersion test of metals. Elastic fields radiated by the array into copper block, therefore, are of our interest. To this end, we make advantage of the angular spectrum approach (ASA) and developed a software package for modeling elastic fields in immersed solids from linear, phased and concave arrays excited by time-harmonic and transient waves. The model has shown to be effective and efficient for treating elastic fields in immersed solids both in near- and far-field ranges. Here we present the basic theory of array and the model for elastic fields in the case of phased arrays.

Geometry and notation for investigation of elastic fields in immersed solids from phased array. Geometry and notation for investigation of elastic fields from curved source in immersed solids.

Basic Theory of Beam-Forming by an Phased Array [6-9]
An ultrasonic phased array used in NDT usually consists of a number of linearly aligned rectangular elements such as the one shown in Figure 5, where the has N' elements and the elements are rectangular with the dimension of 2ax2b, spaced with d and located in a fluid at plane z = 0. In homogeneous fluids and in far-filed, the conventional theory of array applies [6,7]. The far-field beam-pattern of an array is an important factor used to evaluate the performance of the array used in certain specified situation [8,9]. A rectangular piston-like radiator has a beam-pattern of two-dimensional sinc function [6]. By means of the array theory, the beam pattern can be obtained which is the product of the beam-pattern of a linearly-aligned point element array and that of a rectangular piston source [6,10]. Since the elements can be excited separately by either time-harmonic or pulse waves, the phases (or delay times in the pulse case) and amplitudes of exciting waves to the elements can be given separately. Specifying the phases in different ways, the array can generate different kinds of beams, such as steered and focused beams. For details of the basic theory, refer to references [6-9].
A Model for Elastic Fields in Immersed Solids from Phased Arrays [6,12-14]
The above introduced basic theory can only be used to treat beam-patterns in the far-field range of a homogeneous fluid. While our interest is mostly in the immersion test of block in near-field. As is well known, when a beam radiated by the array impinges on the fluid/solid interface, it will experience mode conversion and refraction. The mode conversion results in different modes' waves simultaneously propagating in the solid, for example, longitudinal wave (LW) and shear wave (SW), which are the two most commonly used modes. The refraction brings about the deflection of the incident beam in the solid. This means that the evaluation of such an elastic field is not an easy job. To deal with near-fields from arrays both in homogeneous and in layered media, we have developed a model based on the ASA.

A general procedure of applying the ASA to the calculation of elastic fields in immersed solids involves two main steps. The first step is the analysis of a source or a field from the source at a certain plane in a fluid, that is, the decomposition of the source or the field into a set of plane waves; and the second step is the synthesis of fields in the immersed solids. A field at point (x, y, z) in an immersed solid is synthesized in such a way that the decomposed plane waves propagating from the fluid into the immersed solid at point (x, y, z) are superposed with consideration of mode conversion at the fluid/solid interface. Since the ASA represents a field from a source as a set of plane waves, the solutions of fields from finite sources with different shapes reduce to the solutions to the plane wave problems. This is very helpful to solve wave problems involving layered media. Solution of elastic fields in immersion measurement is one of the examples.

For planar sources (e.g. phased arrays and tilt linear arrays), the sources can be directly decomposed by means of the 2-D spatial Fourier transform (see Figure 5). For curved sources (e.g. concave arrays), however, the Fourier transform can not be directly applied (see Figure 6). Thus a field, here called initial field, from a concave array is first calculated in a certain plane (which is normally close to the source and parallel to the fluid/solid interface), and then decomposed into an angular spectrum of plane waves by means of the Fourier transform. The latter method can be applied to the sources with arbitrary shape. The initial field can be thought of as a secondary source, a type of unbaffled planar source, equivalent to the primary curved source. In this sense the latter case is unified to the former case.

Fig 7: Magnitudes of the LW angular spectra of the nonfocused phased array in copper in the cases of steering angles (a) 0°, and (b) 10°, respectively. Fig 8: Magnitude of LW velocity field of the nonfocused phased array for steering angles (a) 0°, and (b) 10°, respectively. Fig 9: Magnitude of the LW velocity field of the nonfocused phased array with focusing the LW for steering angles (a) 0°, and (b) 10°, respectively. The focal point is supposed to lie in copper on the beam axis at a distance of 80 mm from the array.

Examples of Elastic fields in Immersed Copper

Here are some examples given to the LW fields radiated by a phased array with N' = 32, d = 0.5 mm, 2a = 0.4 mm, and 2b = 12 mm. The array is positioned in the x-y plane at z = 0 and has its center in the origin of the coordinates above a flat surface copper block immersed in water (see Figure 5). The densities and the sound speeds used for calculations are 1000 kg/m3 and 1500 m/s, respectively in water, and 8960 kg/m3, 4660 m/s (LW) and 2260 m/s (SW), respectively in copper. The array is excited by a time-harmonic wave with frequency 3 MHz and unit amplitude. The calculation of the angular spectrum is implemented according to the method proposed in [12-14]. In contrast to the fluid case, the angular spectra of the array in immersed copper are derived from the corresponding angular spectrum in the fluid by using the transmission coefficients at the fluid/solid interface [6]. The simulation of elastic fields is performed from based on the method in [6].

The examples of the LW beam-patterns, the LW nonfocused and focused elastic fields in immersed copper are shown in figures 7-9, respectively. Figure 7 shows the angular spectra of the nonfocused array at steering angles of 0 and 10 degrees, which are presented as color scaled 2-D images. Figure 8 shows the nonfocused LW field radiated by the array at steering angles of 0 and 10 degrees. Figure 9 shows the focused LW field radiated by the array at steering angles of 0 and 10 degrees. The fields are both in the y=0 plane.

