A method for detecting cavities in structures using elastic waves

Abstract

The present invention relates to a method for detecting cavities in a structure using elastic waves, and includes the steps of: installing a plurality of acceleration sensors capable of measuring acceleration in three-axis directions in the structure; applying a plurality of three-dimensional loads having elastic waves around the acceleration sensors and measuring the three-dimensional accelerations with the acceleration sensors; generating two-dimensional distributed load data by superimposing the three-dimensional loads applied to each acceleration sensor; calculating superimposed acceleration data by superimposing the measured three-dimensional accelerations; and applying the two-dimensional distributed load data and the superimposed acceleration data to a two-dimensional elastic wave inverse analysis algorithm.

Description

[Technical Field]

[0001] The present invention relates to a method for detecting cavities in a structure using elastic waves. [Background technology]

[0002] Although effective non-destructive testing techniques have been developed for analyzing defects in the concrete walls of nuclear power plants, such as impact echo, electromagnetic wave detection, and ultrasonic techniques, there are currently no standards for non-destructive evaluation of nuclear power plant structures.

[0003] Non-destructive evaluation techniques are mainly applied to assess shallow spalling damage in bridge top plates and road pavements, but there have been few cases where they have been successfully used to assess the integrity of nuclear power plant containment building walls that are more than 1m thick.

[0004] Representative elastic wave-based nondestructive testing methods include impact echo, impact response, surface wave spectroscopy, etc. However, because these methods are based on the principle of travel-time tomography, which uses the wave arrival time to estimate information about the interior of a structure, they have the problem of only being able to provide limited information about the continuous material state of the target structure.

[0005] For the sound operation of nuclear power plant structures, it is necessary to develop a highly reliable defect diagnosis technique that can accurately diagnose cavity problems in containment buildings. Summary of the Invention [Problem to be solved by the invention]

[0006] SUMMARY OF THE INVENTION It is therefore an object of the present invention to provide a method for detecting cavities in a structure using elastic waves. [Means for solving the problem]

[0007] The object of the present invention is achieved by a method for detecting cavities in a structure using elastic waves, comprising the steps of: installing a plurality of acceleration sensors capable of measuring acceleration in three-axis directions in the structure; applying a plurality of three-dimensional loads having elastic waves around the acceleration sensors and measuring the three-dimensional accelerations with the acceleration sensors; generating two-dimensional distributed load data by superimposing the three-dimensional loads applied to each acceleration sensor; calculating superimposed acceleration data by superimposing the measured three-dimensional accelerations; and applying the two-dimensional distributed load data and the superimposed acceleration data to a two-dimensional elastic wave inverse analysis algorithm.

[0008] The structure may include a containment building for a nuclear power plant.

[0009] The plurality of acceleration sensors may be arranged at intervals in the height direction of the containment building, and the load may be applied to both sides of each acceleration sensor.

[0010] The load is applied using an impact hammer, and the i-th distributed load L i is calculated using the following formula:

[0011]

number

[0012] where A is the cross-sectional area to which the load is applied, which is calculated by multiplying the diameter of the impact hammer tip by the length of the applied load distribution, and I ij is the load data in the jth impact column of the ith acceleration sensor row, and N L is the number of loads at the i-th acceleration sensor position.

[0013] In the superimposed acceleration data, the superimposed acceleration m at the kth response measurement position k is calculated using the following formula:

[0014]

number

[0015] where a ij k is the response measured by the kth accelerometer when a load is applied to the jth impact column of the ith accelerometer row, and N s is the number of accelerometers, and N L is the number of loads at the i-th acceleration sensor position. [Effects of the Invention]

[0016] According to the present invention, there is provided a method for detecting cavities in a structure using elastic waves. [Brief explanation of the drawings]

[0017] [Figure 1] 1 is a flowchart of a cavity detection method according to an embodiment of the present invention.

[0018] [Figure 2] 1 shows the arrangement of a load and an acceleration sensor in a cavity exploration method according to an embodiment of the present invention.

[0019] [Figure 3] 1 illustrates a distribution load simulation in a cavity detection method according to an embodiment of the present invention.

[0020] [Figure 4] 10 shows a distributed load acting perpendicularly to an analysis region in a cavity exploration method according to an embodiment of the present invention.

