Method and device for detecting and characterizing a reflecting element in an object

Abstract

A method and device are provided for determining a mode of detection of an element that reflects ultrasonic waves, wherein it comprises at least the following steps: For each point P of a given volume Zr, determining an ultrasonic field value A.sub.ij.sup.m (P) for N emitter-receiver pairs (i, j) and for one detection mode m, computing a number $C^m\left(P,\stackrel{.fwdarw.}{n}\right)=\underset{i,j=1}{\overset{N}{.Math.}}{c}_{\mathrm{ij}}^m\left(P,\stackrel{.fwdarw.}{n}\right)$
of reflections of the wave where $c_{\mathrm{ij}}^m=\{\begin{array}{cc}1& \mathrm{if}.Math.{\stackrel{.fwdarw.}{n}}_{\mathrm{ij}}^m\left(P\right).Math.\stackrel{.fwdarw.}{n}.Math.=1\\ 0& \mathrm{if}\mathrm{not}\end{array}$
with {right arrow over (n)}.sub.ij.sup.m(P) the normal formed by the forward direction {right arrow over (d)}.sub.i and the backward direction {right arrow over (d)}.sub.j of the ultrasonic wave emitted and reflected by the reflecting element, computing the energy value E.sup.m(P,{right arrow over (n)}) for each point P of the zone Zr, with {right arrow over (n)} and for a plurality of modes m with $E^m\left(P,\stackrel{.fwdarw.}{n}\right)=\underset{i,j=1}{\overset{N}{.Math.}}{A}_{\mathrm{ij}}^m\left(P\right)c_{\mathrm{ij}}\left(P,\stackrel{.fwdarw.}{n}\right)$
comparing the obtained energy values E.sup.m(P,{right arrow over (n)}) to one or more threshold values Es, selecting the one or more energy values that meet the one or more threshold-value conditions, and deducing a mode to be used to construct a representation of the reflecting element.

Claims

1. A method for determining at least one detection mode m of reconstruction of a reflecting object having a portion capable of generating specular reflections of ultrasonic waves, within a given volume Zr, comprising at least the following steps: for each point P of the given volume Zr, determining an ultrasonic field value A.sub.ij.sup.m(P) for N emitter-receiver pairs (i, j) and for the one detection mode m, computing a number $C^m\left(P,\stackrel{.fwdarw.}{n}\right)=\underset{i,j=1}{\overset{N}{.Math.}}{c}_{\mathrm{ij}}^m\left(P,\stackrel{.fwdarw.}{n}\right)$ of reflections of the wave where $c_{\mathrm{ij}}^m=\{\begin{array}{cc}1& \mathrm{if}.Math.{\stackrel{.fwdarw.}{n}}_{\mathrm{ij}}^m\left(P\right).Math.\stackrel{.fwdarw.}{n}.Math.=1\\ 0& \mathrm{if}\mathrm{not}\end{array}$ with {right arrow over (n)}.sub.ij.sup.m(P) the normal formed by the forward direction {right arrow over (d)}.sub.i and the backward direction {right arrow over (d)}.sub.j of an ultrasonic wave emitted and reflected by the reflecting object, computing an energy value E.sup.m (P,{right arrow over (n)}) for each point P of the given volume Zr, with {right arrow over (n)} the normal to the reflecting object, and for a plurality of detection modes m, by summing, over the N emitter-receiver pairs (i, j) the product of the ultrasonic field value and the number of reflections of the wave: with $E^m\left(P,\stackrel{.fwdarw.}{n}\right)=\underset{i,j=1}{\overset{N}{.Math.}}{A}_{\mathrm{ij}}^m\left(P\right)c_{\mathrm{ij}}\left(P,\stackrel{.fwdarw.}{n}\right)$ comparing one or more obtained energy values E.sup.m(P,{right arrow over (n)}) to a given criterion and on the basis of the obtained result selecting the at least one reconstruction mode m to be used to detect and characterize the reflecting element.

2. The method as claimed in claim 1, wherein the at least one reconstruction modes m to be used are determined by choosing the energy values E.sup.m(P,{right arrow over (n)}) that are higher than a threshold value Es.

