JP7235265B2 - Concrete structure diagnostic method and its diagnostic device - Google Patents

Description

The present invention relates to a diagnostic method and diagnostic apparatus for a concrete structure in which conductor rods (eg, anchor bolts, etc.) are partially embedded in concrete.

2. Description of the Related Art Conventionally, a non-destructive inspection technique is known, which uses an electromagnetic pulse method to discover problems such as deterioration and defects in fixed portions of conductor rods partially embedded in concrete. The electromagnetic pulse method is a type of inspection method that uses elastic waves (acoustic waves). It is a non-destructive inspection technology that grasps the state of an object by generating elastic waves from the conductor rod itself, and receiving and analyzing the elastic waves.

Patent Documents 1, 2, etc. disclose a concrete structure for diagnosing problems and defects such as deterioration in fixed portions of conductor rods (for example, post-installed anchor bolts) partially embedded in concrete using an electromagnetic pulse method. A method for diagnosing is described. Post-installed anchor bolts are anchor bolts that are fixed by drilling pilot holes in concrete and partially embedding anchor bolts.

In the diagnostic methods described in Patent Documents 1 and 2, etc., a post-installed anchor bolt that is partially embedded in concrete is targeted for diagnosis, and a coil is placed around the exposed portion of the anchor bolt that is exposed from the concrete surface. and place the sensor on the concrete surface around the anchor bolt. Then, a pulse current output from the power supply unit is passed through the coil to generate a pulse magnetic field from the coil. The generated pulsed magnetic field acts on the anchor bolt to generate elastic waves. The elastic wave passes through the concrete and is received by a sensor installed on the concrete surface. By analyzing the reception results with an analysis processing device, defects such as deterioration of anchor bolt fixing parts and construction defects such as defects are detected. I am diagnosing.

JP 2014-228324 A JP 2015-99060 A

However, in the conventional diagnostic method for concrete structures, a method for evaluating construction defects of anchor bolts has not been sufficiently established. Workers visually identify defective anchor bolts. Therefore, not only does the accuracy of determination of construction defects vary depending on the experience of the inspector, but also the determination process is complicated, which is disadvantageous and inconvenient.

SUMMARY OF THE INVENTION Accordingly, it is an object of the present invention to provide a diagnostic method for concrete structures and a diagnostic apparatus for the same, which are capable of evaluating construction defects in concrete structures simply, accurately, and quantitatively. do.

In order to achieve the above object, a concrete structure diagnostic method according to claim 1 of the present invention is a concrete structure diagnostic method configured by partially embedding a conductor rod in concrete, wherein A coil is placed on the exposed part exposed from the concrete surface, one sensor is placed on the concrete surface around the conductor rod or on the exposed part of the conductor rod, and a pulse current is passed through the coil to An elastic wave generation process that generates a pulse magnetic field and applies the pulse magnetic field to a conductor rod to generate an elastic wave, a reception process that receives the elastic wave with a sensor and generates a reception signal, and a reception signal that is filtered Data processing that obtains the time-axis waveform and the frequency spectrum obtained by Fourier transforming the received signal and outputs the data processing result, and judgment processing that determines whether the conductor rod has been improperly constructed on the concrete based on the data processing result. , wherein the judgment processing includes evaluation index processing that obtains a plurality of evaluation indices based on the data processing results, and evaluation that assigns evaluation points for evaluating construction defects to the plurality of evaluation indices based on a threshold value. and point assignment processing, wherein the evaluation index processing includes an evaluation index of the waveform duration until the maximum amplitude of the time-axis waveform is attenuated to a predetermined amplitude, and an evaluation index of the center-of-gravity frequency of the spectrum in the frequency spectrum. , the evaluation index of the standard deviation of the spectrum, the evaluation index of the number of spectrum peaks, and the evaluation index of the correlation coefficient between the spectrum of the defective construction and the average spectrum of a plurality of standard constructions, and in the evaluation point assignment process, Evaluation index of waveform duration, evaluation index of center-of-gravity frequency of spectrum, evaluation index of standard deviation of spectrum, evaluation index of number of spectrum peaks, evaluation of correlation coefficient between spectrum of poor construction and average spectrum of multiple standard constructions Characterized by diagnosing by adding evaluation points to indicators based on thresholds and totaling them ,

