WO2007071735A1 - Procede universel de modelisation des interactions entre au moins une onde et au moins un objet - Google Patents
Procede universel de modelisation des interactions entre au moins une onde et au moins un objet Download PDFInfo
- Publication number
- WO2007071735A1 WO2007071735A1 PCT/EP2006/070020 EP2006070020W WO2007071735A1 WO 2007071735 A1 WO2007071735 A1 WO 2007071735A1 EP 2006070020 W EP2006070020 W EP 2006070020W WO 2007071735 A1 WO2007071735 A1 WO 2007071735A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- interface
- sources
- point
- matrix
- medium
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
Definitions
- the present invention relates to a universal method for modeling interactions between at least one wave and at least one object, the surface of each object defining an interface between at least two media.
- French Pat. No. 2,847,051 a method for evaluating a physical quantity representative of an interaction between a wave and an obstacle is known.
- the implementation of this method is of great utility in applications such as non-destructive testing. However, it can be applied in a simple way only in the case of a single interface separating two media. Indeed, this method is not generalizable to interactions with objects having multiple media, in particular because of the very many reflections and transmissions generated by the ultrasonic wave incident to the passage of the interfaces.
- the subject of the present invention is a method for modeling the interactions between at least one wave and at least one object, the surface of each object defining an interface between at least two media of different physical properties, a method that is simple to implement and which is applicable in different domains of physics, for continuous, sinusoidal or pulsed temporal regimes, whatever the composition of the studied system.
- system which is, in the sense of the present invention, a domain of the space in which take place the interactions that one seeks to model.
- the method according to the invention is a method of modeling the interactions in a system between at least one wave and at least one object, the surface of each object defining an interface between at least two media, and it is characterized in that it comprises the following steps: one chooses the set of elementary characteristic functions corresponding to the considered domain of application (El), the physical properties of each medium considered composing the system are defined, the geometric structure of each object of the system is defined by meshing it, and at least one test point is positioned on the surface of each mesh, at each test point the minus one test quantity for each medium considered in order to establish continuity equations for the boundary conditions, we associate on each side of each mesh at least one elemental point source, we position the objects with respect to each other in the space, we associate media to the volumes delimited by the objects, we determine the type of boundary conditions for each interface, we build the global matrix of interactions between the different objects, this matrix consisting of at least one matrix block characterizing the interactions between objects taken in pairs, these interactions being related to the propagation of the wave in the environment chosen by the elementary characteristic functions chosen
- macroscopic quantities are calculated in at least part of the system.
- the physical quantities created by the set of elementary point sources considered are visualized.
- closed-volume objects whose surface is closed each time and defines a boundary between the medium outside the object and a medium internal to the object are defined and open-volume objects whose surface is open because it represents an interface between semi-infinite media.
- this interface is created, it is bounded laterally by a set of boundaries that are those of the workspace in which the interaction is studied.
- FIG. 1 is a simplified flowchart of a mode of implementation of the method of the invention
- - Figures 2 to 8 are simplified diagrams of systems on which is implemented the method of the invention
- FIGS. 9 to 19 are very simplified diagrams of non-homogeneous media on which the method of the invention is implemented
- FIGS. 20 to 27 are diagrams relating to the application of the method of the invention to non-destructive eddy current checks.
- the important feature constituting the core of the invention is to spread the boundary conditions step by step, thanks to layers of point sources responsible for the synthesis of quantities in a given medium, these layers of sources being arranged on each side of each of the interfaces.
- the modeling of these interactions consists, according to the invention, of equate the response of each of the media, taken separately, to a point solicitation, which results in an elementary characteristic function per medium considered.
- the physical nature of the sources used corresponds, of course, to the nature of the wave that is to be synthesized.
- the calculation of the response to a solicitation gives, in general, a scalar, vector or tensor potential and the characteristic quantities of the wave.
- the total response of the system is calculated medium by medium, summing the contributions of all the point sources radiating in the medium considered.
- the global resolution implies that the boundary conditions at the passage of each of the interfaces are satisfied by imposing the continuity of the potential and its derivative, according to the normal at the considered interface, to a constant close. This set of conditions makes it possible to obtain said global matrix.
- the number of equations obtained at each test point determines the choice of the number of point sources placed on each side of the interface.
- electromagnetism if the considered mediums are conductive, it will be noted that the potential is a vector, as well as its component according to the normal to the surface of the considered interface. For each test point, six equations with six unknowns (ie three components of A and three components of
- the sources composing these triplets are homogeneous with current elements respectively oriented along the Cartesian axes of these triplets.
- the equation of the system is thus made in a global manner using the single global matrix that takes into account all the interactions between the different media. Then, we calculate the value of the energy radiated by each source using the inverted global matrix and user-defined boundary conditions, as described below.
- said elementary characteristic function is a function of Green.
- a line segment has represented an interface portion on each side of which, or on one side of which, spherical surfaces containing at least one point source have been represented, these surfaces being tangent to the interface.
- These point sources and the spherical surfaces which contain them result from a mesh which models, within the meaning of the aforesaid French Patent 2,847,051, either a transducer or an interface.
- circles in light gray correspond to intrinsic boundary conditions
- circles in dark gray correspond to boundary conditions set by the user.
- Each mesh is associated with a spherical surface containing at least one point source. In a particular case, the point of contact between these spherical surfaces and the interface is a test point, but it is understood that one can distribute several test point in each mesh.
- the total number of parameters unknown to the sources must be equal to the total number of equations available, the number of equations being adjusted by a corresponding choice of the total number of test points.
- CLUs reflect the presence of excitation in the system.
- the figure above represents for example a valid CLU towards the medium n + 1.
- the sources of the Middle n + 1 do not exist, because one does not calculate a magnitude in the Middle n, but it is understood that one does not exclude the possibility of placing sources in the Middle n + 1, these sources radiating in the Middle n.
- the electrostatic potential expressed in volts, on an electrode of a capacitor, is constant and generally fixed by the user.
- the coordinate block (i, j) represents the part of boundary conditions computed on the interface of the object i and due to the influence of the object j (the objects i and j interact in a common medium).
- the total number of source components must be adapted to the total number of test points.