3. The method as claimed in claim 1, wherein the at least one reconstruction modes m to be used are determined while limiting a number of reconstruction modes to a given value, providing the best compromise between energy and dimension of a selected reconstruction zone.

4. The method as claimed in claim 1, wherein a computation of energy takes into account reflection coefficients R({right arrow over (n)}ij,{right arrow over (n)}) corresponding to an interaction of the wave with a planar surface defined by the normal {right arrow over (n)} to a point P: $E^m\left(P,\stackrel{.fwdarw.}{n}\right)=\underset{i,j=1}{\overset{N}{.Math.}}{A}_{\mathrm{ij}}^m\left(P\right)R\left({\stackrel{.fwdarw.}{n}}_{\mathrm{ij}},\stackrel{.fwdarw.}{n}\right)c_{\mathrm{ij}}\left(P,\stackrel{.fwdarw.}{n}\right).$

5. The method as claimed in claim 1, wherein the energy value is determined by taking into account the number of reflections, weighted by an arbitrary tolerance: $c_{\mathrm{ij}}^m\left(P\right)=\{\begin{array}{cc}1& \mathrm{if}.Math.{\stackrel{.fwdarw.}{n}}_{\mathrm{ij}}^m\left(P\right).Math.\stackrel{.fwdarw.}{n}.Math.=1\mathrm{.Math.}\\ 0& \mathrm{if}\mathrm{not}\end{array}.$

6. The method as claimed in claim 1, wherein the reflecting object is a planar defect, like a fault or notch, located in a part to be inspected.

7. A device for determining at least one mode of detection of an element that reflects ultrasonic waves, within a given volume Zr and by specular reflection, and for characterizing one or more defects in an inspected part A, comprising an ultrasound transducer for emitting and receiving ultrasonic waves and at least one processing device in which an estimator is executed, said processing device being configured to determine and execute one or more reconstruction modes obtained by implementing the steps of the method as claimed in claim 1 in order to reconstruct an image of the inspected part from signals received by the ultrasound transducer.

8. The application of the method as claimed in claim 1 to a detection of defects in a part subjected to ultrasonic waves.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The method and device are for example used to detect defects in a part subjected to ultrasonic waves.

(2) The invention will be better understood and other advantages will become apparent on reading the following description of nonlimiting examples that are given by way of illustration, which description is given with reference to the figures, which show:

(3) FIGS. 1a, 1b, 1c, an illustration of the various types of path,

(4) FIG. 2, an illustration of a reflection counter according to the prior art,

(5) FIG. 3, an illustration of a path respecting Snell-Descartes law,

(6) FIG. 4, an illustration of a reconstruction zone within an object, and, FIG. 5, a map of the estimator of the prior-art reflection counter for one reconstruction mode,

(7) FIG. 6, an example device for implementing the method according to the invention,

(8) FIG. 7, a result of mapping of the most favorable detection zones, which was obtained after implementation of the method according to the invention, and

(9) FIG. 8, an illustration of the prediction by the invention of the ability to detect two identical defects located in two different regions of the object A.

DETAILED DESCRIPTION

(10) The idea implemented for the method according to the invention notably consists in determining an adapted reconstruction mode, allowing an image of a defect to be precisely and reliably obtained, for one known inspection configuration and one type of sought-after defect. The mode thus obtained may then be used for the detection and characterization of defects in an object to be inspected.

(11) To be applied, the method according to the invention requires there to be known beforehand a certain number of pieces of information on the part to be inspected and on the testing device. The required pieces of information are then the same as those required to obtain the image of the defect.

