JP7571966B2 - Non-destructive testing system, non-destructive testing method and non-destructive testing program - Google Patents

Description

The present invention relates to a non-destructive testing system that measures changes in the length and cross-sectional area of objects to be measured, such as foundation piles of buildings, and inspects the presence or absence of damage or defects and the soundness of the objects, a non-destructive testing method that uses this non-destructive testing system, and a non-destructive testing program that causes a computer system to execute this non-destructive testing method.

Non-destructive inspection techniques that diagnose the soundness of concrete structures such as foundation piles are becoming increasingly important in terms of the safety of concrete structures that are subject to damage caused by earthquakes, fatigue cracks, and man-made problems. Conventionally, the surface of the object to be measured is struck with an impact hammer to transmit impact elastic waves inside the object, and the impact elastic waves reflected inside the object are captured by an accelerometer. The positions of cracks, enlarged diameter parts, pile tips, etc. that have occurred inside the object are determined from the propagation time and propagation speed of the reflected waves (see Patent Document 1).

The inspection method described in Patent Document 1 utilizes the phenomenon in which impact elastic waves propagating through foundation piles are reflected at boundaries with different impedances to diagnose pile length, changes in pile cross section, damage, defects, etc. In the case of the inspection method described in Patent Document 1, the waveform of the impact elastic wave differs depending on how the impact hammer strikes the object being measured, so hundreds of strikes per measurement point and extensive experience are required. The aim is to detect the tip, cracks, changes in shape, etc. of the object being measured from the waveform of the reflected wave of the impact elastic wave, but identifying defects such as cracks is determined from the object being measured, the surrounding conditions, and multiple measurement results, which requires extensive experience and knowledge.

Furthermore, in the case of long concrete structures such as foundation piles, the propagation speed of the impact elastic wave is required to determine the position of the tip of the concrete structure from the travel time of the impact elastic wave, but this differs depending on the material and restraint state of the object being measured. If the material and restraint state of the object being measured are unknown, the propagation speed of the impact elastic wave cannot be accurately determined, resulting in judgment errors and reduced accuracy.

JP 2014-224756 A

The present invention has been made to solve the above problems, and aims to provide a nondestructive inspection system capable of inspecting an object to be measured with high accuracy without requiring extensive experience or knowledge, a nondestructive inspection method using this nondestructive inspection system, and a nondestructive inspection program for causing a computer system to execute this nondestructive inspection method.

In order to achieve the above object, the present invention according to claim 1 comprises an automatic impact device to be provided on a part of an object to be measured, a detector to be provided on another part of the object to be measured and detect a reflected wave of an elastic wave generated by the automatic impact device, a waveform selector connected to the detector and for selecting a part of the waveform output by the detector, and an analysis device connected to the waveform selector and for selecting a feature amount for the signal output by the waveform selector by an empirical mode decomposition technique and for performing an analysis for inspecting the object to be measured using the feature amount, wherein the analysis device comprises a reflected wave feature extraction logic circuit that uses the empirical mode decomposition technique to extract features of healthy reflected waves that have propagated through a calibration sample output by the waveform selector and returned, and a reflected wave feature extraction logic circuit that uses the empirical mode decomposition technique to extract features of healthy reflected waves that have propagated through a calibration sample output by the waveform selector and returned, The gist of the non-destructive inspection system is that it includes: a feature selection logic circuit that selects a feature of a reflected wave that has propagated through an object and returned; a judgment logic circuit that judges defects in the object being measured using the features extracted by the reflected wave feature extraction logic circuit and the feature of the object being measured selected by the feature selection logic circuit; a reflected wave emphasis filter coefficient calculation circuit that converts the feature of the object being measured using design information of the object being measured and calculates a reflected wave emphasis filter coefficient of the object being measured from the converted feature; a time axis expansion/compression circuit that prepares a plurality of filters and expands or compresses the time axis of each of the plurality of filters; and a filter bank processing logic circuit that processes the plurality of filters in parallel using the reflected wave emphasis filter coefficients .

The invention described in claim 2 is also summarized as the non-destructive inspection system described in claim 1, characterized in that the non-destructive inspection system further comprises a design information storage device connected to the analysis device for storing design information data of a calibration sample .

According to an example of the present invention, it is possible to provide a non-destructive inspection system capable of inspecting an object to be measured with high accuracy without requiring extensive experience and knowledge, a non-destructive inspection method using this non-destructive inspection system, and a non-destructive inspection program for causing a computer system to execute this non-destructive inspection method.