- the vector block CONDITIONS n ° i includes the value of the considered CLUs.
- the corresponding CONDITIONS block is equal to the value of these sources.
- the corresponding CLU matrix is an identity matrix.
- the elementary matrices [CLI] consist of sub-matrix types noted in the examples described below M and Q. These sub-matrices are calculated according to the fields of application (ultrasound, electromagnetism , acoustics, etc.) to respectively ensure the continuity of the magnitudes at the interfaces, for example the normal and tangential components of a vector quantity, or else a scalar quantity and a normal vector component.
- step E 1 Since the invention applies to very diverse physical domains, such as electromagnetism, magnetostatic, acoustics, eddy currents or thermal, it is first necessary (step E 1) to to choose by the user all the elementary characteristic functions corresponding to the considered field of application.
- the user then defines the properties of the environments in which the interactions to be modeled occur (step E 2). These properties include all relevant parameters of the media.
- the type of media properties will depend on the problem being studied. For example, for electrostatics, relative permittivity is an essential parameter.
- This step will be used to create Correspondence Indicators that specify the environments adjacent to each object.
- the objects are associated with two media according to the formulation: i ( ⁇ , ⁇ ) where i designates the object and ⁇ and ⁇ the media adjacent to i.
- Each object has its Correspondence Indicator, which makes it possible to determine which sources to take into account in the interactions, as explained below with reference to several simplified examples.
- step E 2 there are two possibilities: either go through steps E 3, E 4 and E 5, or go directly to step E 5.
- steps E3 and E4 only when it is a question of taking into account wave propagation phenomena.
- the user defines the frequency of the waves to be transmitted by the transmitting transducer.
- the program calculates (step E 4) the maximum distance between the point sources (that is to say their pitch, for a regular configuration of these sources), according to the characteristics of the interactions. This step is preferably less than half wavelength at the origin of these interactions.
- step E 5 the user defines closed volume objects (hereinafter referred to as OVF).
- OVF closed volume objects
- a closed volume object is such that its surface is necessarily closed (for example, a sphere, a pyramid or a cube).
- This surface is a boundary between the external environment and an environment internal to the object, of different properties. If the user does not wish to calculate quantities in this environment, he must then apply boundary conditions (determined by himself and called here CLU) on its surface.
- the volume of this object may tend towards zero (for example for flat electrodes of a capacitor).
- an open-volume object (here called OVO) is such that the surface of this object is open (in theory, this volume is closed to infinity).
- the problem to be solved contains only closed-volume objects, then there is no need to limit the workspace.
- the problem of studying the electric field in the space created by two charged spheres does not require a working space.
- the example, described below with reference to FIGS. 6 to 8, shows two unbounded systems in space.
- a medium is defined by its homogeneous physical properties.
- An environment is necessarily closed, either because it is contained in an FVO, or because it is bounded by one or more open-volume objects (here referred to as OVO) and by the boundaries delimiting the workspace.
- OVO open-volume objects
- To define a medium it is necessary to know its location and its physical properties. In the following description, the media are referenced by numbers.
- the surfaces of the objects define the interfaces between two environments. To define an object, it is necessary to know:
- the objects are referenced by letters.
- the creation of the objects corresponds to a discretization of the surfaces of these objects in sets of points P which are associated the sets of source points responsible for synthesizing the waves propagating in the system. These objects determine interfaces on which Limit Conditions (CL) are applied.
- CL Limit Conditions
- step E 5 the user either goes through a step E 6 defining OVO (if such objects are present) to arrive at step E 7, or goes directly to step E 7 on the other hand.
- the definition and / or the mesh of all or part of the system may be provided by a device or software external to the invention.
- Step E 7 consists in determining boundary conditions, which constitute one of the essential conditions for implementing the method of the invention. These boundary conditions are placed on an interface, which is, as specified above, a separating surface between two media of different properties. Another feature of the invention is to associate layers of point sources only to the surfaces of objects (whether OVO or OVF). The invention takes into account two types of boundary conditions:
- CLI intrinsic boundary conditions
- CLUs User-defined boundary conditions
- these CLUs will be satisfied by a single set of sources located on one side of the interface.
- boundary conditions reflect the presence of energy sources in the system.
- Such a CLU is, for example in a capacitor, the electrical scalar potential difference between the electrodes, which is often known as set by the user.
- This step E 8 begins first by determining the coupling matrices between interfaces and requires a so-called masking operation.
- Masking refers to the fact that some objects in the system may screen each other
- the global matrix is block-sequenced.
- a matrix containing the main information namely the relative position of the media and the objects and their properties, was created from the correspondence indicators. From this matrix, using tests to determine which objects are coupled and according to which types of CL, we write the global matrix in blocks so as to systematically take into account all the interactions. By classifying the source groups in order of interactions, the blocks will be placed mainly on the diagonal and around. With the Object-Medium Correspondence Indicators created in the previous step, it is easy to test whether two or more objects have a common medium.
- the matrices obtained will be written according to the formulation: This can be read as the interaction of the object j on the object i in their common medium ⁇ , the sources of j being in the medium ⁇ .
- the sources by Aj ⁇ ⁇ the sources associated with the object j, being in the medium ⁇ and radiating in the medium ⁇ .
- matrix M or Q because the boundary conditions on the surface are imposed by the user (for example electrodes fed with a determined voltage).
- step E9 the program resolves the matrix equation (step E9), this step essentially consisting of inverting the global matrix. Since some quantities (for example the capacitance of a capacitor) do not need to know the numerical values of the CLUs to be calculated, they can therefore be obtained directly after the inversion, which the program does directly (step E 10 ). In the case where such quantities do not exist, or if, in addition to such quantities, it is necessary to calculate other quantities the user intervenes to supply the program with the missing data (step EI 1).
- some quantities for example the capacitance of a capacitor
- the missing data for the calculation of the physical quantities necessary for the modeling of the interactions are requested from the user according to the types of boundary conditions which have been fixed at the step E 7 (for example the value of the constant potential on an electrode metallic).
- step E 12 the step of defining the observation domain to be displayed.