(12) These pieces of information relate to:

(13) The part to be inspected: Knowledge of the geometry of the part, through its thickness, Knowledge of the velocities of propagation of the waves of interest (longitudinal (L) waves and transverse (T) waves), Other physical properties to be taken into account: anisotropy, attenuation of the waves in question, etc. to model the propagation of a wave in the part to be studied, and to perform the computation of the field,

(14) The one or more phase-array transducers of the sensor: The elementary division: linear, matrix array, annular, etc., The number of elements in question: N, The pitch, measured center-to-center, between the constituent elements of the sensor, in order to compute the times-of-flight required between an emitter, a focus point of the part and a receiver, the coordinates of the elements of the sensor being known, The type of coupling: coupling by immersion or by contact, The geometry of the wedge of the sensor, and its physical properties (velocity of the waves passing through it), in order to estimate the time-of-flight of the ultrasonic wave in the wedge. The total time-of-flight is the sum of the travel time in the wedge and the travel time in the material to be inspected,
The one or more positions of the sensor on the part during the acquisition of the signals: The configuration of the emitting and receiving elements, i.e. for each shot, the definition: of the one or more elements active in emission, with their associated delay law, of the one or more elements active in reception, with their associated delay law.

(15) The orientation of the sought-after planar defect, defined by its normal {right arrow over (n)},

(16) The set of reconstruction modes to be considered,

(17) The zone of interest Zr in which the reconstructions are performed.

(18) These parameters are implemented in the computation of the tool or estimator of the reconstruction mode according to the invention.

(19) For each possible reconstruction mode m, the invention consists in determining, on the basis of a specular-echo estimator, the capacity of this mode to be able to deliver or not, by specular reflections, the image of a planar defect, of known orientation, located at any point P of the zone of interest Zr or reconstruction zone. To do this, the estimator must deliver, for all this reconstruction zone Zr, an estimation of the ultrasonic energy reflected following this type of reflection, for example. The higher the energy computed by the estimator, the better the detection capacity of the reconstruction mode in question will be. The estimator is in addition capable of delivering comparable energy values for all the reconstruction modes in question. The adopted reconstruction mode will then be able to be implemented in a system for detecting and characterizing defects. The steps allowing this estimator to be obtained are detailed further on in the description.

(20) FIG. 6 illustrates an example device allowing the estimator according to the invention to be implemented during the inspection of a part. The device comprises a sensor 61 (ultrasound transducer for example) comprising a plurality of emitter-receiver elements 62(i,j), taking the form of a linear array for example, said elements being adapted to emit and receive ultrasonic waves, the detector for example being positioned on a holder 63. The emitter-receivers emit ultrasonic waves that are reflected within the part, which may contain a defect D. The reflected signals are captured by the receiving matrix array (receivers) and digitized using a principle known to those skilled in the art. The corresponding digital signals are, for example, stored in a file or a memory (not shown for the sake of simplicity) in order to be processed in real-time or subsequently. Simultaneously, the position of the sensor corresponding to a capture of signals reflected and captured by the receiving matrix array of the sensor is stored.

(21) The method according to the invention has allowed a tool or estimator that will allow, for example, one or more reconstruction modes that are the most effective at detecting defects in an examined part to be detected to be obtained. One way of implementing this estimator for example consists in transmitting the experimental data of the digitized signals and the position of the detecting sensor to a processing device 65 comprising a processor 66 on which the estimator 67 is executed in order to reconstruct an image of the inspected part from signals received from the detecting sensor and stored for example in a database 68.

(22) The processor may also comprise an output connected to a device 69 for displaying values thus obtained possibly taking the form of a map allowing an operator to locate those zones of the part in which a defect will possibly be best detected, as a function of a reconstruction mode, and therefore to select the best reconstruction mode.

(23) The method and device according to the invention may be used in the case of immersion tests, which assume that the device is submerged in a liquid, water in most cases, the waves then propagating through the liquid before refracting into the material. In another application, it is employed for contact tests, which assume that the sensor is placed on a wedge that then forms the intermediate medium between the sensor and the part to be inspected.

(24) For the implementation of the method according to the invention, a mesh is defined in the detection zone, in order to mark out the points P considered in the method according to the invention. The letter P designates the points of the mesh whatever their coordinates. The mesh is defined as a compromise between the obtainment of a quality image and the computation time. For example, the pitch of the mesh will be about /6 with the value of the wavelength or even comprised in the interval [/8, /4].

(25) The detection amplitude (i.e. the energy Ed) for a defect D may then be determined by summing the unitary contributions, of each emitter-receiver pair, computed at each of the points P of the mesh of the reconstruction zone Zr.