1 is a diagram showing an overview of a non-destructive inspection system according to a first embodiment of the present invention. 2 is a schematic diagram illustrating the hardware resources constituting the analysis device of FIG. 1 as a logical structure. 2 is a schematic flowchart illustrating a processing procedure of a nondestructive inspection method according to the first embodiment. 1 is a schematic diagram illustrating an example of the structure of an automatic impact device used in a non-destructive inspection system according to a first embodiment. FIG. 10 is a schematic diagram illustrating another example of the structure of the automatic impact device used in the non-destructive inspection system according to the first embodiment. FIG. 11 is a schematic diagram illustrating yet another example of the structure of the automatic impact device used in the non-destructive inspection system according to the first embodiment. FIG. FIG. 2 is a circuit diagram illustrating an example of a waveform selector used in the nondestructive inspection system according to the first embodiment. 8(a) is a diagram showing the waveform of a signal input to the waveform selector shown in FIG. 7, and FIG. 8(b) is a diagram showing the waveform of a signal output from the waveform selector of FIG. FIG. 13 is a diagram for explaining three intrinsic mode functions (IMFs) when an input signal is a reflected wave from a healthy concrete structure and the input signal is decomposed into ten intrinsic mode functions (IMFs) by the Empirical Mode Decomposition method. FIG. 13 is a diagram for explaining three intrinsic mode functions IMF when the input signal is decomposed into ten intrinsic mode functions IMF by the empirical mode decomposition technique, in a case where the object to be measured is a concrete structure in which the presence of cracks, scratches, fissures, etc. is suspected. FIG. 11 is a diagram illustrating a feature amount included in the third intrinsic mode function IMF3 shown in FIG. 10. 3A to 3C are diagrams for explaining an outline of processing of a filter bank used in the nondestructive inspection method according to the first embodiment. FIG. 11 is a diagram for explaining the processing of a filter bank that performs filtering in parallel using a continuous wavelet technique in the nondestructive inspection method according to the first embodiment.

Below, a non-destructive inspection system according to a first embodiment of the present invention will be described by way of example with reference to the drawings. In the following description of the drawings, identical or similar parts are given identical or similar reference symbols. However, it should be noted that the drawings are schematic, and the relationship between thickness and planar dimensions, the ratio of sizes of each component, etc., differ from the actual ones. Therefore, specific thicknesses, dimensions, sizes, etc. should be determined more flexibly, taking into account the gist of the technical ideas that can be understood from the following description. In addition, it goes without saying that there are parts in which the dimensional relationships and ratios differ between the drawings.

The non-destructive inspection system according to the first embodiment shown below is an example of a method for realizing the technical idea of the present invention and an apparatus used in the method, and the technical idea of the present invention does not specify the material, shape, structure, arrangement, etc. of the components, or the procedure of the method, etc., as described below. In particular, the non-destructive inspection system according to the first embodiment below illustrates a case in which the object to be measured is a long, rod-shaped foundation pile, but the object to be measured in the non-destructive inspection system according to the first embodiment is not limited to foundation piles, and various concrete structures such as the inner or outer walls of buildings and bridge girders are objects to be measured. The object to be measured is not limited to these examples. The technical idea of the present invention is not limited to the contents described in the non-destructive inspection system according to the first embodiment, and various modifications can be made within the technical scope defined by the claims described in the claims.

First Embodiment
<About the device configuration>
As shown in Fig. 1, the nondestructive inspection system according to the first embodiment includes an automatic impact device 21a provided on a part of the object 11 to be measured, an elastic wave detector 31 provided on another part of the object 11 to be measured, a waveform selector 32 connected to the detector 31 and selecting a part of the waveform output by the detector 31, a waveform display device 34 connected to the waveform selector 32 and displaying the waveform output by the waveform selector 32, an analysis device 35 connected to the waveform display device 34 and performing an analysis for inspecting the soundness of the object 11 to be measured using a signal output by the waveform display device 34, and a waveform data storage device 36 connected to the waveform display device 34 and storing waveform data. The analysis device 35 selects a feature quantity for the signal output by the waveform selector 32 by a method of empirical mode decomposition, and performs an analysis for inspecting the soundness of the object 11 to be measured using the feature quantity. The automatic impact device 21a is connected to an impact control device 22 that controls the automatic impact operation performed by the automatic impact device 21a, and the impact control device 22 can control the command for the operation of the automatic impact device 21a using data from the analysis device 35.

The object to be measured 11 is an object of non-destructive testing performed by the non-destructive testing system according to the first embodiment, and is, for example, an elastic body such as a long, rod-shaped foundation pile. In the following description of the non-destructive testing system according to the first embodiment, it is assumed that the object to be measured 11 is an existing foundation pile, and that the portion of the pile head side of the object to be measured 11 is exposed above ground, and the other portion is buried underground. Note that in the first embodiment, the object to be measured 11 may be a hollow concrete pile. The automatic impact device 21a automatically impacts the pile head of the object to be measured 11, thereby generating elastic waves in the object to be measured 11.