- the user has gone through step E 6 or not, his freedom of choice for the calculations and observations will be greater or less: if the work space has not been limited during this step (the system, seat of interactions, includes only OVF), there is no restriction on the viewing area. If not, it must define the workspace that will be automatically bounded.
- step E 13 the program begins with the mesh of the spaces given by the user. Then, it calculates the physical quantities intervening in the studied interactions, according to the same procedure as that used for the computation of the matrices of coupling, applied this time to the points of the space of visualization.
- FIGS. 2 to 8 illustrate the writing technique of the overall matrix according to the invention.
- FIGS. 2 and 3 show the case of a planar multilayer system 1, FIG. 2 being a very simplified perspective view, and FIG. 3 a sectional view along III-11I of FIG. 2.
- the system 1 shown in FIGS. 2 and 3 has the general shape of a rectangular parallelepiped whose six faces are boundaries of the working space.
- the interface between the Middle 1 and the Middle 2 is referenced e
- that between the Middle 2 and the Middle 3 is referenced b.
- Two OVFs a and d, in this case capacitor electrodes, are in contact with the lower and upper faces of the work space.
- Middle 2 has a first inclusion b, and a second inclusion c, which is in contact with the interface f.
- Middle 0 defines the outside of the workspace, which is not known and does not interest the user.
- the six different interfaces are referenced as follows: a: between the 0 and 1 b media: between the 2 and 3 c: between the 2 and 4 d: between the 5 and 0 e: between the 1 and 2 f: between Media 2 and 5
- the ULC relative to the electrode is written Pa (potential imposed by this electrode at the interface a), and the CLU imposed by the electrode d is written Pd (potential imposed by this electrode at the interface d) .
- Interface a Aai oi (on the electrode side a)
- Interface b Abi 2 (in medium 2) and Ab 2 3 (in medium 3)
- Interface c Ac 4 2 (in the middle 2) and AC 2 4 (in the middle 4)
- Interface d Ad 502 (on the electrode side d)
- Interface e Ae 2 1 (in medium 1) and Ae 1 2 (in medium 2)
- Interface f Af 5 2 (in medium 2) and Af 2 5 (in the middle 5)
- FIGS. 4 (simplified perspective view) and 5 (sectional view along V-V of FIG. 4) show an example of a closed system 2.
- the system 2 comprises a hollow tube 3 rectangular section, which is called here object a and which contains the medium 1 and the medium 2 (delimited by the object b). Electrodes 4 and 5 are applied to two opposite faces of the object a. In the cross-section of FIG. 5, the different point source layers are indicated: Aaj o on the outer face of the object a, Ab 2 - I and Abi 2 on each side of the interface between the media 1 and 2, respectively in the middle 1 and in the middle 2. It will be noted that at the point of application of the electrodes 4 and 5, the point sources are represented differently from the others, because they are primary sources which indicate the presence of active surfaces.
- the CL Pa vector therefore contains CLUs and CLIs.
- FIGS. 6 to 8 The last two examples are illustrated in FIGS. 6 to 8. These are two variants of micro-capacitors with variable or adjustable capacitance, as described respectively in: "V-shaped micromechanical tunable capacitor for RF applications", A Cruau et al, DTIP2004, Montreux and in "A high-Q tunable micromechanical capacitor with movable dielectric for RF applications", JB. Yoon et al, Intl. EDM 2000, San Francisco. Only their particularities relating to the application of the process of the invention will be set forth here.
- the first variant represented in FIG. 6, comprises four sets of three electrodes, each arranged in a plane. These four planes are all perpendicular to a common plane (not shown) and their intersections with this plane form two "V" arranged one slightly above the other.
- Each set of electrodes comprises three identical electrodes in the form of rectangular ribbon, whose main axes are parallel to each other and to said common plane.
- the electrodes of one of the two sets of the internal "V" are opposite those of the other set of this "V", and the same is true of the "V” electrode sets. outside.
- the electrode planes of the interior “V” are movable relative to those of the planes of the outer "V", but in each "V", the two sets of electrodes are fixed relative to each other.
- One object is called one of the sets of electrodes (the one on the left, as seen in the drawing), object b the two sets of electrodes of the "V” inside, and object c the other set of electrodes of the "V” outside.
- the boundary conditions on each of the electrodes are set by the respective potentials applied to these electrodes.
- Medium 1 is the one surrounding all the electrodes.
- P a , P b and P c the potential respectively applied to objects a, b and c.
- variable capacitor comprises two sets of electrodes each arranged in a plane and a set of dielectric bars, these three planes being parallel to one another.
- the set of dielectric bars of the inner plane is movable relative to the other two games, which are fixed relative to each other.
- the set of electrodes of one of the outer planes (the lower plane, as seen in the drawing) is called object a, that of the central plane object b and that of the other outer plane object c. All these elements are surrounded by the medium 1, and the dielectric contains the medium 2.
- the boundary conditions on the electrodes of the device of FIG. 7 are fixed by the electrical potentials which are applied to them.
- the potentials applied to the objects a and c are also respectively called P a and P c .
- FIG. 8 shows the different point sources on the different objects a, b and c.
- the coupling matrix is:
- the matrix Qac and in the latter case, the matrices Mac (Qac) and Mca (Qca), are partially hidden because the object b is partially screened between the objects a and c.
- the global matrix of this system it is possible to establish, for a system comprising several objects, the global matrix of this system, and thus to find all the components of the source vector J.
- the physical quantities representative of the interactions of the breast of the system can be calculated at any point of observation of the system. Note that for this application to electromagnetism, the number of unknown parameters of the sources per mesh is equal to the number of available equations (six, in this case).
- the system described could be a conductive object composed of stratified media 1, 2 and 5, placed facing an excitation field created by the object d, the objects b or c being able to represent a crack or an inclusion.
- a condition called gauge then allows to completely determine A.
- Coulomb gauges, or Lorentz gauges are commonly used. Other conditions on the potential coexist.
- FIG. 9 shows a system 6 comprising two transducers T and
- transducers whose active faces (vis-à-vis the system 6) are flat and parallel to the interfaces, but it is understood that the invention can also be implemented when these faces are not parallel to the interfaces.