[0021] [Figure 5] 4 shows superimposed acceleration in a cavity exploration method according to an embodiment of the present invention.

[0022] [Figure 6] 10A and 10B are diagrams for explaining measurement and superposition of elastic wave and acceleration data.

[0023] [Figure 7] FIG. 1 is a conceptual diagram of elastic wave full waveform inverse analysis. DETAILED DESCRIPTION OF THE INVENTION

[0024] The present invention will now be described in more detail with reference to the drawings.

[0025] The accompanying drawings are merely examples shown to more specifically explain the technical concept of the present invention, and the concept of the present invention is not limited to the accompanying drawings. Also, in the accompanying drawings, sizes and intervals may be exaggerated to explain the relationship between each component.

[0026] In this invention, elastic waves are applied to the surface of the structure to be diagnosed, and the acceleration response that passes through, is reflected, and is refracted in the wall is measured. The time history of the load and the acceleration response waveform data are used to perform full waveform inversion of the elastic waves to detect cavities inside the wall.

[0027] In the following description, a containment building of a nuclear power plant is used as an example of a structure, but the present invention can also be applied to general architectural structures, concrete structures, reinforced concrete structures, and the like.

[0028] Furthermore, the arrangement of the acceleration sensor and the load in the following description is not limited to this, and can be modified into various forms.

[0029] The cavity detection method according to the present invention will be described with reference to FIGS.

[0030] First, a plurality of acceleration sensors capable of measuring acceleration in three axial directions are installed on a structure (S100).

[0031] As shown in Figure 2, 15 acceleration sensors may be attached at regular intervals along the height of the wall where precise cavity diagnosis is required.

[0032] An acceleration sensor is a device that outputs an electrical signal proportional to acceleration within a certain frequency range. The acceleration sensor attached to the surface of the containment building wall measures instantaneous acceleration response in three axial directions. For example, the acceleration sensor may be PCB's 356A16.

[0033] Next, a plurality of three-dimensional loads having elastic waves are applied around the acceleration sensor, and the three-dimensional acceleration is measured by the acceleration sensor (S200).

[0034] The load may be applied to each acceleration sensor three times in each of the two lateral directions, a total of six times, as shown in Figure 2. The intervals between the loads may be constant.

[0035] In this way, there are three impact loads on each side along the corresponding row of each acceleration sensor, for a total of six impact loads, and the impact loads are applied repeatedly to all acceleration sensor rows, for a total of 90 impact loads.

[0036] The loads may be applied one by one in sequence, multiple at the same time, or all at the same time.

[0037] The load may be applied using an impact hammer, which is capable of applying elastic waves over a wide frequency range and can be used in a variety of directions and positions. The impact hammer may be, but is not limited to, PCB's 086D20.

[0038] Next, to apply the 3D load data applied to the wall surface of the nuclear power plant containment building and the measured acceleration response to the 2D elastic wave inverse analysis algorithm, the load time history and the acceleration response time history are superimposed, respectively. Here, the order in which the load time history and the acceleration response time history are superimposed is not limited.

[0039] This will be explained in more detail below.

[0040] The three-dimensional loads applied to each acceleration sensor are superimposed to generate two-dimensional load distribution data (S300). This step is a step of superimposing surface loads.

[0041] As shown in Figures 3 and 4, the two-dimensional analysis domain is assumed to have a constant elastic modulus and line load in the direction perpendicular to the analysis surface.

[0042] The load data applied by the impact hammer is extracted by a data collection device.

[0043] The six loads applied to each acceleration sensor in the row direction are summed up, divided by the cross-sectional area A to which the loads are applied, and superimposed as a distributed load Li.

[0044]

number

[0045] where A is the cross-sectional area to which the load is applied, which is calculated by multiplying the diameter of the impact hammer tip by the length of the applied load distribution, and I ij is the load data in the jth impact column of the ith acceleration sensor row, and N L is the number of loads at the i-th acceleration sensor position.

[0046] Examples of surface load superposition are as follows:

[0047]

number

[0048] The superimposed line load L1 applied to the first sensor row is a total of six (N L ) and divide by the cross-sectional area to which the load is applied.