The detector 31 is provided at the pile head of the object to be measured 11, and is an acoustic sensor or the like that detects the reflected wave that propagates through the object to be measured 11 and returns when the pile head is struck by the automatic striking device 21a. In this embodiment, the automatic striking device 21a and the detector 31 are provided on the same surface at the pile head of the object to be measured 11, but this is not limited thereto, and they may be provided on the side of the pile depending on the situation at the inspection site and the structure of the object to be measured, and the automatic striking device 21a and the detector 31 may be provided on different surfaces. In addition, considering the accuracy of the inspection, it is preferable to perform the measurement in the arrangement as in this embodiment. The waveform selector 32 is preferably a high pass filter (HPF) or the like that selects a part of the waveform output by the detector 31 so that the analysis device 35 estimates the pile length of the object to be measured 11 and performs the analysis required for non-destructive testing of the object to be measured 11 based on the output of the detector 31, but is not limited to an HPF and may be a band pass filter (BPF) or the like. Alternatively, if the automatic impact device 21a generates periodic elastic waves such as sine waves, a lock-in amplifier or the like can be used for the waveform selector. FIG. 7 shows a circuit showing the configuration of an HPF as an example of the waveform selector 32. If the waveform selector 32 is an active HPF using an operational amplifier or the like, a stabilized power supply 33 is connected to the waveform selector 32 as shown in FIG. 1. The waveform data storage device 36 is a secondary memory such as a semiconductor memory (RAM), magnetic disk, optical disk, magneto-optical disk, or magnetic tape for A/D converting waveform data (analog data) into digital data and storing the digital data. The waveform data storage device 36 of the first embodiment can store multiple waveform data.

The analysis device 35 is further connected to a waveform data storage device 36, a design information storage device 41, a defect data storage device 43, a feature amount storage device 44, an analysis data storage device 38, an output device 37, and an input device 42. The feature amount storage device 44 is a secondary memory for storing feature amount data (digital data). The design information storage device 41 is a secondary memory for storing design information data (digital data) of the calibration sample.
The defect data storage device 43 is a secondary memory for storing defect data (digital data). The program storage device 39 is a secondary memory such as a ROM for storing a computer software program that defines the processing procedure of the analysis device 35. The program storage device 39 can store a program recorded on a computer-readable recording medium such as a USB memory, an SD card, a flexible disk, an SSD, a CD-ROM, an MO disk, a cassette tape, or an open reel tape.

Alternatively, the computer software program required for the processing procedure of the analysis device 35 can be stored in the program storage device 39 via an information processing network such as the Internet. A part or all of the waveform data storage device 36, the design information storage device 41, the defect data storage device 43, and the feature storage device 44 can exist as part of a file in a storage device that is a common hardware resource, or each can exist as an individual hardware resource. The analysis data storage device 38 is a primary memory such as an SRAM for storing analysis data (digital data), a register, or a cache memory. The analysis data storage device 38 of the first embodiment can temporarily store and read out intermediate data sequentially output in the processing stage of the analysis data storage device 38 as necessary. The analysis device 35 performs calculations for wave theory analysis of the measured object 11 using the data stored in the waveform data storage device 36, the design information storage device 41, the defect data storage device 43, the feature storage device 44, and the analysis data storage device 38.

The output device 37 outputs the waveforms stored in the analysis data storage device 38 and the calculation results of the analysis device 35, and in the first embodiment, has a display that displays these on a screen. The output device 37 is not limited to a display. For example, it may be a data recorder that records the calculation results together with the measurement results, or a plotter or printer that prints the waveforms and the calculation results. The input device 42 is composed of a keyboard, a mouse, a light pen, or a reading device such as a flexible disk device or a USB memory. Using the input device 42, the person performing the analysis can specify input/output data, set operators required for the analysis, the order of the polynomial, frequency values, the value of the allowable error, the degree of error, etc. Furthermore, it is also possible to set analysis parameters such as the form of output data through the input device 42, and it is also possible to input instructions such as executing and canceling calculations.

The logical hardware resource configuration of the analysis device 35 constituting the computer system shown in FIG. 1 is expressed as shown in FIG. 2. The analysis device 35 constituting the non-destructive inspection system according to the first embodiment includes a control unit 351 having a sequence circuit 352, a reflected wave feature extraction logic circuit 353, a feature selection logic circuit 354, a reflected wave emphasis filter coefficient calculation circuit 355, a time axis expansion/compression circuit 356, a filter bank processing logic circuit 357, and a judgment logic circuit 358 as shown in FIG. 2, and can measure the change in the length and cross-sectional area of the measured object 11, as well as inspect the presence or absence of damage, defects, etc., and soundness. The reflected wave feature extraction logic circuit 353 constituting the analysis device 35 is a logic circuit that extracts the features of the reflected wave that propagates through a sound sample for calibration output by the waveform display device 34 and returns, using the empirical mode decomposition (EMD) method, and stores them in the feature storage device 44.

The feature selection logic circuit 354 is a logic circuit that selects the feature of the reflected wave that has propagated through the object to be measured 11 and returned by using the empirical mode decomposition method. The reflected wave emphasis filter coefficient calculation circuit 355 is a logic circuit that converts the feature of the object to be measured 11 using the design information of the object to be measured 11 and calculates the reflected wave emphasis filter coefficient of the object to be measured 11 from the converted feature. The time axis expansion/compression circuit 356 is a logic circuit that prepares multiple extensions of the time axis of the filter and expands or compresses the time axis, taking into account the variation in the propagation speed of the elastic wave. The filter bank processing logic circuit 357 is a logic circuit that processes multiple filters in parallel using the reflected wave emphasis filter coefficient. In other words, the filter bank processing logic circuit 357 performs filter bank processing in parallel using the continuous wavelet method, since the mother function can be determined in advance. The judgment logic circuit 358 is a logic circuit that judges defects in the object to be measured 11 using the features extracted by the reflected wave feature extraction logic circuit 353 and the features of the object to be measured selected by the feature selection logic circuit 354. In other words, the judgment logic circuit 358 judges whether the defect is a scratch in the object to be measured 11 or a reflection from the tip of the object to be measured 11.