- a source layer called a primary source
- an interface is modeled by two layers of sources, called secondary sources.
- FIG. 10 two layers of primary sources As and A) which model the two transducers S and T, and 2 (nl) secondary source layers relating to the (n-1 ) interfaces. These sources have been represented as small circles tangent to the corresponding interfaces.
- the total ultrasound field in each planar medium of the system 6 is obtained by superposition of the vectors generated by the two sets of sources situated just above and just below the medium considered.
- this total vector is calculated for the different media as follows:
- Middle C Summation of the fields generated by As and A / Middle C 2 : Summation of the fields generated by A / and A ⁇ Middle C 3 : Summation of the fields generated by A 2 and Aj
- Middle Q nI Summation of the fields generated by A ,, _ 2 and A H- i
- Middle C n Summation of the fields generated by A n . ⁇ and Aj-
- M and Q are coupling matrices of objects that interact in the same medium.
- the matrix Q characterizes a scalar quantity (an acoustic pressure in the case of ultrasonic waves, an electric scalar potential in electrostatic, or a magnetic scalar potential in magnetostatic), while the matrix M characterizes a vector quantity (speed of propagation of ultrasonic waves in the present example, electric field in electrostatic, or magnetic field in magnetostatic).
- Boundary and continuity conditions result in the following relationships:
- source layers are distributed on each side of the n + 1 interfaces Io, Ii, ... I n .
- These two source vector layers are denoted A 111 (for sources above interfaces) and A * m (for sources below interfaces), with 0 ⁇ m ⁇ n.
- the first (Io) and last (I n ) interfaces are adjacent to three different source layers.
- two sets of additional sources of As and A T are due to the fact that the active faces of the transducers are applied, by hypothesis, against these first and last interfaces.
- the interface Io coincides with the active face of the transducer S, we will consider here, for the presentation of the method of the invention, I 0 and S as two different interfaces.
- Io represents the passive region (devoid of energy source) and S the active region (a source of energy is present).
- S the active region (a source of energy is present).
- I n and T two different interfaces.
- Middle 2 Summation of the fields produced by A t and A 2 .
- Middle 3 Summation of the fields produced by A 2 and A 3 .
- the ultrasonic field is produced only by a single layer of sources:
- Middle 0 the field is produced by Ao sources.
- Middle n + ⁇ the field is produced by the sources A n .
- the field is produced by three layers of sources:
- Middle 1 Summation of the fields produced by A 0 , Aj and As.
- Middle n Summation of the fields produced by A n . ⁇ N and A and A T.
- the conditions at the following interfaces must first be satisfied.
- the ultrasonic pressure P and the speed V in the direction normally normal to the interfaces' planes must be continuous.
- the interfaces Io and I n are only adjacent to passive zones. Passing these passive interfaces, the pressure
- the number of unknown source vectors is also 2 (n + 1) NM J -M 2 -
- the global source vector ⁇ A ⁇ can be calculated.
- the transducers S and T are shown in contact with the interfaces I0 and I1 of an object 1 comprising n different media. These two interfaces delimit half-spaces marked by the 0 and n + 1 media.
- the difference between FIGS. 11 and 12 lies in the distribution of the source points in the vicinity of the interfaces Io and I n .
- Middle 1 Summation of the fields generated by A 0 and Ai.
- Middle 2 Summation of the fields generated by A 1 and A 2 .
- Middle 3 Summation of the fields generated by A 2 and A 3 .
- Middle n Summation of the fields generated by A n . ⁇ N and A.
- the field is generated by a single layer of sources: Middle 0: the field is generated by Ao sources. Middle n + ⁇ : the field is generated by the sources A ".
- Interfaces I 0 and I 11 are adjacent to both active and passive areas.
- the active faces of the transducers S and T are arranged in the active areas. These areas are respectively called I Q and I ⁇ .
- the passive zones that do not include the active faces of the transducers are called Io and I n (without exponents).
- braces ( ⁇ ) are replaced by zeros in the case of the 0 and n rank interfaces, whereas the terms in square brackets (()) are replaced by zeros when it comes to active zones (/ o s and P n ) and interfaces of ranks 0 and n.
- the velocity vector at the interface level in the medium 2 can be expressed in the form of a transmission coefficient matrix T such that: and in the medium 1 in the form of a matrix of reflection coefficients R such that:
- the value of the N point sources distributed on the surface of a transducer is calculated for example in the following manner. It is assumed here that the transducer is in contact with a fluid, but it is understood that one can easily transpose the presentation in the case where the transducer is in contact with a solid body.
- a m be the value of the mth point source.
- the pressure p m at a distance rm from this point source is given by:
- the component in an Xs direction of speed is:
- V s is a dimension vector (3N x 1) of the speed components in N points x and A s is the dimension vector (3N x 1) with the relative physical quantities 3N point sources.
- M ss is the matrix of dimensions (3N x 3N) linking the two vectors V s and A s .
- the transpose of the column vector V s is a ranked vector of dimension (1 x 3N).
- the elements of this vector are denoted v ", the index j being able to take the values 1, 2 or 3 and indicating the direction of the Cartesian component of the velocity of propagation of the waves.
- the exponent n can take any value between 1 and N and it corresponds to the point of the active face of the transducer for which the speed component is defined.
- the first index y of x can take the values 1, 2 or 3 and indicates which direction xi, x 2 or x 3 refers to x.
- the index m of x and r can take the values from 1 to 3N, depending on the point source considered, while the index n can take any value between 1 and N depending on the considered point of the active face of the transducer for which the velocity component is calculated.
- X 11 be one of the points x (located at the rear of the active face of the transducer).
- Pr is a dimension vector (M x 1) and has the pressure values in M observation points
- V 7 - is a dimension vector (3 M x 1) with all three components at any point. observation.
- the expression of V ⁇ is similar to that of Vs taken from the expression (18-K). The only difference is that its dimension is (3M x 1) and not (3N x 1).
- the matrix Mrs is the same as that of Mss in the equation (18-N) if the points of observation are identical to the points of the active face of the transducer for which the speed components are chosen so as to obtain the source vector As in the equation (18-P).