Comparatively observing figures 7 and 8, we can find the corresponding lobes of the field in figure 8 to those of the beam-pattern in figure 7. This demonstrates that a beam-pattern in an immersed solid can be easily determined from the angular spectra of the array in the solid. By comparing figures 8 and 9, it can be seen that the main lobe of the focused field in figure 9 is narrower in width and higher in maximal amplitude than that of the nonfocused field in figure 8. This illustrates that the focusing increases the lateral resolution and echo strength around the focal zone.

References

1. T. Stepinski, P. Wu, Ultrasonic Inspection of Nuclear Copper Canisters, SKB Projektrapport 97-01, December 1996.
2. A.B. Day, An Investigation into Optimisation of NDT of Spent Nuclear Canisters - Phase III, report TWI 220343/1/95, April 1995.
3. T. Stepinski, L. Ericsson , B. Eriksson and M. Gustafsson, Quasi Frequency Diversity Processing of Ultrasonic Signals - A Review, in Advances in Signal Processing for Non-Destructive Evaluation of Materials, X.P.V. Maldague, ed. Kluwer Academic Publ., 1994, pp. 49-58.
4. M.G. Gustafsson, "Statistical Aspects of the Split Spectrum Technique". Doctoral Dissertation, Acta Universitatis Upsaliensis, Uppsala University, Sweden 1995. ISBN 91-554-3490-8.
5. M.G. Gustafsson, "Nonliner Ultrasonic Clutter Supression Using Noncoherent Detector Statistics and Split Spectrum Processsing". UPTEC 97 063R. April 1997.
6. T. Stepinski, and P. Wu, Ultrasonic Inspection of Nuclear Copper Canisters, SKB Projektrapport 97-08, August 1997.
7. A. Macovski, "Ultrasonic imaging using arrays," Proc. IEEE 67: 484-495 (1979).
8. W.H. Kummer, "Basic array theory," Proc. IEEE 80: 127-140 (1992).
9. B.D. Steinberg and H.M. Subbaram, Microwave Imaging Techniques. John Wiley & Sons, Inc., New York, Chap. 3 (1991).
10. P. M. Morse and K. U. Ingard, Theoretical Acoustics. McGraw-Hill Book Company, New York, Chap. 7, (1968).
11. L.M. Brekhovskikh, Waves in Layered Media. 2nd Edition, Academic, New York, Chap. I, (1980).
12. P. Wu, R. Kazys and T. Stepinski, "Analysis of the numerically implemented angular spectrum approach based on the evaluation of two-dimensional acoustic fields--Part I: Errors due to the discrete Fourier transform and discretization," J. Acoust. Soc. Am. 99: 1339-1348 (1996).
13. P. Wu, R. Kazys and T. Stepinski, "Analysis of the numerically implemented angular spectrum approach based on the evaluation of two-dimensional acoustic fields--Part II: Characteristics as a function of angular range," J. Acoust. Soc. Am. 99: 1349-1359 (1996).
14. P. Wu, R. Kazys and T. Stepinski, "Optimal selection of parameters for the angular pectrum approach to numerically evaluate acoustic fields," J. Acoust. Soc. Am, 101: (1997).

Tadeusz Stepinski
was born in Szczecin Poland, in 1950. He obtained his M.Sc. degree in electrical engineering from the Technical University of Szczecin, and a Doctor of Technology degree from the Technical University of Warsaw, Poland in 1973, and 1983, respectively.
From 1973 to 1984 he was with the Institute of Industrial Automation in the Department of Electrical Engineering of the Technical University of Szczecin, where he served as a lecturer teaching courses on the subjects of systems identification and modelling, automatic control, industrial control systems and electronic circuits.
In 1984 he joined Swedish company AB Sandvik Bergstarand where he was actively involved in developing eddy current systems for non-destructive testing of hot steel and in research on the application of modern signal processing methods to non-destructive evaluation. Since 1988 he has been with Uppsala University where he is Associate Professor at the Department of Material Science. He heads a research group working with the application of modern signal processing to NDE. His research interests are in the area of signal processing, pattern recognition and neural networks applied to ultrasound and eddy current data. Dr. Stepinski is a member of IEEE, Acoustical Society of America, ASNT, he is a chairman of the Committee for New Techniques and Research in FOP (Swedish Society for NDT).
Address:
Signals and Systems Group
Uppsala University
P O Box 528
SE-751 20 Uppsala, Sweden
Phone: +46 - 18 - 4711076 Fax: +46 - 18 - 555096
E- Mail: Tadeusz.Stepinski@signal.uu.se
http://www.signal.uu.se/

Ping Wu
received the B.S. degree in electrical engineering in 1982 from Nanjing Aeronautic and Astronautic University, Nanjing, China, and the M.S. and Ph.D. degrees from Jiaotong University, Xi'an, China, in 1987 and 1991, respectively, in the fields of biomedical ultrasonics and ultrasonic engineering. From 1992 to 1994, he worked as post-doc in the Fraunhofer Institute for Biomedical Engineering, St. Ingbert, Germany, in the area of ultrasonic medical image processing and ultrasound tissue characterisation. Since late 1994 when he joined the Department of Technology at Uppsala University, Uppsala, Sweden, he has been working on elastic wave propagation and scattering and on ultrasonic imaging with applications to NDT.
E- Mail: Ping.Wu@signal.uu.se


Eider Martinez, M.Sc., lab research assistant.