A computer system can be configured using a microprocessor (MPU) implemented as a microchip for the analysis device 35. In addition, a digital signal processor (DSP) with enhanced arithmetic operation functions and specialized for signal processing, or a microcontroller (microcomputer) equipped with memory and peripheral circuits and intended for embedded device control, can be used as the analysis device 35 constituting the computer system. Alternatively, the main CPU of a current general-purpose computer can be used for the analysis device 35. Furthermore, a part or all of the configuration of the analysis device 35 can be configured with a programmable logic device (PLD) such as a field programmable gate array (FPGA).

When all or part of the analysis device 35 is configured using a PLD, the waveform data storage device 36, design information storage device 41, defect data storage device 43, feature storage device 44, and analysis data storage device 38 can be configured as memory elements such as memory blocks included in part of the logic blocks that make up the PLD. Furthermore, the analysis device 35 may have a structure in which a CPU core-like array and a PLD-like programmable core are mounted on the same chip. This CPU core-like array includes a hard macro CPU that is mounted in advance inside the PLD, and a soft macro CPU configured using the logic blocks of the PLD. In other words, the PLD may have a configuration in which software processing and hardware processing are mixed.

The control unit 351, the reflected wave feature extraction logic circuit 353, the feature selection logic circuit 354, the reflected wave emphasis filter coefficient calculation circuit 355, the time axis expansion/compression circuit 356, the filter bank processing logic circuit 357, and the judgment logic circuit 358 are connected to each other via the A bus 359a and the B bus 359b. The sequence circuit 352 controls the processing procedures of the control unit 351, the reflected wave feature extraction logic circuit 353, the feature selection logic circuit 354, the reflected wave emphasis filter coefficient calculation circuit 355, the time axis expansion/compression circuit 356, the filter bank processing logic circuit 357, and the judgment logic circuit 358 according to a computer software program. In FIG. 2, a configuration is illustrated in which the impact control device 22, the waveform data storage device 36, the waveform display device 34, the design information storage device 41, and the defect data storage device 43 are connected to the A bus 359a. On the other hand, the B bus 359b is shown to have a feature memory device 44, an analysis data memory device 38, a program memory device 39, an output device 37, and an input device 42 connected to it, but is not limited to the configuration shown in FIG. 2.

The reflected wave feature extraction logic circuit 353, feature quantity selection logic circuit 354, reflected wave emphasis filter coefficient calculation circuit 355, time axis expansion compression circuit 356, filter bank processing logic circuit 357, and judgment logic circuit 358, etc., which are hardware resources constituting the analysis device 35 shown in FIG. 2, formally express hardware resources focusing on logical functions, and do not necessarily mean functional blocks that exist independently as physical areas on a semiconductor chip, but do not deny the actual configuration of programmable logic components implemented on a semiconductor chip such as the "logic block" of a PLD. If some or all of the configuration of the analysis device 35 is configured with a PLD such as an FPGA, the program counter of the sequence circuit 352 shown in FIG. 2 and data buses such as the A bus 359a and the B bus 359b can be omitted.

As shown in FIG. 4, the automatic impact device 21a used in the nondestructive inspection system according to the first embodiment is a solenoid-type device that drives a plunger 211 with an electromagnetic solenoid 212. In the solenoid-type automatic impact device 21a shown in FIG. 4, the plunger 211 driven by the electromagnetic solenoid 212 moves the impact portion 213 downward, and the tip of the impact portion 213 impacts a part of the object 11 to be measured.

The electromagnetic solenoid 212 is stored inside a solenoid container 216 that houses the electromagnetic solenoid 212. As shown in FIG. 4, the striking section 213 has a two-stage cylindrical shape, with the outer diameter of the upper cylinder being larger than that of the lower cylinder that serves as the hammer section. A hole is drilled in the top of the upper cylinder, into which the tip of the plunger 211 is inserted. The striking section 213 is surrounded by a guide section 217, the underside of which is fixed to the surface of the object to be measured 11 via a joint 215. A stepped through hole is opened in the center of the guide section 217, and the inner diameter of the through hole on the upper stage is larger than that of the through hole on the lower stage.

A spring 214 is stored between the step of the through hole and the bottom surface of the upper cylinder of the striking section 213 so as to surround the lower cylinder. A part of the upper surface of the guide section 217 near the outer periphery and a part of the side of the solenoid container 216 are connected by an L-shaped connecting plate 219. The L-shaped lower end of the connecting plate 219 is parallel to the upper surface of the guide section 217 and is fixed to a part of the upper surface of the guide section 217 near the outer periphery by a bolt 220c. A buffer material 218 is inserted between the side of the solenoid container 216 and the rising part of the connecting plate 219. The rising part of the connecting plate 219 is fixed to a part of the side of the solenoid container 216 by a bolt 220a and a bolt 220b via this buffer material 218. By adjusting the elastic force of the spring 214 and the driving force of the electromagnetic solenoid 212, any desired striking force can be obtained.