In the above configuration, preferably, the evaluation index of the waveform duration time expresses a phenomenon in which the convergence time of the time-axis waveform becomes longer due to loose fixing of the conductor rod due to poor construction, and the evaluation index of the center-of-gravity frequency of the spectrum is It expresses the phenomenon that the frequency spectrum shifts to the low frequency side due to poor construction, and the evaluation index of the standard deviation of the spectrum expresses the degree of concentration of the frequency spectrum that appears due to poor construction. The numerical evaluation index expresses multiple frequency spectra that appear due to poor construction when the conductor bar has a complicated structure, and the correlation between the spectrum of poor construction and the average spectrum of multiple standard constructions. The numerical evaluation index expresses the correlation coefficient between the frequency spectrum appearing due to poor construction and the FFT waveform obtained by averaging the spectrum due to a plurality of standard constructions .
Preferably, in the evaluation index processing, an evaluation index of a cross-correlation function between the time-axis waveform of the defective construction and the average time-axis waveform of the plurality of standard constructions is obtained , and in the evaluation point assignment processing, the time-axis of the defective construction is further obtained. Evaluation points are given based on a threshold to the evaluation index of the cross-correlation function between the waveform and the average time-axis waveforms of multiple standard constructions .
The evaluation index of the cross-correlation function between the time-axis waveform of defective construction and the average time-axis waveform of a plurality of standard constructions is preferably normalized by the magnitude of the two functions that constitute the cross-correlation function.
The evaluation index of the cross-correlation function between the time-axis waveform of construction failure and the average time-axis waveform of a plurality of standard constructions is preferably such that the greater the degree of construction failure, the greater the difference in the shape of the time-axis waveform, and the higher the cross-correlation function. It expresses a phenomenon that becomes smaller .
The conductor rod is preferably a conductive member including a post-installed anchor bolt.

Further, according to a seventh aspect of the present invention, there is provided a concrete structure diagnosis apparatus in which a conductor rod is partially embedded in concrete, and the conductor rod is exposed from the concrete surface. a coil placed on the exposed portion of the conductor bar, one sensor placed on the concrete surface around the conductor rod or placed on the exposed portion of the conductor rod; An elastic wave generating means for generating a pulse magnetic field and causing the pulse magnetic field to act on a conductor rod to generate an elastic wave, and a waveform receiving section for receiving the elastic wave with a sensor and generating a reception signal.

Further, in the diagnostic apparatus of the present invention, there are provided a time-domain waveform processing unit that filters a received signal to obtain a time-domain waveform, a frequency spectrum processing unit that performs Fourier transform on the received signal to obtain a frequency spectrum, and a time-domain waveform processing unit. A waveform evaluation index unit that obtains an evaluation index for waveform duration based on the processing results of (1), and a spectrum center-of-gravity frequency evaluation index, a spectrum standard deviation evaluation index, and the number of spectrum peaks based on the processing results of the frequency spectrum processing unit a spectrum evaluation index part that obtains the evaluation index of, waveform duration evaluation index, spectrum centroid frequency evaluation index, spectrum standard deviation evaluation index, spectrum peak number evaluation index, construction defect spectrum and multiple standards an evaluation point assigning unit that assigns evaluation points for evaluation of poor construction based on a threshold to the sixth evaluation index of the correlation coefficient with the average spectrum of construction, and totals the evaluation points. .

In order to achieve the above object, the diagnostic apparatus for concrete structures according to claim 8 of the present invention includes a coil arranged in an exposed portion of a conductor rod exposed from the concrete surface, and a concrete structure surrounding the conductor rod. One sensor placed on the surface or placed on the exposed part of the conductor rod, and a pulse current is passed through the coil to generate a pulse magnetic field from the coil, and the pulse magnetic field is applied to the conductor rod. An elastic wave generating means for generating an elastic wave and a waveform receiving section for receiving the elastic wave with a sensor and generating a reception signal are provided.