- the expression giving M ⁇ s is slightly different from that of Mss taken from the equation (18-N). Therefore, its dimension is (3 M x 3 N) as marked in equation (18-R) below:
- f ⁇ x " ⁇ m , r" t is identical to that in equation (18-O).
- the definitions of the indices / and m are the same as those in equation (18-O).
- the exponents n of the variables x and r can take any value between 1 and M, depending on the point considered. It should be noted that M ⁇ s is not a square matrix when M is different from N.
- the matrices Q 11 and M 11 are similar to the matrices Qn and Mn of equation (18), the only difference being that the matrices Q 11 and M 11 give the values of pressure and speed for the points at interfaces in the source distribution configuration Ai, while the matrices Qn and Mn give the pressure and velocity values for the points at the interfaces in the source distribution configuration Ai. It should be noted that these two matrices use the material properties of medium 1. Since the distribution pattern Ai is located just above the interface, while the distribution pattern Aj is located just below the interface, the normal velocity vectors relative to the points situated on either side of the interface are of the same amplitude and opposite directions. The pressures generated by these two layers of point sources are of the same amplitude. We then have:
- a and Ai are different although they are in the same location.
- A represents the equivalent point source distribution when the two media 1 and 2 are identical. In other words, if we replace medium 2 with medium 1, there is no real interface for A. However, for Aj * there is a real interface between media 1 and 2.
- the high-frequency reflection coefficient is similar to the plane-incidence reflection coefficient at normal incidence. On the other hand, it is slightly different in low frequencies. The reason is that the example presented here refers to spherical waves with a point source. The plane waves are modeled by superimposing a large number of point sources. The characteristics of the process of the invention will now be described with reference to a multilayer system and to interactions on the active face of the transducer. The geometrical problem considered here is identical to that relating to that of FIG. 10. The difference in the boundary conditions between the present case and that exposed at the beginning of the description, for which the velocity of the waves at the boundaries S and T is specified and denoted respectively V ⁇ 0 and Vro, whereas in the case described here, it is not specified.
- two point source layers are introduced at each interface, including S and T, as shown in FIG. 14.
- the total value of the sources below the S interface is equal to As + A 0 and Aj + A n above the interface T.
- the interface T is referenced Io and the interface T is referenced I n .
- the other interfaces are respectively referenced Ii to I n- ].
- Middle 2 Summation of the fields generated by A 1 and A 2 .
- Middle 3 Summation of the fields generated by A 2 and A 3 .
- Middle n-1 Summation of the fields generated by A n _ 2 and A n . ⁇ .
- Middle 1 Summation of the fields generated by A o, Ai and As.
- Middle n Summation of the fields generated by A n- j, A n and A T.
- the field is produced by a single layer of point sources: Inside the transducer S: The field is generated by Ao sources only .
- the field is generated by A n sources only.
- the value of the sources is determined from the conditions at the interfaces as follows.
- the dimension of the matrix M is 2 [( «- I) / V + Mi + M 2 ] x 2 [(w-1 W + M 1 + M 2 ]
- the number of unknown source vectors is also 2 [(wl) N + ⁇ / i + M 2 ]. As a result, the global source vector is uniquely obtained.
- each transducer may include active areas and passive areas.
- the surface velocity of Vos (or Vor) waves in the absence of any interface effects is specified.
- the surface velocity is zero when it is not influenced by any other source.
- a ⁇ o and An are point sources distributed over the transducers S and T when the interaction effect between them is ignored.
- a 50 and A ⁇ represent point source distributions for modeling the field generated by a transducer in a homogeneous medium in the absence of any other transducer or diffuser.
- the field A ⁇ o is reflected by the transducer T.
- This field is modeled by introducing a new layer of sources An distributed over the transducer T By equating the velocity vectors on the surface S and produced by the sources An and those due to the reflection of the field A ⁇ o, we obtain:
- the field generated by An is then reflected by the transducer S.
- This field is modeled by introducing a layer of sources A 52 distributed over the surface of the transducer S.
- the source vectors at the S and T transducers can be expressed as follows:
- Equation (44) can also be written:
- the method of the invention also applies to the case of diffusion, and in particular when the diffuser has any shape and is assumed to be homogeneous, as illustrated in FIG. 16.
- a diffuser may be an inclusion, in a layer of medium homogeneous (called medium 1) of a material (called medium 2).
- medium 1 a layer of medium homogeneous
- medium 2 a material
- a J first layer
- Ai second layer
- two layers of sources are taken into account point As and Ay modeling the S and T transducers.
- the total value of the field in the medium 1 is obtained by superimposing three layers of sources As, A T and Ai, whereas the field in the middle 2 is due to a single layer of sources A].
- This matrix equation is solved for the source vectors A ⁇ , A ⁇ , Ai and Ai.
- any point of the medium 1 (hereinafter referred to as the point in question) does not "see” all the point sources of the layer A], and it may not see all the point sources of the two transducers.
- this problem did not exist because the point in question was never in a masked area.
- FIG. 17 shows the point in question, marked P and located in the middle 1 and three layers of point sources (A ⁇ -, A-, and A 7 -) which contribute to the generation of the calculated field at point P.
- the radiation pattern relating to these three layers of point sources is such that that represented by small croissants in Figure 17.
- the point P then receives radiation from all point sources A ⁇ located between the points A and B as well as some of the point sources Aj located between the points C and D (on the medium 2) and all point sources A 7 - located between points E and F.
- the point P is excited only by the point sources A 7 - situated between the points E and G. (the radius joining P to G is tangent to the middle 2).
- FIG. 18 shows, between the transducers S and T, three media instead of the two media of FIG. 17.
- the medium 2 has the same inclusion as that of FIG. 17, and is extended by plane layers.
- This medium 2 is delimited by two interfaces II, 12 bypassing the inclusion and extending by plane surfaces parallel to the active faces of the transducers S and T and distant from each other by a distance D less than the maximum thickness of the inclusion. (maximum thickness of the inclusion, as determined on a perpendicular to the active faces of the transducers).
- the median plane of these flat surfaces of the interfaces II, 12 passes substantially through the center C of the middle 2 (the drawing shows its trace TR, which passes through C). It is called the medium 1 between the transducer S and the interface II, and the medium 3 between the transducer T and the interface 12.