The automatic impact device 21a shown in FIG. 4 can generate not only pulse or impulse elastic waves in the object 11 to be measured by a single impact, but also sinusoidal elastic waves in the object 11 to be measured by periodic impact. Therefore, if sinusoidal elastic waves of about 2 to 4 KHz are generated, it becomes possible to perform an integrity test using low-frequency elastic waves, and inspections such as the high-frequency impact elastic wave method using elastic waves of various frequencies up to about 1 MHz.

The automatic impact device used in the nondestructive inspection system according to the first embodiment is not limited to the solenoid-type automatic impact device 21a shown in FIG. 4. For example, a pendulum-type automatic impact device 21b as shown in FIG. 5 may be used. In the pendulum-type automatic impact device 21b shown in FIG. 5, an impact part 221 provided at one end of an arm 222a impacts a part of the object 11 to generate a pulse-like or impulse-like elastic wave in the object 11 to be measured. At the other end of the arm 222a, a spring drive part 222b is provided, which is connected in a direction perpendicular to the arm 222a. The arm 222a and the spring drive part 222b form an L-shaped structure. The other end of a spring 223, one end of which is fixed to the spring drive part 222b, is fixed to a spring fixing part 224.

A rotation shaft is provided at the corner of the L-shape where the arm 222a and the spring drive unit 222b are connected, perpendicular to the plane formed by the arm 222a and the spring drive unit 222b. This rotation shaft is fixed so that it can rotate freely by a rotation shaft support unit 225 fixed to the base 227. A rod-shaped support 229 is provided at the end of the base 227 perpendicular to the main surface of the base 227, and a hook 228 that determines the height of the striking unit 221 is provided on part of the support 229. In addition, a rod-shaped stopper 226 is fixed to the top surface of the base 227 near the base of the support 229. The striking speed of the striking unit 221 can be changed by adjusting the distance between the hook 228 and the stopper 226.

The automatic impact device used in the non-destructive inspection system according to the first embodiment may be a motor-type automatic impact device 21c as shown in FIG. 6. In the motor-type automatic impact device 21c shown in FIG. 6, the impact part 231 connected to the rotational movement of the belt 236 driven by the motor 234 moves linearly in the vertical direction, and the tip of the impact part 231 strikes a part of the object to be measured 11, so that a pulse-like or impulse-like elastic wave can be generated in the object to be measured 11. In order to enable the rotational movement of the belt 236, a lower pulley 235a directly connected to the motor 234 and an upper pulley 235b facing the lower pulley 235a are provided at a distance from the lower pulley 235a. In order to connect the impact part 231 to the rotational movement of the belt 236, a rod-shaped impact part 231 is fixed to the lower side of a link part 232 provided on the guide rail 233 with its tip facing vertically. It is also possible to have a structure in which the link portion 232 and the belt 236 are supported without being fixed, allowing them to move freely in the falling direction, and the belt 236 moves downward before the striking portion 231, causing the striking portion 231 to fall freely, so that the tip of the striking portion 231 strikes a part of the object to be measured 11.

Figure 7 is a circuit diagram showing an op-amp feedback type HPF as an example of a waveform selector used in the non-destructive inspection system according to the first embodiment. As shown in Figure 7, two voltage source type (VCVS) type secondary HPFs are connected in series. In the first stage secondary HPF, a second AC coupling capacitor C5 is connected in series to a first AC coupling capacitor C4 located on the input terminal IN side, and the second AC coupling capacitor C5 is connected to the positive terminal of the first operational amplifier. The positive terminal of the first operational amplifier is grounded via a first bias resistor R5. One terminal of a first input resistor R2 is connected to a connection node between the first AC coupling capacitor C4 and the second AC coupling capacitor C5, and the other terminal of the first input resistor R2 is connected to the negative terminal of the first operational amplifier. The high-level power supply side of the first operational amplifier is grounded via a first bypass capacitor C2, and the low-level power supply side of the first operational amplifier is grounded via a second bypass capacitor C9. The high-level power supply for the first operational amplifier can be supplied from the stabilized power supply 33 shown in FIG. 1. The first bypass capacitor C2 and the second bypass capacitor C9 are intended to reduce noise from the stabilized power supply 33 and to reduce the impedance of the power supply line of the stabilized power supply 33.

