Apparatus for estimating elastic constant

Abstract

To estimate an elastic constant of a desired part of an inspected member having acoustic anisotropy by using an actual member without using a sample or the like.SOLUTION: A second probe 4 having one element 3 that transmits a sound wave toward a first probe 2 having a plurality of transmitting and receiving elements 1 is disposed with an inspection target member 11 interposed therebetween, a start timing of the transmitting and receiving elements 1 of the first probe 2 is obtained on the basis of a reception situation of the sound wave from the second probe 4 input to the first probe 2, and an elastic constant of the inspection target member 11 is estimated as an estimated elastic constant CI on the basis of the obtained start timing.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an elastic constant estimation device that estimates the elastic constant of an inspection target member (inspected member) that has, for example, acoustic anisotropy. [Background technology]

[0002] Various ultrasonic flaw detection tests have been known as non-destructive testing methods for metals (for example, Patent Document 1). In ultrasonic flaw detection tests, ultrasonic waves are propagated to an inspected portion of a component to be inspected, and the size and shape of the flaw are estimated based on the strength and reflection range of the ultrasonic waves reflected from the flaw or other defect, thereby evaluating the condition of the flaw or other defect in the inspected portion of the component to be inspected.

[0003] For example, centrifugally cast stainless steel is used for piping in power generation equipment, and non-destructive testing using ultrasonic flaw detection is regularly carried out on the welds of the piping (centrifugally cast stainless steel).Centrifugal cast stainless steel contains coarse crystal grains, which may cause scattering, attenuation, and acoustic anisotropy of ultrasonic waves, resulting in bending of the ultrasonic path.

[0004] When conducting non-destructive testing on centrifugally cast stainless steel (a test object component with acoustic anisotropy), a phased array ultrasonic method is used, which uses a probe consisting of multiple elements and adjusts the timing of transmission and reception of each element to perform flaw detection. By using the phased array ultrasonic method, the accuracy of flaw detection for the test object component can be improved.

[0005] When performing non-destructive testing using phased array ultrasonics on a test object (e.g., a component with acoustic anisotropy: an acoustically anisotropic material), determining the elastic constants allows for more accurate testing. While the elastic constants can be determined experimentally using test specimens, it is currently not possible to experimentally determine the elastic constants for a desired portion of the test object (e.g., a portion corresponding to the location of the test object). For this reason, there is a demand for the establishment of a technology that can determine the elastic constants of any portion of the test object. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Application Laid-Open No. 2014-106130 Summary of the Invention [Problem to be solved by the invention]

[0007] The present invention has been made in view of the above circumstances, and aims to provide an elastic constant estimation device that can estimate the elastic constant of an inspected member (particularly a member having acoustic anisotropy) itself. [Means for solving the problem]

[0008] In order to achieve the above object, the elastic constant estimation device of the present invention according to claim 1 is characterized by comprising: a first probe having a plurality of elements for receiving sound waves; a second probe having one element for emitting sound waves toward the first probe across the test member; and control means for deriving, based on the reception status of sound waves input from the second probe to the first probe, the activation status of the transmitting and receiving elements of the first probe in a state in which sound waves are focused at a portion of the test member that is in contact with the second probe, and for estimating the elastic constant of a desired portion of the test member as an estimated elastic constant CI based on the derived activation status.

[0009] The present invention of claim 1 receives sound waves from a second probe (longitudinal wave vertical probe) with a first probe (multiple transmitting / receiving elements) and derives the activation status (activation timing of each element: measured focal law) from the first probe (multiple transmitting / receiving elements) in a state where the sound waves are focused on the second probe, and calculates the value of a convergence function (objective function) from the measured activation status (measured focal law) obtained based on the derived activation status and the theoretical activation status (theoretical focal law), and sets the elastic constant at a desired location of the inspected member as an estimated elastic constant CI based on the value of the convergence function.

[0010] This makes it possible to estimate the elastic constant of the test member itself (especially a member having acoustic anisotropy) without using a sample or the like.