(26) For one reconstruction mode m, the specular-echo estimator, in its simplest expression, firstly consists in computing the following unitary quantities, for each point P of the zone of interest and for the following parameters: A.sub.ij.sup.m(P): the ultrasonic field for each set of emitter(s) i and receiver(s) j in the configuration defined during the acquisition. By configuration, reference is being made to the set of acquisition parameters that must be taken into account for the computation of the field, which parameters were listed above. The configuration may be computed by means of a software package for simulating the propagation of elastodynamic waves, for example the aforementioned software package CIVA developed and sold by the CEA, which is available in the publication, CIVA: An expertise platform for simulation and processing NDT data, Ultrasonics volume 44 Supplement, 22 Dec. 2006, Pages e975-e979, Proceedings of Ultrasonics Intemational (UI'05) and World Congress on Ultrasonics (WCU), c.sub.ij.sup.m (P): the reflection counter is equal to 1 if {right arrow over (n)}.sub.ij.sup.m(P) is collinear to the normal ii to the defect and to 0 if not. One way of estimating c.sub.ij.sup.m (P) may be given by:

(27) $\begin{array}{cc}{c}_{\mathrm{ij}}^m\left(P\right)=\{\begin{array}{cc}1& \mathrm{if}.Math.{\stackrel{.fwdarw.}{n}}_{\mathrm{ij}}^m\left(P\right).Math.\stackrel{.fwdarw.}{n}.Math.=1\\ 0& \mathrm{if}\mathrm{not}\end{array}& \left(3\right)\end{array}$
with {right arrow over (n)}.sub.ij.sup.m (P) the normal formed by the forward direction {right arrow over (d)}.sub.i and the backward direction {right arrow over (d)}.sub.j, for the reconstruction mode m, corresponding respectively to the path of the ultrasonic wave associated with the set of the one or more emitters i, and to the path of the ultrasonic wave associated with the set of the one or more receivers j. di designates the direction of the path of the ultrasonic wave originating from the element i and arriving at P, and dj the direction of the ultrasonic path reflected at P and retuming to element j of the sensor. These paths may be direct or via a reflection from the inspected part.

(28) The unitary contributions of each element of the sensor, for a given mode m and for a point P of the mesh then result from the product of the field A.sub.ij.sup.m (P) and of the reflection counter c.sub.ij.sup.m (P).

(29) The sought-after final ultrasonic energy (for selecting the reconstruction mode most suitable for detecting and characterizing a defect) is then determined by summing all the unitary contributions for each point P of the reconstruction zone. The points of the reconstruction zone are for example distributed in the form of a grid the vertices of which correspond to the points P. Denoting this energy E.sup.m(P,{right arrow over (n)}), one way of estimating it is given by the following relationship:

(30) $\begin{array}{cc}{E}^m\left(P,\stackrel{.fwdarw.}{n}\right)=\underset{i,j=1}{\overset{N}{.Math.}}{A}_{\mathrm{ij}}^m\left(P\right)c_{\mathrm{ij}}^m\left(P,\stackrel{.fwdarw.}{n}\right).& \left(4\right)\end{array}$

(31) For the reconstruction mode m, an energy value E.sup.m(P,{right arrow over (n)}) corresponding to the ultrasonic energy collected by the sensor (all of the elements) after reflection from a planar defect of known orientation is obtained for one point P of the mesh.

(32) The energy is computed for a plurality of possible reconstruction modes m, for all of the points P of the mesh and in the reconstruction zone in question. These energy values may be represented in the form of a map, or of a table that indicates, for each given point P of the mesh, the corresponding energy value, for one reconstruction mode.

(33) The method will then exploit these results in order to define the reconstruction mode m that is most suitable for detecting and characterizing a planar defect present in the reconstruction zone.

(34) According to one variant embodiment, the method will select the maximum energy value in the table and select the reconstruction mode m corresponding to this value, in order to execute for example a defect-seeking algorithm.