Further, in the diagnostic apparatus of the present invention, there are provided a time-domain waveform processing unit that filters a received signal to obtain a time-domain waveform, a frequency spectrum processing unit that performs Fourier transform on the received signal to obtain a frequency spectrum, and a time-domain waveform processing unit. A waveform evaluation index unit that obtains an evaluation index of waveform duration and a cross-correlation function between the time-axis waveform of construction failure and the average time-axis waveform of multiple standard constructions based on the processing results of , and a frequency spectrum processing unit. Based on the processing result, a spectrum evaluation index unit that obtains an evaluation index of the spectrum center-of-gravity frequency, a spectrum standard deviation evaluation index, and an evaluation index of the number of spectrum peaks, a waveform duration evaluation index, and a spectrum-center-of-gravity frequency evaluation index , the evaluation index of the standard deviation of the spectrum, the evaluation index of the number of peaks in the spectrum, and the evaluation index of the correlation coefficient between the spectrum of the construction defect and the average spectrum of multiple standard constructions. and an evaluation point giving unit that gives the evaluation points of each and totals them.
For example, a conductor rod is a conductive member that includes a post-installed anchor bolt.

According to the concrete structure diagnosis method and the configuration of the diagnosis apparatus according to claim 1 of the present invention, only one sensor for detecting elastic waves is required, so the manufacturing cost is low, and the subsequent data Processing is also simplified, multiple evaluation indices are converted into points from the signals received by the sensors, and the total evaluation points are used to evaluate construction defects in concrete structures, so construction defects in concrete structures can be evaluated simply and accurately. can be
Further, according to the concrete structure diagnosis method of claim 3 and the configuration of the diagnosis device of claim 8 of the present invention, the construction failure time, which is the time-axis waveform of the elastic wave detected by the sensor , is Since the evaluation index of the cross-correlation function between the axial waveform and the average time-axis waveform of multiple standard constructions is used to evaluate the construction failure of the concrete structure, even if there is a phase shift, the phase shift can be evaluated. The correlation can be calculated without being greatly affected, and the construction failure of the concrete structure can be evaluated more accurately.