[0011] The elastic constant estimation device of the present invention according to claim 2 is the elastic constant estimation device according to claim 1, wherein the control means determines the timing of the incident wave based on the reception state of the wave input from the second probe to the first probe, and based on the state in which the timing of the determined incident wave is inverted on the time axis (time reversal: time-reversed wave), derives activation timings of the plurality of elements of the first probe as actual activation timings for focusing sound waves on a portion of the test member in contact with the second probe, and calculates the timing of the incident wave based on the specifications (number of elements, layout, dimensions) of the first probe. The device is characterized by having an activation status derivation function that derives the theoretical activation timing of the plurality of elements estimated based on the assumed elastic constants (provisional elastic constants) of the inspected member as the theoretical activation timing, and an estimation function that compares the measured activation timing derived by the activation status derivation function with the theoretical activation timing, and sets the elastic constant (provisional elastic constant + Δ) of the inspected member at the location where the second probe comes into contact as the estimated elastic constant CI (provisional elastic constant + Δ) based on the assumed elastic constant of the inspected member taking into account the result of the comparison (provisional elastic constant + Δ with the provisional elastic constant corrected).

[0012] In the present invention of claim 2, the timing of the incident wave to the first probe, which is focused on the desired location (discontinuity such as a defect) by the sound wave emitted from the second probe (for example, a longitudinal wave vertical probe), is determined, and the measured activation timing ti (measured focal law) is determined based on the situation in which the timing of the incident wave is reversed on the time axis (time-reversed wave), and the value of the convergence function is determined according to the difference between the measured activation timing (measured focal law) and the theoretical activation timing Ti (theoretical focal law based on the situation of the theoretical inverted wave) based on the theoretical elastic constants (provisional elastic constants).

[0013] When the value of the convergence function converges to the desired state (when the measured activation timing and the theoretical activation timing are close to each other), the theoretical elastic constant (provisional elastic constant PC+Δ corrected during the convergence process) used to determine the theoretical activation timing (theoretical focal law) is used as the estimated elastic constant CI, which is the elastic constant at the desired location of the inspected component.

[0014] Specifically, when obtaining the theoretical activation timing (theoretical focal law), a provisional elastic constant (provisional elastic constant) is assumed, and when the measured activation timing and the theoretical activation timing are close to each other (when the value of the convergence function converges to the desired state), the elastic constant at which the convergence function converges (provisional elastic constant + Δ) is taken as the estimated elastic constant CI, which is the elastic constant of the inspected component.

[0015] Furthermore, the elastic constant estimation device of the present invention according to claim 3 is characterized in that in the elastic constant estimation device described in claim 2, the estimation function of the control means calculates at least the sum of squared residuals of the time index of the theoretical activation timing (the sum of squared residuals of the time index of the measured activation timing and the theoretical activation timing) multiple times, and when the sum of squared residuals converges to a desired state (when it becomes smallest: provisional elastic constant + Δ with the provisional elastic constant corrected), the converged elastic constant (provisional elastic constant + Δ) based on the assumed specifications of the inspected member at the time of convergence is set as the estimated elastic constant CI.

[0016] In the present invention according to claim 3, the sum of squared residuals is converged, that is, the provisional elastic constant assumed when determining the theoretical activation timing is corrected so that the sum of squared residuals between the measured activation timing and the theoretical activation timing is minimized (the value of the convergence function is determined), and when the sum of squared residuals is minimized (when the value of the convergence function converges to the desired state), the converged elastic constant (provisional elastic constant + Δ), which is the assumed elastic constant after correction, is taken as the estimated elastic constant CI.

[0017] Furthermore, the elastic constant estimation device of the present invention according to claim 4 is characterized in that in the elastic constant estimation device described in claim 3, the estimation function of the control means adjusts the assumed specifications (provisional elastic constants) of the inspected component each time a predetermined number of times from the start of control when calculating the residual sum of squares of the time index of the theoretical activation timing multiple times.

[0018] According to the present invention as set forth in claim 4, the elastic constant can be accurately estimated from the start of control.