(35) For example, it is possible to compare the one or more energy values E.sup.m(P,{right arrow over (n)}) to a threshold value Es and, on the basis of the result obtained from the comparison, to select the energy values higher than this threshold value and therefore the modes to be used to detect and characterize a defect in a given reconstruction zone.

(36) Another way of proceeding consists in using an interval [Emin, Emax] of energy values to select the modes to be used, thereby limiting the choice to a given number of reconstruction modes. In this variant, the choice of the one or more reconstruction modes to be used results from a compromise between the number of reconstruction modes and the energy values that allow a good visualization of the defects.

(37) According to another variant embodiment, it is possible to improve the estimator defined in particular by formula (4) by taking into account other physical quantities such as the reflection coefficients R({right arrow over (n)}.sub.ij,{right arrow over (n)}) corresponding to the interaction of the wave with a planar surface defined by the normal {right arrow over (n)} to the point P. R({right arrow over (n)}.sub.ij,{right arrow over (n)}) may be defined by analytical formulae or by a software package for simulating the propagation of elastodynamic waves known to those skilled in the art, such as the aforementioned software package CIVA.

(38) One way of taking into account the reflection coefficients is the following:

(39) $\begin{array}{cc}{E}^m\left(P,\stackrel{.fwdarw.}{n}\right)=\underset{i,j=1}{\overset{N}{.Math.}}{A}_{\mathrm{ij}}^m\left(P\right)R\left({\stackrel{.fwdarw.}{n}}_{\mathrm{ij}},\stackrel{.fwdarw.}{n}\right)c_{\mathrm{ij}}\left(P,\stackrel{.fwdarw.}{n}\right).& \left(5\right)\end{array}$

(40) Another variant consists in applying an arbitrary tolerance (E) to the reflection counter c.sub.ij.sup.m (P), the following values of the counter being considered:

(41) $\begin{array}{cc}{c}_{\mathrm{ij}}^m\left(P\right)=\{\begin{array}{cc}1& \mathrm{if}.Math.{\stackrel{.fwdarw.}{n}}_{\mathrm{ij}}^m\left(P\right).Math.\stackrel{.fwdarw.}{n}.Math.=1\mathrm{.Math.}\\ 0& \mathrm{if}\mathrm{not}\end{array}& \left(6\right)\end{array}$

(42) The estimator according to the invention in particular allows a map that is more precise than those obtained by implementing known prior-art methods to be obtained and a reconstruction mode that is best suited to the geometry of a part and to a defect to be selected.

(43) FIG. 7 illustrates the result obtained by implementing the method according to the invention. This specular-echo estimator may be represented in the form of a map showing the spatial distribution of energy in the zone of interest Zr. In this case, the amplitude of the energies at each point P is coded using a palette of colors, which has been represented in FIG. 7 by zones in shades of gray with a scale corresponding to the amplitude of the specular echoes.

(44) An example map obtained with formula (4) is illustrated in FIG. 8 for a TTT reconstruction mode and for two planar defects located in two locations in the part. In this example, by virtue of the information provided by this estimator, it is possible to predict that the ability of the reconstruction mode in question to detect a vertical planar defect will be proportional to the calculated energy. Thus, in the regions in which the energy is low (or even zero), this type of defect will not be detected, and conversely, in the regions in which the energy is high, the detection capacity will be maximal. For example, in the zone 80, which is a region of high detectability, the planar defect D1 is correctly detected over all its height, as is illustrated in 81. For the zone 83 of lower detectability, the vertical defect D2 is poorly detected, this being shown in 84.

(45) The description applies to various reconstruction modes, for example, the TLT mode, the TTL mode, the LLT mode or other reconstruction modes known to those skilled in the art, such as those described in the document by Jie Zhang et al, entitled Defect detection using ultrasonic arrays: the multi-mode total focusing method, NDT&E International, 43(2010) 123-133.

(46) The method according to the invention makes it possible to quantitatively predict, for a given configuration, the capacity of a reconstruction mode to be able to detect a reflecting element in a precise region of an inspected zone.

(47) The applications are many and diverse, and include the detection and characterization (nature, position, orientation) of defects in nondestructive testing using ultrasound.