1 is a schematic configuration diagram showing a diagnostic apparatus for concrete structures in Example 1 of the present invention; FIG. 2 is a functional block diagram showing a configuration example of an information processing apparatus in FIG. 1; FIG. FIG. 2 is a diagram showing measurement conditions of elastic waves in FIG. 1 ; FIG. 2 is a diagram showing measurement conditions of elastic waves in FIG. 1 ; FIG. 2 is a plan view showing a test body produced as a concrete structure; FIG. 6 is an enlarged side view showing an adhesive resin-encapsulated anchor bolt embedded in the specimen of FIG. 5; FIG. 10 is a plan view showing another test body produced as a concrete structure; FIG. 8 is an enlarged side view showing a metal-based sleeve expandable anchor bolt embedded in the specimen of FIG. 7; 1 is a flow chart showing a diagnostic method for a concrete structure in Example 1 of the present invention. FIG. 3 is an explanatory diagram of a waveform energy ratio in FIG. 2; FIG. 3 is an explanatory diagram of a waveform energy ratio in FIG. 2; FIG. 3 is an explanatory diagram of a waveform duration time in FIG. 2; FIG. 3 is an explanatory diagram of the center-of-gravity frequency of the spectrum in FIG. 2; FIG. 3 is an explanatory diagram of the standard deviation of the spectrum in FIG. 2; FIG. 3 is an explanatory diagram of the number of peaks in the spectrum in FIG. 2; FIG. 3 is a diagram for explaining how to assign evaluation points to the waveform energy ratio in FIG. 2; FIG. 3 is a diagram for explaining how to assign evaluation points to waveform durations in FIG. 2 ; FIG. It is a figure which shows the summary of a threshold value and an evaluation point. It is a figure which shows the evaluation point graph of the construction failure example. FIG. 3 is a functional block diagram showing a diagnostic device for concrete structures in Example 2 of the present invention; 6 is a flow chart showing a concrete building diagnosis method in Example 2. FIG. FIG. 10 is a diagram showing FFT waveforms when 1 to N standard constructions to be evaluated are measured. FIG. 10 is a diagram showing an average of FFT waveforms when 1 to N standard constructions to be evaluated are measured; It is a figure which shows comparison of the FFT waveform when a new evaluation object is set to standard construction, and the average (x) of the FFT waveform of standard construction. It is a figure which shows the comparison of the FFT waveform of the FFT waveform when evaluation object is made into poor construction, and the average (x) of the FFT waveform of standard construction. FIG. 10 is a diagram showing a summary of thresholds and evaluation points in Example 2; FIG. 11 is a diagram showing an evaluation point graph of an example of poor construction in Example 2; FIG. 3 is a schematic configuration diagram showing a diagnostic apparatus for concrete structures in Example 3 of the present invention; 29 is a functional block diagram showing a configuration example of an information processing device in FIG. 28; FIG. FIG. 29 is a diagram showing measurement conditions for the elastic wave of FIG. 28; FIG. 29 is a diagram showing measurement conditions for the elastic wave of FIG. 28; Fig. 10 is a flow chart showing a concrete building diagnosis method according to a third embodiment of the present invention; It is a figure which shows the summary of a threshold value and an evaluation point. FIG. 30 is a functional block diagram showing a configuration example of an information processing apparatus similar to that of FIG. 29 in Embodiment 4 of the present invention; FIG. 35 is a schematic diagram showing a first configuration example of the evaluation analysis device in FIG. 34; FIG. 36 is a schematic diagram showing an example of a teacher signal used in the evaluation analysis device of FIG. 35; FIG. 36 is a schematic diagram showing identification results in the evaluation analysis device of FIG. 35; FIG. 36 is a graph showing input data during known standard construction that is input to the input layer during learning in the evaluation analysis device of FIG. 35; FIG. 36 is a graph showing known input data at the time of depth filling construction that is input to the input layer during learning in the evaluation analysis apparatus of FIG. 35; FIG. 36 is a graph showing input data at the time of known front filling construction that is input to the input layer during learning in the evaluation analysis device of FIG. 35; FIG. 36 is a graph showing input data at the time of unknown standard construction that is input to the input layer at the time of identification in the evaluation analysis device of FIG. 35; FIG. 36 is a graph showing input data at the time of unknown back-filling construction input to the input layer at the time of identification in the evaluation analysis device of FIG. 35; FIG. 36 is a graph showing input data at the time of unknown front filling construction input to the input layer at the time of identification in the evaluation analysis apparatus of FIG. 35; FIG. 35 is a schematic diagram showing a modification of the evaluation analysis device in FIG. 34; FIG. 45 is a schematic diagram showing an example of a teacher signal used in the evaluation analysis device of FIG. 44; FIG. 45 is a schematic diagram showing identification results in the evaluation analysis device of FIG. 44; FIG. 30 is a functional block diagram showing a configuration example of an information processing apparatus similar to that of FIG. 29 in Embodiment 5; 10 is a flow chart showing a concrete building diagnostic method in Example 5. FIG. FIG. 5 is a diagram showing time-axis waveforms when 1 to N standard constructions to be evaluated are measured. FIG. 10 is a diagram showing an average of time-axis waveforms when 1 to N standard constructions to be evaluated are measured. When the new evaluation target is the standard construction, (A) is the time axis waveform, (B) is the cross-correlation function with the average time axis waveform of FIG. 50, and (C) is the normalized cross-correlation function. be. (A) is a time-axis waveform, (B) is a cross-correlation function with the average time-axis waveform of FIG. 50, and (C) is a normalized cross-correlation function when the new evaluation target is construction failure. be. FIG. 12 is a diagram showing a summary of thresholds and evaluation points in Example 5; FIG. 11 is a diagram showing an evaluation point graph of a poor construction example of Example 5;

Modes for carrying out the invention will become apparent from the following description of preferred embodiments, read in conjunction with the accompanying drawings. However, the drawings are for illustrative purposes only and do not limit the scope of the invention.

(Diagnostic device for concrete structures)

1(a) and 1(b) are schematic configuration diagrams showing a diagnosis apparatus for concrete structures according to Embodiment 1 of the present invention. FIG. b) is a configuration diagram of a waveform receiving section (for example, an analog/digital conversion device, hereinafter referred to as an "AD conversion device") in FIG. It is a functional block diagram showing an example.

As shown in FIG. 1( a ), the diagnostic apparatus for concrete structures diagnoses a conductor rod (for example, a post-installed anchor bolt) 10 partially embedded in concrete 1 . It is a device for diagnosing failures and defects such as deterioration in the embedded fixing part. The post-installed anchor bolt 10 is to be fixed by drilling a pilot hole in the concrete 1 and partially embedding the anchor bolt 10 . A plurality of first sensors (for example, AE (Acoustic Emission) sensors) 21-1 to 21-4 for detecting elastic waves are arranged on the surface of the concrete 1 around the anchor bolt 10. It is fixed with an adhesive tool (glue gun) or the like.