- Middle 2 Summation of fields produced by A i and A ?
- Middle 3 Summation of fields produced by A 2 and A 7 -
- the example of Figure 18 becomes that shown in Figure 19.
- the same elements as those of Figure 18 are assigned the same references, and that, by hypothesis, the media 1 and 3 have the same properties and the distance D is zero, the medium 2 is entirely surrounded by the medium 1, and the two interfaces II and 12 are reduced to a virtual interface IO passing through the center C of the middle 2 and parallel to the active faces of T and S.
- the difference with respect to the case shown in Figure 16 lies in the determination of point sources.
- the four source layers are As, Ar, Al and A * i
- the layers A 1 and A 2 are identical to those of FIG. 16, while the layer A ] of Figure 16 is divided into two layers A] and A * 2 in Figure 19, and the layer A 1 of Figure 16 is divided into two layers A 2 and A 1 in Figure 19.
- Equation (51) is modified as follows to fit the case of Figure 19:
- Equation (53) is written in the following matrix form:
- equation (54) allows, unlike the resolution of equation (50), to solve the problems related to masked areas.
- Green 's theory is one of many computational techniques that can be used in the field of electromagnetism to solve equations for inhomogeneous waves [Ney], such as:
- A represents the magnetic potential vector
- J the current density
- Green's theory makes it possible to formulate the problem in a global manner, which is perfectly connected with the DPSM method of the invention, according to which, as specified above, the contributions of all the point sources radiating in the medium considered are summed up. .
- G (r, r ') being the function of Green or the elementary solution related to equation (F-4), and f the source in volume V.
- This solution can be considered as a spherical wave, decreasing according to 1 / R and having a singularity at the point R - »0.
- J (r ') is a current density, expressed in A / m 2 .
- J represents a finite sum of Ns point sources:
- J 1 is equivalent to the current density J (r ') which has been integrated in an elementary volume dv and which represents an elementary source vector (source in triplet, that is to say to three elementary Cartesian components, expressed in unit of current multiplied by a length, namely in A. m in the metric system).
- the DPSM method of the invention performs the summation of the contributions of all the point sources in a given network of observation points P. For example, if we have Np observation points, for each value of r. the potential magnetic vector consists of three Pae vector vectors example, Ax is given by the expression:
- R ⁇ is the distance between the excitation points / and the observation point /
- Jz 1 are not exactly located at the same point V 1 and the matrix Wxx, Wyy, and Wzz may be slightly different.
- B a ⁇ are block matrices in which the indices a and ⁇ respectively indicate the derivation axis and the axis of orientation of the elementary point source components.
- Equations (68) and (69) express the electric and magnetic fields according to the DPSM method of the invention.
- Green's theory has been adapted to the DPSM process to implement the concept of point sources. Boundary conditions are described below for imposing potential continuity across surfaces in the DPSM / Green formulation.
- one of the important characteristics of the DPSM formulation consists in placing on each side of an interface virtual point source networks (triplets of current sources in the case of electromagnetism) in order to synthesize the fields in the fields. media separated by this interface.
- the method of the invention is not limited to simple geometries, but can be applied to all kinds of surfaces and objects with multiple interfaces.
- the magnetic potential A x is calculated in the medium 1 by superimposing the effects of the sources J 5 and J n , whereas the magnetic potential A 2 in the medium 2 results solely from the sources J 12 .
- FIG. 21 shows a configuration for which there are sources J s (situated in the middle 1), an interface between the media with their associated layers of DPSM sources J M and J 42 (respectively oriented outwards to radiate in the middle 1 and inwards to radiate in the middle 2).
- the boundary conditions according to the process of the invention then become:
- J 9 Since J 9 is well known, one can also write: J s - I - J 8 , F being the identity matrix 3 ⁇ / S x 3 / Vs, which gives:
- the unknowns J 41 and J 42 are obtained by inverting equation (74b), in which the global matrix is always square, since W ] S is a dimension matrix.
- W n is a matrix of dimension 3Npx3Np
- W 12 is a matrix of dimension 3Npx3Np
- FIG. 20 shows a block of electrically conductive material ME1 considered to have a semi-infinite thickness above which, at a distance D from the upper face of the block, a conductive loop B 1 supplied by a voltage V (t ).
- the MEl block here has the shape of a rectangular parallelepiped and is considered semi-infinite insofar as its upper face is the one facing the loop B1 and its dimensions are greater than the wavelength of the voltage V (t).
- the loop B1 induces in the ME1 block a magnetic field CM.
- FIG. 21 diagrammatically shows the distribution of virtual point sources according to the method of the invention.
- a layer of primary excitation point sources Js in the plane of the loop B1 and secondary point source layers J AX and J 42 on each side of the air / block interface ME1, the sources J A2 being outside, air side, as specified above.
- the pitch of the sources Js is greater than that of the other sources because the loop B1 is a single turn.
- FIGS. 22 and 23 show, in plan view, the distribution of the real part of the sources J 41 and J ⁇ 2 respectively, in a plane parallel to the air / block interface ME 1, the block B 1 being made of aluminum and the frequency of the voltage V (t) being 1000Hz.
- FIGS. 24 and 25 show the distribution of these sources in the case where the loop B1 is inclined by 10 °, the distance between the center of the loop and the block being 50 mm, and the frequency of the voltage V (t) being 100 Hz.
- FIGS. 26 and 27 show, for the case corresponding to FIGS. 24 and 25, the real and imaginary parts of the magnetic field H at three different levels, namely in the plane of the upper face of the block ME1, in a plane located 5 mm from this face, inside the block and 10 mm from this face, still inside the block. All the coordinates of these figures 22 to 27 are graduated in millimeters.
- the invention can be implemented for anisotropic media, for which the elementary characteristic functions can also be obtained analytically.
- the DPSM sources are, in most cases, so-called “fictitious” sources, because they are arranged (on both sides of the interfaces in particular) for the sole purpose of synthesizing a physical quantity in a given region of the space, instead of the real influence of the objects that make up the system to be modeled.