As shown in FIG. 7, the second-stage secondary HPF is also connected in series with the third AC coupling capacitor C6 located on the output terminal side of the first operational amplifier, and the fourth AC coupling capacitor C7 is connected to the positive terminal of the second operational amplifier. The positive terminal of the second operational amplifier is grounded via the second bias resistor R6. One terminal of the second input resistor R3 is connected to the connection node between the third AC coupling capacitor C6 and the fourth AC coupling capacitor C7, and the other terminal of the second input resistor R3 is connected to the negative terminal of the second operational amplifier. Although not shown in the figure, the high-level power supply side of the second operational amplifier is grounded via the third bypass capacitor, and the low-level power supply side of the second operational amplifier is grounded via the fourth bypass capacitor, just like the first operational amplifier. The high-level power supply of the second operational amplifier can also be supplied from the stabilized power supply 33 shown in FIG. 1. The third and fourth bypass capacitors are intended to reduce noise from the stabilized power supply 33 and to reduce the impedance of the power supply line of the stabilized power supply 33.

<Non-destructive testing methods>
The procedure of the nondestructive inspection according to the first embodiment will be described with reference to the flow chart shown in FIG. 3. First, in step S11, the design information of the calibration sample is input via the input device 42 of the nondestructive inspection system according to the first embodiment, and is stored in the design information storage device 41 of the nondestructive inspection system. In the nondestructive inspection method according to the first embodiment, the elastic wave applied by the automatic impact device 21a is considered to be a short-term pulse or impulse. The reflection of the elastic wave depends on the characteristics when the elastic wave is applied and the characteristics of the reflecting surface. In particular, the magnitude changes depending on the impedance ratio of the boundary surface, and the reflection is similar to that of the free end. In a sound concrete structure used as a calibration sample, the reflection from the tip of the calibration sample basically indicates the main factor of the reflected wave signal.

In the case of a healthy concrete structure to be used as a calibration sample, the length, size, shape, as well as the bulk modulus and density are determined by design, so the design information for the calibration sample is stored in the design information storage device 41. Next, in step S12, the automatic impact device 21a automatically impacts the pile head of the calibration sample, and the detector 31 detects healthy elastic waves that propagate through the calibration sample and return. The waveform of the elastic wave detected by the detector 31 as shown in Figure 8(a) is partially filtered by the waveform selector 32 shown in Figure 7, and after selection, it becomes the waveform shown in Figure 8(b). The waveform of the healthy elastic wave output from the waveform selector 32 can be observed on the waveform display device 24 shown in Figure 1.

Then, in step S13, the reflected wave feature extraction logic circuit 353 constituting the analysis device 35 of the nondestructive inspection system extracts features of healthy elastic waves (reflected waves) according to the following empirical mode decomposition procedure:
(a) In step S12, data of the input signal x(t) from which a portion of the reflected wave waveform has been selected by the waveform selector 32 shown in Fig. 7 is stored in the waveform data storage device 36. The vibration peaks and valleys of the input signal x(t) of a healthy elastic wave appear alternately. Therefore, in step S13, the reflected wave feature extraction logic circuit 353 reads the data of the input signal x(t) from the waveform data storage device 36, detects all extreme values of the input signal x(t) of a healthy elastic wave, and stores the data of all the extreme values in the analysis data storage device 38.
(b) The reflected wave feature extraction logic circuit 353 reads the data of the maximum and minimum points from the analysis data memory device 38, interpolates the maximum and minimum points to obtain the peak envelope (upper envelope) e _max (t) and the valley envelope (lower envelope) e _min (t), and stores the data of the peak envelope e _max (t) and the valley envelope e _min (t) in the analysis data memory device 38.

(c) The reflected wave feature extraction logic circuit 353 reads the data of the peak envelope and the valley envelope from the analysis data storage device 38, calculates the local average m(t) = (e _min (t) + e _max (t) ) / 2 of the peak envelope and the valley envelope, and stores the data of the local average m(t) in the analysis data storage device 38.
(d) The data of the input signal x(t) is read from the waveform data storage device 36, and the data of the local average m(t) is read from the analysis data storage device 38. The reflected wave feature extraction logic circuit 353 then repeats steps (a) to (d) by considering the difference between the input signal x(t) and the local average, y(t) = x(t) - m(t), as an input.
(e) When the standard deviation of the local average becomes equal to or smaller than the threshold, the reflected wave feature extraction logic circuit 353 regards the difference y(t) between the force and the local average as the intrinsic mode function IMF, and ends the loop. In this way, the reflected wave feature extraction logic circuit 353 approximates the peak envelope and the valley envelope, extracts the line passing between them as the first intrinsic mode function IMF1=y(t), and stores the data of the first intrinsic mode function IMF1 in the analysis data storage device 38.

(f) The reflected wave feature extraction logic circuit 353 repeats the procedure of steps (a) to (f) using the residual _r1 (t) = x(t) - y(t) between the input signal x(t) of a healthy elastic wave and the intrinsic mode function IMF as a new input signal x(t). That is, the reflected wave feature extraction logic circuit 353 subtracts _y1 (t) + _y2 (t) + _y3 (t) + ... + _yk (t) from the original input signal x(t) and repeats this process multiple times until convergence is achieved.
(g) The reflected wave feature extraction logic circuit 353 extracts the first intrinsic mode function IMF1 to the k-th intrinsic mode function IMFk (all intrinsic mode functions IMF), and ends when there is only one extreme value of x(t). If the final residual is r _final (t), then

x(t) =y ₁ (t)＋y ₂ (t)＋y ₃ (t)＋……＋y _k (t)＋r _final (t)……(1)

k is the number of channels of the intrinsic mode function IMF. According to the formula (1), the reflected wave feature extraction logic circuit 353 can decompose the vibration.