A plurality of sensors 21-1 to 21-4 are fixed on the concrete surface on a circumference centered on the anchor bolt 10, for example, four at 90° intervals. In the anchor bolt 10, a second sensor (for example, a small AE sensor) 22 for detecting elastic waves is arranged at the upper end of the exposed portion exposed from the concrete surface and fixed with a glue gun or the like.
Furthermore, a coil (for example, ring coil) 23 for generating a pulse magnetic field is arranged around the exposed portion of the anchor bolt 10 . The upper end of the coil 23 is installed so as to match the upper end of the anchor bolt 10, for example.

A coil unit 24 for generating a pulse current is connected to the coil 23 via a cable. Furthermore, an electromagnetic pulse power supply 25 for power supply is connected to the coil unit 24 via a cable. The coil unit 24 and the electromagnetic pulse power supply 25 constitute an elastic wave generating means for generating a pulse magnetic field from the coil 23 by applying a pulse current to the coil 23 and applying the pulse magnetic field to the anchor bolt 10 to generate an elastic wave. are doing.

An AD converter 26 as a waveform receiver is connected to the plurality of sensors 21-1 to 21-4 and 22 via cables. The AD conversion device 26 amplifies the analog elastic wave signals received by the sensors 21-1 to 21-4, converts them into a first received signal S26a composed of digital signals, and converts the analog elastic wave signals received by the sensor 22. is amplified and then converted into a second received signal S26b composed of a digital signal.

As shown in FIG. 1B, the AD conversion device 26 converts, for example, an analog elastic wave signal into a digital signal with a sampling frequency of 100 MHz and a data length of 10 ms. An information processing device 30 is connected to the AD conversion device 26 via a signal line. The information processing device 30 is a device for calculating input reception signals S26a and S26b and evaluating problems and defects such as deterioration of the fixed portion of the anchor bolt 10. For example, a personal computer (PC) or the like may It is configured.

FIGS. 3(a), (b) and FIGS. 4(a), (b) are diagrams showing the measurement conditions of the elastic wave of FIG.

The post-installed anchor bolt 10 shown in FIG. 1 is to be fixed by drilling a pilot hole in the concrete 1 and partially embedding the anchor bolt 10 in the pilot hole. , metal anchor bolts, and other anchor bolts). Adhesive anchor bolts use capsule-type or injection-type adhesive that fills pre-drilled holes in the base material and hardens through a chemical reaction to physically fix the fixing part. Fixing methods such as capsule fixing, inorganic tube fixing, and resin injection fixing are available. Metal anchor bolts are divided into metal expansion anchor bolts (for example, metal sleeve expansion type) and other metal anchor bolts. A metal-based sleeve expansion type anchor bolt is mechanically fixed to the inner wall of the pilot hole by opening the sleeve, which is the expanded portion, in the pilot hole drilled in advance in the base material.

FIGS. 3(a) and 3(b) show an example of measurement conditions using an adhesive resin encapsulated anchor bolt 10A. FIG. b) is an enlarged side view of the anchor bolt;

As shown in FIG. 3(a), the concrete sensors 21-1 to 21-4 are, for example, AE sensors weighing 70 g. The exposed portion sensor 22 of the anchor bolt 10A is, for example, a small AE sensor weighing 8 g. A distance between the anchor bolt 10A and the sensors 21-1 to 21-4 is, for example, 100 mm. The five sensors 21-1 to 21-4, 22 are fixed with a glue gun, for example, and measure simultaneously. The sensors 21-1 to 21-4 are not limited to measurements at the same time, and may be measured one by one.

As shown in FIG. 3(b), the concrete 1 is provided with a pilot hole 2. As shown in FIG. The pilot hole 2 has a diameter of 19 mm and a depth of 130 mm, for example. be done. The anchor bolt 10A is made of chromium molybdenum steel (SNB7) with a diameter M of 16 mm and a length of 230 mm, for example. The anchor bolt 10A is composed of, for example, a fixing portion 11 having a length of 130 mm inserted into the pilot hole 2 and an exposed portion 12 having a length of 100 mm exposed from the concrete surface.