[0049] A total of 15 loads, each of which is a sum of the loads for each row of acceleration sensors, are applied as distributed loads in the vertical direction of the finite element analysis model.

[0050] Next, the measured three-dimensional accelerations are superimposed to calculate superimposed acceleration data (S400).

[0051] In order to apply the 3D response measured in situ to the 2D elastic wave inversion algorithm, the measured acceleration response must be superimposed. The superimposed acceleration response m at the kth response measurement location is k is as follows:

[0052]

number

[0053] where a ij k is the response measured by the kth accelerometer when a load is applied to the jth impact column of the ith accelerometer row, and N s is the number of accelerometers, and N L is the number of loads at the i-th acceleration sensor position.

[0054] An example of the superposition of the measured responses is as follows:

[0055]

number

[0056] 15 (N s ) sensor rows, 6 (N L A total of 90 impact loads are applied to the load sequence. The superimposed response m1 at the position of sensor 1 is calculated by summing all 90 responses measured by sensor 1 to the 90 impact loads.

[0057] The above-mentioned superposition of the impact load and acceleration data will be described in more detail below.

[0058] FIG. 6 is a diagram for explaining the superposition of impact load and acceleration data in the present invention.

[0059] The example impact load and acceleration data superposition shown in FIG. 6 can be summarized as follows in Table 1:

[0060]

[0061] [Table 1]

[0062] - Superimposed load

[0063] The experiment is conducted on a three-dimensional structure, but the inverse analysis algorithm is based on a two-dimensional plane deformation rate problem, so the three-dimensional load data must be converted to two dimensions.

[0064] In the inverse analysis algorithm, loads are applied simultaneously at multiple positions on the surface of the analysis domain. Since it is not possible to simultaneously apply loads during the experiment, multiple loads are applied individually and then superimposed.

[0065] The experiment involves applying a three-dimensional impact load and then converting the impact load into a two-dimensional distributed load according to the process described above.

[0066] The superimposed distributed load is used to simulate a two-dimensional uniformly distributed load applied in the vertical direction of the analysis domain.

[0067] -Superimposition of acceleration response

[0068] Through experiments, acceleration responses to individual impact loads can be obtained as shown in Table 2, A.

[0069] A two-dimensional distributed load composed of the impact loads applied in the experiment is simultaneously applied to the analysis domain.

[0070] To perform the inverse analysis, the response at the acceleration sensor position when distributed loads are simultaneously applied as shown in Table 2, B, is required.

[0071] When a distributed load is applied simultaneously, the calculated response at a particular acceleration sensor location is equal to the sum of all acceleration responses measured by that acceleration sensor in the experiment.

[0072] The superimposed acceleration time history is compared with the acceleration response calculated by elastic wave propagation analysis, and an inverse analysis is performed.

[0073]

[0074] [Table 2]

[0075] Finally, the distributed load data and superimposed acceleration data are applied to a two-dimensional elastic wave inverse analysis algorithm (S500), which calculates the distribution of material properties in the structure and enables the diagnosis of cavities.

[0076] Full waveform seismic inversion is a technique that derives unknown specific values ​​of a system by analyzing the entire measured waveform that has passed through, reflected, and refracted from a structure using an inversion algorithm.

[0077] In order to explore cavities in the containment building of a nuclear power plant using elastic wave measurement responses, we assume that the walls of the containment building of a nuclear power plant are in a two-dimensional elastic domain and perform full waveform inversion of elastic waves.

[0078] This is done through an optimization process in which elastic waves are generated on the surface of the containment building wall, the elastic waves that move and reflect inside the wall are measured, and the difference between the measured response waveform and the calculated response waveform is minimized to estimate the distribution of material properties in the containment building wall.

[0079] In order to diagnose cavities inside the walls of a nuclear power plant containment building, it is necessary to carry out "1. Time domain elastic wave analysis for an analysis area based on the walls of the nuclear power plant containment building" and "2. Elastic wave excitation and response measurement for the walls of the nuclear power plant containment building."

[0080] The present invention relates to "2. Elastic wave excitation and response measurement of the wall of the nuclear power plant containment building" and a method for applying the measured data to an inverse analysis algorithm.

[0081] Algorithm for elastic wave full waveform inversion technique

[0082] The flow of the inverse analysis will be described below with reference to FIG.