- the potential v2 has a value determined by the capacitance coefficients of the system (see “Study, design, realization and test of a variable capacity MEMS RF for the treatment of frequencies between 0.5 and 20GHz" thesis
- each conductor is represented by a set of point sources Ai (internal sources radiating outwards - light gray squares) and a corresponding set of test points placed on the surface, where the boundary conditions apply. (black round dots)
- the objects A and B were respectively meshed with N1 and N2 test points, and respectively include N1 and N2 source points (A x and A 1 layers) which are here electrostatic charges.
- N1 and N2 source points A x and A 1 layers which are here electrostatic charges.
- P x be the vector containing the potential values at all the test points of the object A (P 1 is set by the user), and P 2 the vector containing the values of the potential at all the test points of the object.
- the DPSM method translates this problem by the following system of equations, given the properties listed above:
- Ai is the vector of internal charges of the conductor i
- Qij the coupling matrices DPSM
- Pi the potential vector of the conductor i.
- the unknown values are A ⁇ , A 2 and v 2 .
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)
Abstract
Description
Claims
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CA2634903A CA2634903C (fr) | 2005-12-23 | 2006-12-20 | Methode universelle de modelisation des interactions entre une onde au minimum et un objet au minimum, la surface de chaque objet definissant une interface entre deux milieux au minimum |
US12/158,757 US8126686B2 (en) | 2005-12-23 | 2006-12-20 | Universal method for modeling the interactions between at least one wave and at least one object, the surface of each object defining an interface between at least two media |
EP06841519A EP1964008A1 (fr) | 2005-12-23 | 2006-12-20 | Procede universel de modelisation des interactions entre au moins une onde et au moins un objet |
JP2008546450A JP5292574B2 (ja) | 2005-12-23 | 2006-12-20 | 少なくとも1個の波と、各物体の表面が少なくとも2個の媒質間の界面を画定している、少なくとも1個の物体との間の相互作用をモデル化する汎用的方法 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
FR0513219 | 2005-12-23 | ||
FR0513219A FR2895544B1 (fr) | 2005-12-23 | 2005-12-23 | Procede universel de modelisation des interactions entre au moins une onde et au moins un objet, la surface de chaque objet definissant une interface entre au moins deux milieux |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2007071735A1 true WO2007071735A1 (fr) | 2007-06-28 |
Family
ID=36570826
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/EP2006/070020 WO2007071735A1 (fr) | 2005-12-23 | 2006-12-20 | Procede universel de modelisation des interactions entre au moins une onde et au moins un objet |
Country Status (6)
Country | Link |
---|---|
US (1) | US8126686B2 (fr) |
EP (1) | EP1964008A1 (fr) |
JP (1) | JP5292574B2 (fr) |
CA (1) | CA2634903C (fr) |
FR (1) | FR2895544B1 (fr) |
WO (1) | WO2007071735A1 (fr) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2011091932A1 (fr) | 2010-01-26 | 2011-08-04 | Centre National De La Recherche Scientifique (Cnrs) | Procede d'estimation de defauts dans un objet et dispositif de mise en oeuvre |
WO2011092210A1 (fr) | 2010-01-26 | 2011-08-04 | Centre National De La Recherche Scientifique (Cnrs) | Procédé de modélisation des interactions d'une onde impulsive avec un milieu |
WO2018127538A1 (fr) | 2017-01-06 | 2018-07-12 | Centre National De La Recherche Scientifique | Procede de simulation de forces appliquees a une aile dans un ecoulement de fluide |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
FR3021405B1 (fr) * | 2014-05-23 | 2018-07-27 | Centre National De La Recherche Scientifique (Cnrs) | Dispositif et procede de mesure d'une grandeur physique d'un ecoulement de fluide |
WO2019090700A1 (fr) * | 2017-11-10 | 2019-05-16 | 盐城普兹机械有限公司 | Prise de force d'embrayage dynamique |
CN111352832B (zh) * | 2020-02-24 | 2024-06-07 | 腾讯科技(成都)有限公司 | 自动测试页面的方法、装置 |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH07147254A (ja) * | 1993-11-24 | 1995-06-06 | Hitachi Ltd | 高速三次元不純物分布計算方法 |
JP2003270280A (ja) * | 2002-03-18 | 2003-09-25 | Toshiba Corp | 電磁波解析方法および装置 |
FR2847051B1 (fr) | 2002-11-12 | 2005-02-04 | Centre Nat Rech Scient | Procede pour evaluer une grandeur physique representative d'une interaction entre une onde et un obstacle |
-
2005
- 2005-12-23 FR FR0513219A patent/FR2895544B1/fr not_active Expired - Fee Related
-
2006
- 2006-12-20 CA CA2634903A patent/CA2634903C/fr not_active Expired - Fee Related
- 2006-12-20 US US12/158,757 patent/US8126686B2/en not_active Expired - Fee Related
- 2006-12-20 WO PCT/EP2006/070020 patent/WO2007071735A1/fr active Application Filing
- 2006-12-20 EP EP06841519A patent/EP1964008A1/fr not_active Ceased
- 2006-12-20 JP JP2008546450A patent/JP5292574B2/ja not_active Expired - Fee Related
Non-Patent Citations (7)
Title |
---|
AHMAD RAIS ET AL: "Modeling of phased array transducers", JOURNAL OF THE ACOUSTICAL SOCIETY OF AMERICA, AIP / ACOUSTICAL SOCIETY OF AMERICA, MELVILLE, NY, US, vol. 117, no. 4, April 2005 (2005-04-01), pages 1762 - 1776, XP012072864, ISSN: 0001-4966 * |
BANERJEE SOURAV ET AL: "Ultrasonic field modeling in multilayered fluid structures using the distributed point source method technique", J APPL MECH TRANS ASME; JOURNAL OF APPLIED MECHANICS, TRANSACTIONS ASME JULY 2006, vol. 73, no. 4, July 2006 (2006-07-01), pages 598 - 609, XP009083281 * |
CRUAU AURELIE ET AL: "V-shaped micromechanical tunable capacitors for RF applications", MICROSYST TECHNOL; MICROSYSTEM TECHNOLOGIES; SYMPOSIUM ON DESIGN, TEST, INTEGRATION AND PACKAGING OF MEMS/MOEMS DECEMBER 2005, vol. 12, no. 1-2 SPEC. ISS., 25 October 2005 (2005-10-25), pages 15 - 20, XP002385849 * |