FIGS. 4(a) and 4(b) show an example of measurement conditions using a metal-based sleeve expandable anchor bolt 10B. FIG. 4(a) is a perspective view around the anchor bolt, and b) is an enlarged side view of the anchor bolt;

As shown in FIG. 4(a), the concrete sensors 21-1 to 21-4 are, for example, AE sensors weighing 70 g, similar to FIG. 3(a). A sensor 22 is arranged at the upper end of the exposed portion 15 of the anchor bolt 10B via a magnet 27 for applying a static magnetic field. The magnet 27 is for applying a bias static magnetic field in the axial direction of the anchor bolt 10B in order to reinforce the pulse magnetic field generated in the coil 23. As shown in FIG. Similar to FIG. 3(a), the sensor 22 is, for example, a small AE sensor weighing 8 g. A distance between the anchor bolt 10B and the sensors 21-1 to 21-4 is, for example, 100 mm. The five sensors 21-1 to 21-4, 22 and the magnet 27 are fixed with a glue gun, for example. Five sensors 21-1 to 21-4, 22 measure simultaneously. The sensors 21-1 to 21-4 are not limited to measurements at the same time, and may be measured one by one.

As shown in FIG. 4(b), the concrete 1 is provided with a pilot hole 2. As shown in FIG. The pilot hole 2 has, for example, a diameter φ of 35 mm and a depth of 155 mm. The anchor bolt 10B is made of stainless steel (SUS material) with a diameter M of 24 mm and a length of 230 mm, for example. Therefore, the magnet 27 for magnetic field reinforcement is used. The anchor bolt 10B includes, for example, a metal sleeve 13 with a length of 140 mm that is driven into the pilot hole 2, a fixing portion 14 with a length of 155 mm that is inserted into the sleeve 13, and a length of 75 mm that is exposed from the concrete surface. and the exposed portion 15 of the . The depth of the head of the sleeve 13 is, for example, 20 mm before driving and 0.7 mm to 2.8 mm after driving.

FIG. 2 is a functional block diagram showing a configuration example of the information processing device 30 in FIG.
This information processing device 30 includes a time axis waveform processing unit 40 and a frequency spectrum processing unit 50 that perform data processing on the first received signal S26a and the second received signal S26b output from the AD conversion device 26, and the time axis waveform processing unit 40 and the frequency spectrum processing unit 50. A waveform evaluation index unit 60 connected to the output side of the waveform processing unit 40, a spectrum evaluation index unit 70 connected to the output side of the frequency spectrum processing unit 50, the waveform evaluation index unit 60 and the spectrum evaluation index unit 70 and an evaluation point giving unit 80 connected to the output side of the.

The time-axis waveform processing unit 40 receives the first reception signal S26a and the second reception signal S26b output from the AD conversion device 26, filters the first reception signal S26a, and obtains a signal of the first time-axis waveform S44a. is obtained, and the signal of the second time domain waveform S44b is obtained by filtering the second received signal S26b. The time-axis waveform processing unit 40 includes, for example, a 2 MHz low-pass filter (hereinafter referred to as "LPF") 41 that passes only the low frequency components of the received signals S26a and S26b, and the output signal of this LPF 41 as a text file (for example, CSV (Comma Separated Values) file), a thinning unit 43 that performs thinning for simplifying the processing of the signal after file conversion, and only the high frequency components of the signal after thinning. and a 2.5 kHz high-pass filter (hereinafter referred to as “HPF”) 44 for passing. The HPF 44 has a function of outputting signals of a first time domain waveform S44a and a second time domain waveform S44b having a sampling frequency of 1 MHz and a data length of 10 ms, for example.

The frequency spectrum processing unit 50 receives the first received signal S26a and the second received signal S26b output from the AD conversion device 26, and Fourier transforms the received signal S26a (for example, fast Fourier transform, hereinafter referred to as "FFT"). ) to obtain a first frequency spectrum, and an FFT section 51 that obtains a second frequency spectrum by performing an FFT on the received signal S26b, and a text file (for example, , CSV file). The first frequency spectrum S52a and the second frequency spectrum S52b converted into text files have, for example, a frequency range of 0 to 50 kHz, a frequency resolution of 200 Hz, and 250 FFT points.