[0083] (1) Construction of a 2D elastic wave finite element analysis model of the containment building wall base to be diagnosed

[0084] (2) The analysis domain is limited to a finite area, and a wave-absorbing boundary layer (perfectly matched layer) is introduced to absorb the reflected waves generated at the boundary surface of the domain.

[0085] (3) Setting initial assumptions for the mechanical material property distribution of the finite element analysis model

[0086] (4) Elastic waves are applied to the containment building wall to be diagnosed using an impact ammer, and the time history of the load and the acceleration response on the wall surface are extracted using a data collection device.

[0087] (5) Applying the time history of surface loads applied to the target containment building walls to the finite element analysis model to calculate the behavior of elastic waves transmitted, reflected, and refracted by the walls.

[0088] (6) The responses calculated by elastic wave finite element analysis and the responses measured on the target containment building walls are reflected in the elastic wave inverse analysis algorithm to update the material properties of the finite calculation domain.

[0089] (7) Repeat steps (5) and (6) to minimize the objective function that indicates the difference between the calculated response and the measured response, and determine the optimal solution for the distribution of material properties inside the containment building wall to be diagnosed.

[0090] The following describes cavity diagnosis using full waveform inversion of elastic waves.

[0091] Inverse analysis can derive the optimal solution for the distribution of material properties (Lame constants) of the walls of the containment building of a nuclear power plant.

[0092] If the reconstructed material property values ​​are different from the material property values ​​of the concrete used in the containment building and are evaluated to be significantly lower than the surrounding material property values, the corresponding area can be evaluated as a cavity.

[0093] The reconstructed material property distribution can be used to generate B-Scan images of the containment building walls of a nuclear power plant. The generated tomographic images can identify the location and shape of cavities where material property values ​​are evaluated as low.

[0094] The present invention provides a data processing technique that enables measured elastic wave data to be used for two-dimensional inverse analysis based on plane deformation rates.

[0095] By applying 3D measured data to 2D full waveform inversion analysis, cavities can be detected in specific sections of concrete walls.

[0096] To evaluate the condition inside the thick walls of a nuclear power plant containment building, elastic waves, which transmit more energy than ultrasonic waves or electromagnetic waves, can be used to obtain more reliable diagnostic results.

[0097] The above-described embodiments are merely examples for explaining the present invention, and the present invention is not limited thereto. A person skilled in the art to which the present invention pertains can implement the present invention by modifying it in various ways, and therefore the technical scope of protection of the present invention should be defined by the appended claims.

Claims

1. A method for detecting cavities in a structure using elastic waves, installing a plurality of acceleration sensors capable of measuring acceleration in three axes on the structure; applying a plurality of three-dimensional loads having elastic waves around the acceleration sensor and measuring three-dimensional acceleration with the acceleration sensor; generating two-dimensional load distribution data by superimposing the three-dimensional loads applied to each acceleration sensor; calculating superimposed acceleration data by superimposing the measured three-dimensional accelerations; and An exploration method comprising the step of applying the two-dimensional load distribution data and the superimposed acceleration data to a two-dimensional elastic wave inverse analysis algorithm.

2. The method of claim 1 , wherein the structure comprises a containment building of a nuclear power plant.

3. the plurality of acceleration sensors are spaced apart in the height direction of the containment building, 3. The method according to claim 2, wherein the load is applied to both sides of each of the acceleration sensors.

4. The load is applied using an impact hammer; The i-th distributed load L in the distributed load data ｉ The method of claim 3, wherein is calculated using the following formula: [Equation 1] where A is the cross-sectional area to which the load is applied, which is calculated by multiplying the diameter of the impact hammer tip by the length of the applied load distribution, and I ｉｊ is the load data in the jth impact column of the ith acceleration sensor row, and N Ｌ is the number of loads at the i-th acceleration sensor position.

5. In the superimposed acceleration data, the superimposed acceleration m ｋ The method of claim 4, wherein is calculated using the following formula: [Equation 2] Here, a ｉｊ ｋ is the response measured by the kth acceleration sensor when a load is applied to the jth impact column of the ith acceleration sensor row, and N ｓ is the number of acceleration sensors, and N Ｌ is the number of loads at the i-th acceleration sensor position.

Images