Figure 9(a) shows a reflected wave from a healthy concrete structure used in empirical mode decomposition as an input signal. Figure 9(b) is a diagram explaining the first intrinsic mode function IMF1 when the input signal from the healthy concrete structure shown in Figure 9(a) is decomposed into 10 intrinsic mode functions IMF by the empirical mode decomposition method. Figure 9(c) is a diagram explaining the second intrinsic mode function IMF2 when the input signal shown in Figure 9(a) is decomposed into 10 intrinsic mode functions IMF by the empirical mode decomposition method. Similarly, Figure 9(d) is a diagram explaining the third intrinsic mode function IMF3 when the input signal shown in Figure 9(a) is decomposed into 10 intrinsic mode functions IMF by the empirical mode decomposition method. In the decomposition by the empirical mode decomposition method as shown in Figure 9, the reflected wave feature extraction logic circuit 353 adopts as an eigenmode the one in which the ratio of the energy of the signal of only the peaks to the energy of the valleys is close to 1 and the relative tolerance is large. In step S13, the reflected wave feature extraction logic circuit 353 selects the front and rear distances that are considered to be approximately equal to the design distance as features. In step S14, the reflected wave feature extraction logic circuit 353 stores the features of healthy elastic waves (reflected waves) extracted in the feature storage device 44.

Next, in step S15, if the design information of the object 11 to be measured is known, the design information of the object 11 to be measured is input via the input device 42 and stored in the design information storage device 41 of the nondestructive inspection system according to the first embodiment. In step S16, the reflected wave feature extraction logic circuit 353 selects the feature of the empirical mode decomposition of the object 11 to be measured and stores it in the feature storage device 44 of the nondestructive inspection system according to the first embodiment. If the object 11 to be measured has a defect such as a crack, a scratch, or a crack, the reflected wave from the defect can be assumed to have similar characteristics to the reflection from the tip of a sound structure, although the size and shape of the reflected wave will change from that of a sound concrete structure. Therefore, the reflected wave feature extraction logic circuit 353 extracts the eigenmode assuming that the reflected wave from the defective part of the object 11 to be measured has the same characteristics as the reflected wave from a sound structure, and stores it in the analysis data storage device 38.

Figure 10(a) is a diagram showing a reflected wave from a measured object 11 including defects such as cracks, scratches, and cracks as an input signal. Figure 10(b) is a diagram explaining a first intrinsic mode function IMF1 when the input signal from the measured object 11 including defects shown in Figure 10(a) is decomposed into 10 intrinsic mode functions IMF by the empirical mode decomposition method. Figure 10(c) is a diagram explaining a second intrinsic mode function IMF2 when the input signal shown in Figure 10(a) is decomposed into 10 intrinsic mode functions IMF by the empirical mode decomposition method. Similarly, Figure 10(d) is a diagram explaining a third intrinsic mode function IMF3 when the input signal shown in Figure 10(a) is decomposed into 10 intrinsic mode functions IMF by the empirical mode decomposition method.

As shown in an enlarged view in FIG. 11, a feature value as circled with the symbol A is found in the third intrinsic mode function IMF3 shown in FIG. 10(d). In step S17, the reflected wave emphasis filter coefficient calculation circuit 355 constituting the analysis device 35 reads out the feature value of the healthy structure from the feature value storage device 44, and uses the read out feature value as a selection filter to calculate the reflected wave emphasis filter coefficient of the measured object 11 as shown in the upper part of FIG. 12. In step S18, the time axis expansion/compression circuit 356 constituting the analysis device 35 prepares multiple extensions of the time axis of the filter in consideration of the variation in the propagation speed of elastic waves, and expands or compresses the time axis as shown in the lower part of FIG. 12.

Then, in step S19, the automatic impact device 21a of the non-destructive inspection system according to the first embodiment automatically impacts the pile head of the object to be measured 11, and the detector 31 detects the reflected wave that propagates through the object to be measured 11 and returns. The reflected wave waveform detected by the detector 31, including reflections from defects such as cracks, scratches, and fissures, is partially selected by the waveform selector 32, and the elastic waves from the object to be measured 11 are observed on the waveform display device 34, after which the data with the selected portion of the waveform is stored in the waveform data storage device 36. In step S20, the waveform data is read from the waveform data storage device 36, and the filter bank processing logic circuit 357 constituting the analysis device 35 performs filter bank processing on the read waveform data as shown in FIG. 13.