[0019] Furthermore, the elastic constant estimation device of the present invention according to claim 5 is characterized in that in the elastic constant estimation device according to claim 3 (claim 3 or claim 4), the estimation function of the control means sets the assumed elastic constant of the inspected member as a provisional elastic constant, corrects the provisional elastic constant to obtain a provisional elastic constant +Δ in order to converge the residual sum of squares to a desired state, and sets the provisional elastic constant +Δ when the residual sum of squares has converged to the desired state as the converged elastic constant.

[0020] In the present invention according to claim 5, in order to converge the sum of squared residuals (to converge the value of the convergence function to a desired state), corrections are repeated to obtain a new provisional elastic constant, which is the provisional elastic constant +Δ, and when the sum of squared residuals converges to the desired state (when the value of the convergence function converges to the desired state and becomes a minimum), the new provisional elastic constant +Δ is taken as the estimated elastic constant CI.

[0021] Furthermore, the elastic constant estimation device of the present invention according to claim 6 is characterized in that, in the elastic constant estimation device according to any one of claims 1 to 5, the inspected member is centrifugally cast stainless steel having acoustic anisotropy, the first probe has a plurality of ultrasonic transmitting / receiving elements, and the first probe is used in an evaluation device that evaluates discontinuities in the centrifugally cast stainless steel by reflected waves based on the transmission and reception of ultrasonic waves by the plurality of ultrasonic transmitting / receiving elements.

[0022] In the present invention according to claim 6, the phased array ultrasonic flaw detection technology can be applied to estimate the elastic constant of a desired portion of centrifugally cast stainless steel. [Effects of the Invention]

[0023] The elastic constant estimation device of the present invention makes it possible to estimate the elastic constant of the test object itself, which is a member to be inspected (particularly a member having acoustic anisotropy). [Brief explanation of the drawings]

[0024] [Figure 1] 1 is a diagram illustrating the overall configuration of an elastic constant estimation device according to an embodiment of the present invention. [Figure 2] FIG. 10 is a conceptual diagram illustrating a situation in which activation timing is estimated by time reversal. [Figure 3] FIG. 2 is a block diagram of a control means. [Figure 4] 3 is a control flowchart. [Figure 5] 1 is a conceptual diagram of a wave in a state using estimated elastic constants. [Figure 6] FIG. 1 is a conceptual diagram of a wave motion without using an estimated elastic constant. DETAILED DESCRIPTION OF THE INVENTION

[0025] FIG. 1 shows the overall configuration of an elastic constant estimation device according to one embodiment of the present invention, FIG. 2 shows the concept of measuring the activation timing, FIG. 3 shows the block configuration of the control means, and FIG. 4 shows the processing flow.

[0026] 1, a first probe 2 having a plurality of (five in the illustrated example) transmitting and receiving elements 1 is provided, and a second probe 4 (longitudinal wave vertical probe) having one element 3 that transmits sound waves toward the first probe 2 is provided, sandwiching an acoustically anisotropic test piece 11 (centrifugally cast stainless steel). Ultrasonic waves are transmitted from the second probe 4 to the first probe 2, and a received signal is input to the first probe 2.

[0027] Although the specifications (number of elements, layout, dimensions) of the first probe 2 have been described as an example of a probe having a plurality of transmitting and receiving elements 1 and arranged parallel to the surface of the inspected member 11, the number of transmitting and receiving elements 1 can be any number greater than one, and it is also possible to use a probe arranged at a desired angle inclined to the surface of the inspected member 11. Furthermore, although centrifugally cast stainless steel having acoustic anisotropy has been described as an example of the inspected member 11, the inspected member 11 is not limited to a member having acoustic anisotropy.

[0028] The command status for transmitting sound waves from the second probe 4 is controlled by the control means 6, and the status of the received signals of each transmitting and receiving element 1 in the first probe 2 is input to the control means 6. Based on the reception status input to the first probe 2 (plurality of transmitting and receiving elements 1), the control means 6 derives the activation status of the transmitting and receiving elements 1 of the first probe 2 in a state where sound waves are focused at a portion (desired portion) of the inspected member 11 that the second probe 4 comes into contact with.

[0029] Then, in the control means 6, the elastic constant of the inspected member 11 is estimated as an estimated elastic constant CI based on the derived activation state of the transmitting and receiving element 1 of the first probe 2.