The waveform evaluation index unit 60 calculates a first evaluation index EI1 for the waveform energy ratio SR and a second evaluation index EI2 for the waveform duration TC based on the signals of the first and second time-axis waveforms S44a and S44b output from the HPF 44. is a request. Based on the signals of the first and second frequency spectra S52a and S52b after file conversion, the spectrum evaluation index unit 70 calculates a third evaluation index EI3 for the center-of-gravity frequency SC of the spectrum and a fourth evaluation index EI4 for the standard deviation SD of the spectrum. , and the fifth evaluation index EI5 of the number of peaks SP of the spectrum.

Here, the waveform energy ratio SR, which is the first evaluation index EI1, is the ratio of the waveform energy of the concrete surface to the waveform energy of the anchor bolt exposed portion. The phenomenon in which the waveform energy of the part increases is expressed by the ratio. The waveform continuation time TC, which is the second evaluation index EI2, is the time required for the time-axis waveform to attenuate to, for example, less than 10% of the maximum amplitude of the crest of the wave. It expresses an event in which the convergence time of the waveform on the time axis becomes long. The center-of-gravity frequency SC of the spectrum, which is the third evaluation index EI3, is the center-of-gravity frequency of a specific frequency spectrum, and expresses an event in which the frequency spectrum shifts to the low-frequency side due to poor construction. The standard deviation SD of the spectrum, which is the fourth evaluation index EI4, is the sharpness of the peak of a specific frequency spectrum, and expresses the degree of concentration of the frequency spectrum that appears due to poor construction. Furthermore, the number of spectral peaks SP, which is the fifth evaluation index EI5, is the number of peaks in the frequency spectrum. It expresses the number of peaks of multiple frequency spectra appearing by .

The evaluation point giving unit 80 connected to the output side of the waveform evaluation index unit 60 and the spectrum evaluation index unit 70 receives one of the first, second, third, fourth, and fifth evaluation indexes EI1 to EI5. In response to the above, evaluation points for evaluation of poor construction are given based on threshold values, and the total evaluation points and the like are displayed.

(Diagnostic method for concrete structures)

FIG. 5 is a plan view showing a specimen 5A produced as a concrete structure.
As the specimen 5A, assuming an adhesive post-installed anchor bolt for fixing the ceiling plate of a tunnel, a plurality of (eg, 18) adhesive resins are placed on a rectangular parallelepiped concrete 1A (eg, 1800 mm in width and 900 mm in length). Capsule-type anchor bolts 10A were embedded upward under different construction conditions. On the plane of concrete 1A, three types (No. 1, No. 2, No. 3) of anchor bolts 10A with different construction conditions were arranged in the vertical direction, and six anchor bolts 10A were arranged in the horizontal direction. The arrangement interval of each anchor bolt 10A is, for example, 250 mm. Each of the six anchor bolts 10A arranged in the lateral direction is numbered 1 to 6 in the 90° direction from the 270° side when viewed from above.

Of the three types (No. 1, No. 2, No. 3) of anchor bolts 10A, type No. 1 is the one with standard construction, type No. 2 is the one with the tip capsule remaining because the prepared hole 2A is deep, and type No. 3 has an expected anchoring length of 40 mm. and the amount of resin is small.

6(a) to 6(c) are enlarged side views showing an adhesive resin-encapsulated anchor bolt 10A embedded in the specimen 1A of FIG.

The anchor bolt 10A of type No. 1 shown in FIG. 6(a) is of standard construction, is made of, for example, chromium molybdenum steel (SNB7), has a diameter M of 16 mm and a length of is 230 mm. The anchor bolt 10A is fixed by the resin 3 of the resin capsule 4 in the pilot hole 2A (for example, 19 mm in diameter and 130 mm in depth) drilled in the concrete 1A.

The anchor bolt 10A of type No. 2 shown in FIG. 6(b) has the same dimensions as those of FIG. 6(a). Capsule 4 remains and is in a construction failure state.

The anchor bolt 10A of type No. 3 and the pilot hole 2A shown in FIG. 6C have the same dimensions as those of FIG. In order to simulate a defect in which the amount of resin decreases, the resin capsule 4 was intentionally cut and applied, so the resin 3 decreased, for example, an air pocket 6 was generated at 90 mm on the far side, and the estimated fixing length was reduced to 40 mm. It's becoming For example, a polyethylene sleeve and a foamed urethane sheet are wrapped around the air pool 6 to prevent the resin 3 from entering, thereby reproducing the poor construction condition.