As can be seen from FIG. 13, in the filter bank processing, multiple outputs y(t) are obtained by performing parallel processing of filters on a single input x(t). Parallel processing of filters requires processing time, but since the mother function can be determined in advance, it is realized using the continuous wavelet method as shown in FIG. 13. In step S21, the judgment logic circuit 358 constituting the analysis device 35 judges whether or not the object is a scratch in the object 11 to be measured or a reflection from the tip of the object 11 to be measured. When judging whether or not the object is a scratch in the object 11 to be measured or a reflection from the tip of the object 11 to be measured, a deep learning method is used to compare the data stored in the design information storage device 41 and the defect data storage device 43 with multiple outputs y(t) obtained by the filter bank processing. The judgment logic circuit 358 displays the measurement results of the change in the length and cross-sectional area of the object 11 to be measured, the presence or absence of damage, defects, etc., and the inspection results of the soundness of the object 11 to be measured by the output device 37. Furthermore, the judgment logic circuit 358 stores the inspection data on the presence or absence of damage, defects, etc., and soundness determined, together with the measurement results of the changes in the length and cross-sectional area of the object to be measured 11, in the defect data storage device 43 to accumulate the defect data.

According to the non-destructive inspection system and non-destructive inspection method of the first embodiment, the eigenmode obtained from a sound concrete structure is used as an extraction filter for reflected waves from a measured object 11 having defects such as cracks, scratches, and fissures, so that the basic characteristics of the reflected waves can be emphasized and other components other than the reflected waves can be suppressed. Furthermore, according to the non-destructive inspection system and non-destructive inspection method of the first embodiment, multiple extensions of the time axis of the filter can be prepared and filtering can be performed in parallel, so that variations in propagation speed due to aging and the like can be absorbed, and variations in the characteristics of each structure can be absorbed.

Other Embodiments
Although the present invention has been described by the first embodiment, the description and drawings forming a part of this disclosure should not be understood as limiting the present invention. From this disclosure, various alternative embodiments, examples, and operation techniques will become apparent to those skilled in the art. Thus, the present invention naturally includes various embodiments not described here. Therefore, the technical scope of the present invention is defined only by the invention-specific matters related to the claims, which can be interpreted as appropriate from the above description. For example, a non-destructive inspection system, a non-destructive inspection method, and a non-destructive inspection program specify non-destructive in the wording, but it cannot be denied that some fine particles, fragments, etc. may be lost due to impact during inspection, but since the wording non-destructive inspection is used by definition, it can be said that such a case is naturally allowed by a non-destructive inspection system, a non-destructive inspection method, and a non-destructive inspection program.

11...Objects to be measured 21a, 21b, 21c...Automatic impact device 22...Impact control device 24...Waveform display device 31...Detector 32...Waveform selector 33...Stabilized power supply 34...Waveform display device 35...Analysis device 36...Waveform data storage device 37...Output device 38...Analysis data storage device 39...Program storage device 41...Design information storage device 42...Input device 43...Defect data storage device 44...Feature storage device 211...Plunger 212...Electromagnetic solenoid 213...Impact portion 214...Spring 215...Joint portion 216...Solenoid container 217...Guide portion 218...Cushioning material 219...Connecting plates 220a, 220b, 220c...Bolt 221... Impact section 222a...arm 222b...spring drive section 223...spring 224...spring fixing section 225...rotating shaft support section 226...stopper 227...base 228...hook 229...support 231...impact section 232...link section 233...guide rail 234...motor 235a...lower pulley 235b...upper pulley 236...belt 351...control section 352...sequence circuit 353...reflected wave feature extraction logic circuit 354...feature quantity selection logic circuit 355...reflected wave emphasis filter coefficient calculation circuit 356...time axis expansion/compression circuit 357...filter bank processing logic circuit 358...determination logic circuit 359a...A bus 359b...B bus,

Claims (2)

An automatic impact device to be attached to a part of the object to be measured;
a detector for detecting a reflected wave of the elastic wave generated by the automatic impact device, the detector being provided on another part of the object to be measured;
a waveform selector coupled to the detector for selecting a portion of the waveform output by the detector;
an analysis device connected to the waveform selector, for selecting a feature quantity from a signal output by the waveform selector using an empirical mode decomposition technique, and for performing an analysis for inspecting the object under test using the feature quantity;
Equipped with
The analysis device includes:
a reflected wave feature extraction logic circuit that extracts features of a healthy reflected wave output by the waveform selector, propagated through the calibration sample, and returned by using the empirical mode decomposition technique;
a feature selection logic circuit that selects a feature of the reflected wave that has propagated through the object to be measured and returned by the empirical mode decomposition method;
a judgment logic circuit that judges defects in the object to be measured using the features extracted by the reflected wave feature extraction logic circuit and the feature quantity of the object to be measured selected by the feature quantity selection logic circuit;
a reflected wave emphasizing filter coefficient calculation circuit that converts a feature amount of the object to be measured using design information of the object to be measured and calculates a reflected wave emphasizing filter coefficient of the object to be measured from the converted feature amount;
a time axis expansion/compression circuit for expanding or compressing a time axis of each of a plurality of filters;
a filter bank processing logic circuit that processes the plurality of filters in parallel using the reflected wave emphasis filter coefficients;
A non-destructive inspection system comprising : In the non-destructive inspection system,
a design information storage device connected to the analysis device for storing design information data of a calibration sample;
2. The non-destructive inspection system according to claim 1, further comprising:

Images