[0030] Specifically, as shown in Fig. 2(a), sound waves (vibrations) from a second probe 4 (longitudinal wave vertical probe) are received by a first probe 2 (plurality of transmitting and receiving elements 1) to obtain a received signal group 12. As shown in Fig. 2(b), the time reversal method is used to derive the measured activation timing ti (activation pulses Fa1, Fa2, Fa3, Fa4, Fa5 at each transmitting and receiving element 1: measured focal laws that are inverted waves) as the activation state of sound waves from the first probe 2 (plurality of transmitting and receiving elements 1) in a state where sound waves are focused on the second probe 4.

[0031] Furthermore, the elastic constant of the inspected member 11 is estimated as an estimated elastic constant CI based on the derived actual measurement activation timing ti (actual measurement focal law).

[0032] Although details will be described later, the value of the convergence function (objective function S) is found by the sum of squares of the residuals from the obtained measured activation timing ti (measured focal law) and the theoretical activation situation (theoretical activation timing Ti: theoretical focal law), and the elastic constant is estimated as the estimated elastic constant CI based on the convergence situation of the value of the objective function S (in a converged situation). In other words, the elastic constant in a state where the obtained value of the objective function S has converged is estimated as the estimated elastic constant CI.

[0033] This makes it possible to estimate the elastic constant of the test member itself (particularly, a member having acoustic anisotropy) without using a sample or the like.

[0034] As shown in FIG. 3, the control means 6 includes an activation status derivation function 7 and an estimation function 8.

[0035] The activation status derivation function 7 includes an actual activation timing derivation function 15 that derives the activation timing of the multiple transmitting and receiving elements 1 of the first probe 2 as the actual activation timing ti. The activation status derivation function 7 also includes a theoretical activation timing derivation function 16 that derives the theoretical activation timing of the multiple transmitting and receiving elements 1 as the theoretical activation timing Ti, which is estimated based on the specifications of the first probe 2 and the assumed elastic coefficient (provisional elastic constant PC) of the inspected member 11.

[0036] The estimation function 8 compares the measured activation timing ti derived by the activation status derivation function 7 with the theoretical activation timing Ti, and based on the assumed elastic coefficient of the inspected member 11 taking into account the results of the comparison (based on the provisional elastic constant PC+Δ obtained by correcting the provisional elastic constant PC), it sets the elastic constant (provisional elastic constant PC+Δ) of the inspected member 11 that the second probe 4 comes into contact with as the estimated elastic constant CI (provisional elastic constant PC+Δ).

[0037] That is, the estimation function 8 includes an objective function derivation function 17 that calculates the sum of squares of the residuals of the time indexes of the measured activation timing ti and the theoretical activation timing Ti using the following equation (1) to derive the value S of the objective function. S=Σ(ti-Ti) 2 ····(1)

[0038] In addition, when the sum of squared residuals converges to the desired state (when it becomes the minimum), it has a convergence elastic constant derivation function 18 that derives the convergence elastic constant (provisional elastic constant PC+Δ) based on the specifications of the provisional elastic constant PC of the inspected member 11 at the time of convergence.

[0039] Then, the converged elastic constant (provisional elastic constant PC+Δ) converged by the converged elastic constant derivation function 18 is set as the new provisional elastic constant PC by the replacement function 19, and the replaced provisional elastic constant PC is set as the estimated elastic constant CI by the setting function 20.

[0040] That is, the timing of the incident wave focused on the part of the inspected member 11 that the second probe 4 comes into contact with (discontinuous part such as a defect) is determined by the sound wave emitted from the second probe 4, and the measured activation timing ti (measured focal law) is determined based on the situation in which the timing of the incident wave is inverted on the time axis (time-reversed wave).The objective function value S is then determined according to the difference between the measured activation timing ti (measured focal law) and the theoretical activation timing Ti (theoretical focal law based on the situation of the theoretical inverted wave) based on the theoretical elastic constants (provisional elastic constants).