LIEBEAUX N ET AL: "The distributed sources method: a concept for open magnetic cores modelling", EUROPEAN PHYSICAL JOURNAL APPLIED PHYSICS, EDP SCIENCES, LES ULIS, FR, vol. 20, no. 2, November 2002 (2002-11-01), pages 145 - 150, XP008021970, ISSN: 1286-0042 * |
PLACKO D ET AL: "Theoretical computation of acoustic pressure generated by ultrasonic sensors in the presence of an interface", PROCEEDINGS OF THE SPIE - THE INTERNATIONAL SOCIETY FOR OPTICAL ENGINEERING SPIE-INT. SOC. OPT. ENG USA, vol. 4702, 2002, pages 157 - 168, XP008021971, ISSN: 0277-786X * |
PLACKO D ET AL: "Ultrasonic field computation in the presence of a scatterer of finite dimension", PROCEEDINGS OF THE SPIE - THE INTERNATIONAL SOCIETY FOR OPTICAL ENGINEERING SPIE-INT. SOC. OPT. ENG USA, vol. 5047, 2003, pages 169 - 179, XP002385609, ISSN: 0277-786X * |
TRIBIKRAM KUNDU, DOMINIQUE PLACKO: "Modeling of Ultrasonic Field by Distributed Point Source Method (Chapter 2 of Ultrasonic nondestructive evaluation: engineering and biological material characterization)", 2004, CRC PRESS, ISBN 0-8493-1462-3, XP002385842 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2011091932A1 (fr) | 2010-01-26 | 2011-08-04 | Centre National De La Recherche Scientifique (Cnrs) | Procede d'estimation de defauts dans un objet et dispositif de mise en oeuvre |
WO2011092210A1 (fr) | 2010-01-26 | 2011-08-04 | Centre National De La Recherche Scientifique (Cnrs) | Procédé de modélisation des interactions d'une onde impulsive avec un milieu |
US8818075B2 (en) | 2010-01-26 | 2014-08-26 | Centre National De La Recherche Scientifique (Cnrs) | Method for estimating defects in an object and device for implementing same |
WO2018127538A1 (fr) | 2017-01-06 | 2018-07-12 | Centre National De La Recherche Scientifique | Procede de simulation de forces appliquees a une aile dans un ecoulement de fluide |
Also Published As
Publication number | Publication date |
---|---|
FR2895544A1 (fr) | 2007-06-29 |
CA2634903C (fr) | 2016-04-26 |
CA2634903A1 (fr) | 2007-06-28 |
US20100010781A1 (en) | 2010-01-14 |
EP1964008A1 (fr) | 2008-09-03 |
US8126686B2 (en) | 2012-02-28 |
JP5292574B2 (ja) | 2013-09-18 |
FR2895544B1 (fr) | 2008-09-19 |
JP2009521031A (ja) | 2009-05-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Huthwaite | Evaluation of inversion approaches for guided wave thickness mapping | |
WO2007071735A1 (fr) | Procede universel de modelisation des interactions entre au moins une onde et au moins un objet | |
US10296683B2 (en) | Computer implemented apparatus and method for finite element modeling using hybrid absorbing element | |
Chaljub et al. | 3-D numerical simulations of earthquake ground motion in sedimentary basins: testing accuracy through stringent models | |
EP1600897B1 (fr) | Méthode pour générer un maillage hybride conforme en trois dimensions d'une formation hétérogène traversée par une ou plusieurs discontinuités géométriques dans le but de réaliser des simulations | |
EP2407934A2 (fr) | Méthode pour générer un maillage hexa-dominant d'un bassin géometriquement complexe | |
Hermann et al. | Non-conforming hybrid meshes for efficient 2-D wave propagation using the discontinuous Galerkin method | |
EP2275994A1 (fr) | Méthode pour générer un maillage hexa-dominant d'un milieu souterrain faillé | |
EP1107187A1 (fr) | Méthode pour générer un maillage sur une formation hétérogène traversée par une ou plusieurs discontinuités géométriques dans le but de réaliser des simulations | |
Givoli et al. | Time reversal with partial information for wave refocusing and scatterer identification | |
Assous et al. | Time-reversed absorbing condition: application to inverse problems | |
EP3295212B1 (fr) | Procédé et dispositif de sondage par propagation d'ondes | |
Mulder et al. | Time-stepping stability of continuous and discontinuous finite-element methods for 3-D wave propagation | |
Somala et al. | Finite-fault source inversion using adjoint methods in 3-D heterogeneous media | |
Zhao et al. | Direct numerical modeling of time-reversal acoustic subwavelength focusing | |
Meredith et al. | Visualization and Analysis‐Oriented Reconstruction of Material Interfaces | |
WO2018065684A1 (fr) | Procédé de fabrication d'un modèle géologique vectoriel | |
Zhao et al. | Computation of the radiation force exerted by the acoustic tweezers using pressure field measurements | |
Bonnasse-Gahot | High order discontinuous Galerkin methods for time-harmonic elastodynamics | |
Shi et al. | A time-domain finite element boundary integration method for ultrasonic nondestructive evaluation | |
EP1579353A1 (fr) | Procede pour evaluer une grandeur physique representative d une interaction entre une onde et un obstacle | |
Shen et al. | Pulsed-laser source characterization in laboratory seismic experiments | |
Assous et al. | Time reversal for obstacle location in elastodynamics from acoustic recording | |
Zharkov | Geometric properties of acoustic waves generated by a point source in the solar-like interior: effects of acoustic cutoff frequency | |
Cochran | Introduction to Sonar Transducer Design |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
REEP | Request for entry into the european phase |
Ref document number: 2006841519 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2006841519 Country of ref document: EP Ref document number: 2008546450 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2634903 Country of ref document: CA |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWP | Wipo information: published in national office |
Ref document number: 2006841519 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 12158757 Country of ref document: US |