[0041] When the value S of the objective function converges to the desired state (when the measured activation timing ti and the theoretical activation timing Ti are approximate), the theoretical elastic constant (provisional elastic constant: provisional elastic constant PC+Δ corrected during the convergence process) used when determining the theoretical activation timing Ti (theoretical focal law) is taken as the estimated elastic constant CI, which is the elastic constant (at the desired location) of the inspected member 11.

[0042] Specifically, a provisional elastic constant is assumed when obtaining the theoretical activation timing Ti (theoretical focal law), and when the measured activation timing ti and the theoretical activation timing Ti are approximate (when the value S of the objective function converges to the desired state), the provisional elastic constant (provisional elastic constant PC+Δ) is taken as the estimated elastic constant CI, which is the elastic constant of the inspected member 11.

[0043] A specific flow of the elastic constant estimation process will be described with reference to FIG.

[0044] As shown in the figure, when the process starts, in step S1, a sound wave is emitted from the second probe 4 (see Figure 2), and based on the reception status input to the first probe 2 (see Figure 2), an inverted wave is derived (set) in the state in which the sound wave is reflected at the part (desired part) of the inspected member 11 that the second probe 4 (see Figure 2) comes into contact with.

[0045] Then, in step S2, the measured activation timing ti (activation pulses Fa1, Fa2, Fa3, Fa4, Fa5 at each transmitting / receiving element 1: see Figure 2: measured focal law which is an inverted wave) is derived (set) as the activation state of the sound waves from the first probe 2 (see Figure 2) when the sound waves are focused on the second probe 4 (see Figure 2).

[0046] Meanwhile, in step S3, a provisional elastic constant PC is set based on the assumed specifications of the test member 11 (see Figure 2), and in step S4, the theoretical activation timing of the multiple transmitting / receiving elements 1 (see Figure 2) is derived (set) as the theoretical activation timing Ti based on the provisional elastic constant PC.

[0047] In step S5, the value of the convergence function (objective function value S) is derived (obtained) from the actual measured activation timing ti and the theoretical activation timing Ti by the sum of squared residuals. In step S6, it is determined whether the objective function value S has been derived multiple times (a predetermined number of times), and if the objective function value S has not been derived multiple times (less than the predetermined number of times), the assumed specifications (provisional elastic constants) of the inspected member 11 (see FIG. 2) are adjusted in step S7, for example, so that the provisional elastic constant PC is set at another point within a predetermined range, and the process proceeds to step S4 (A).

[0048] That is, the assumed specifications (provisional elastic constants PC) of the inspected member 11 can be adjusted each time for a predetermined number of times from the start of control.

[0049] If it is determined in step S6 that the objective function value S has been derived multiple times, it is determined in step S8 whether the objective function value S has converged. In other words, it is determined whether the residual sum of squares between the actual activation timing ti and the theoretical activation timing Ti has become the smallest.

[0050] If it is determined in step S8 that the objective function value S has converged, then in step S10 the converged provisional elastic constant PC is set as the estimated elastic constant CI, and the process ends. If it is determined in step S8 that the objective function value S has not converged, then in step S9 the provisional elastic constant PC is corrected (the provisional elastic constant PC+Δ is set as the new provisional elastic constant PC), and the process proceeds to step S4 (A). Then, based on the new provisional elastic constant PC, the theoretical activation timings of the multiple transmitting and receiving elements 1 (see FIG. 2) are derived (set) as the theoretical activation timings Ti.

[0051] In other words, when the sum of squared residuals converges to the desired state and the value of the objective function S converges (when it becomes the smallest: when the provisional elastic constant PC is corrected to the provisional elastic constant PC+Δ), the elastic constant at convergence (provisional elastic constant PC+Δ), which is the assumed elastic constant after correction, is used as the estimated elastic constant CI (replaced with the new provisional elastic constant PC).

[0052] When the estimated elastic constant CI is obtained using the above-mentioned elastic constant estimation device, and the elastic constant of the test portion 25 of the test member 11 itself (the portion that the second probe 4 comes into contact with: see Figure 2) is determined and non-destructive testing is performed using the phased array ultrasonic method, ultrasonic waves can be directed toward the test portion 25 of the test member 11, as shown in Figure 5.

[0053] If non-destructive testing is performed using the phased array ultrasonic method without calculating the estimated elastic constant CI and using a sample of the same material, i.e., using an elastic constant other than that of the test member 11 itself, ultrasonic waves will be directed to a position that is different from the test portion 25, which is the desired portion of the test member 11, as shown in Figure 6.

[0054] The elastic constant estimation device of the above-described embodiment can estimate the elastic constant of a desired portion of an acoustically anisotropic test component using the test component itself, without using a sample, etc. Therefore, it becomes possible to direct ultrasonic waves to a desired portion of the test component, and to perform highly accurate flaw detection of an acoustically anisotropic test component using a phased array ultrasonic method.

[0055] The elastic constant estimation device of the present invention makes it possible to estimate the elastic constant of a portion of a test member that corresponds to the test object itself. [Industrial Applicability]

[0056] The present invention can be used in the industrial field of elastic constant estimation devices that estimate the elastic constant of a test member. [Explanation of symbols]

[0057] 1 Transmitting and receiving element 2 1st probe 3 elements 4 2nd probe 6. Control Measures 7. Startup status derivation function 8 Estimation Function 11. Inspected component 12 Received signal group 15 Actual measurement start timing derivation function 16 Theoretical startup timing derivation function 25 Area to be inspected

Claims

1. a first probe having a plurality of transmitting and receiving elements for receiving sound waves; a second probe having one element that sandwiches a desired portion of the test object and emits a sound wave toward the first probe; a control means for deriving an activation state of the transmitting / receiving elements of the first probe in a state where sound waves are focused at a portion of the test member that is in contact with the second probe, based on a reception state input from the second probe to the first probe, and for estimating an elastic constant of a desired portion of the test member as an estimated elastic constant CI, based on the derived activation state. An apparatus for estimating elastic constants, comprising:

2. 2. The elastic constant estimation device according to claim 1, The control means an activation status derivation function that determines the timing of an incident wave based on the reception status of the wave input from the second probe to the first probe, and derives activation timings of the plurality of elements of the first probe as actual activation timings based on the status obtained by inverting the determined timing of the incident wave on the time axis, so that sound waves are focused on a portion of the test member that is in contact with the second probe, and also derives theoretical activation timings of the plurality of elements estimated based on the specifications of the first probe and the assumed elastic constant of the test member as theoretical activation timings; an estimation function of comparing the actual activation timing derived by the activation status derivation function with the theoretical activation timing, and setting the elastic constant of the test member at the portion contacted by the second probe as an estimated elastic constant CI based on an assumed elastic constant of the test member taking into account the result of the comparison; An apparatus for estimating elastic constants, comprising:

3. 3. The elastic constant estimation device according to claim 2, The estimation function of the control means is At least, the residual sum of squares of the time index of the theoretical activation timing is calculated a plurality of times, and when the residual sum of squares converges to a desired state, the converged elastic constant based on the assumed specifications of the inspected member at the time of convergence is set as the estimated elastic constant CI. An elastic constant estimation device characterized by:

4. 4. The elastic constant estimation device according to claim 3, The estimation function of the control means is When the sum of squares of the residuals of the time index of the theoretical start timing is calculated a plurality of times, the assumed specifications of the test member are adjusted each time for a predetermined number of times from the start of control. An elastic constant estimation device characterized by:

5. 4. The elastic constant estimation device according to claim 3, The estimation function of the control means is The assumed elastic constant of the known test member is set as the provisional elastic constant, and in order to converge the residual sum of squares to a desired state, the provisional elastic constant +Δ is obtained by correcting the provisional elastic constant, and the provisional elastic constant +Δ when the residual sum of squares has converged to the desired state is set as the converged elastic constant. An elastic constant estimation device characterized by:

6. 6. The elastic constant estimation device according to claim 1, the test member is a centrifugally cast stainless steel having acoustic anisotropy, The first probe includes: A plurality of ultrasonic transmitting and receiving elements are included, The first probe includes: The ultrasonic transducer is used in an evaluation device that evaluates discontinuities in the centrifugally cast stainless steel by reflected waves based on the transmission and reception of ultrasonic waves by the ultrasonic transducer elements. An elastic constant estimation device characterized by:

Images