FIG. 7 is a plan view showing another specimen 5B produced as a concrete structure.
As another test body 5B, post-installed metal anchor bolts for fixing the jet fan of the tunnel are assumed, and a plurality of (for example, 18) bolts are installed in the rectangular parallelepiped concrete 1B having the same dimensions as the rectangular parallelepiped concrete 1A in FIG. were embedded downward under different construction conditions. On the plane of concrete 1B, three types (No. 4, No. 5, No. 6) of anchor bolts 10B with different construction conditions are arranged in the vertical direction, and six anchor bolts 10B are arranged in the horizontal direction. The arrangement interval of each anchor bolt 10B is the same as in FIG. Each of the six anchor bolts 10B arranged in the lateral direction are numbered 1 to 6 in the 90° direction from the 270° side, as in FIG. The anchor bolts 10B numbered 1 to 3 have already been subjected to a pull-out test, and the three anchor bolts 10B numbered 4 to 6 will be diagnosed in each construction.

Of the three types (No. 4, No. 5, No. 6) of anchor bolts 10B, type No. 4 is a standard construction type (for example, the diameter φ of the pilot hole 2B is 35 mm and the depth is 155 mm), and type No. 5 is the diameter φ of the pilot hole 2B. is enlarged (for example, 38 mm), and Type No. 6 has a deep pilot hole 2B (for example, 170 mm), resulting in poor driving.

FIGS. 8(a) to 8(c) are enlarged side views showing a metal sleeve expandable anchor bolt 10B embedded in the specimen 5B of FIG.
The anchor bolt 10B of type No. 4 shown in FIG. 8(a) is of standard construction, and is made of, for example, a stainless steel material (SUS material) similarly to FIG. is 230 mm. Anchor bolts 10B are fixed by expanded metal sleeves 13 in pilot holes 2B (for example, 35 mm in diameter and 155 mm in depth) drilled in cuboid concrete 1B. For example, the upper end of the metal sleeve 13 is located at a position of 55 mm from the upper end of the exposed portion before being driven, and is at a depth of 0.7 mm to 2.8 mm from the surface of the concrete 1B after being driven.

The anchor bolt 10B of type No. 5 shown in FIG. 8B has the same dimensions as those shown in FIG. The upper end of the metal sleeve 13 has a depth of, for example, 6.6 mm to 8.7 mm from the surface of the concrete 1B.

The anchor bolt 10B of type No. 6 shown in FIG. 8(c) has the same dimensions as those shown in FIG. ), but the depth is as large as 170 mm), resulting in a defective driving condition.

The diagnostic method of the first embodiment using the specimens 5A and 5B shown in FIGS. 5 and 7 will be described below.

FIG. 9 is a flow chart showing a concrete building diagnostic method according to the first embodiment of the present invention.
In this concrete building diagnostic method, the following elastic wave generation processing ST1, reception processing ST2, data processing ST3, evaluation index processing ST4, and evaluation point assignment processing ST5 are sequentially performed.

(I) Elastic wave generation processing ST1
The specimens 5A and 5B shown in FIGS. 5 and 7 and the diagnostic apparatus shown in FIGS. 1 to 4 are prepared in advance, and the electromagnetic pulse power supply 25 is turned on to start the process. Power supply power is output from the electromagnetic pulse power supply 25 and supplied to the coil unit 24 . A pulse current is output from the coil unit 24 , and this pulse current flows through the coils 23 arranged at the anchor bolt exposed portions 12 and 15 . Then, a pulse magnetic field is generated from the coil 23, and the anchor bolt 10 (that is, the anchor bolt 10A of adhesive resin capsule type) vibrates to generate an elastic wave. In the case of the metal sleeve expandable anchor bolt 10B, since a static magnetic field is generated from the magnet 27, a pulse magnetic field is generated from the coil 23 with this static magnetic field as a bias, and the anchor bolt 10B vibrates to generate an elastic wave. occurs. The generated elastic waves pass through the concrete 1A, 1B and propagate to the concrete surface around the anchor bolts 10A, 10B.

(II) Reception processing ST2
The elastic waves propagated to the concrete surface are received by a plurality of sensors 21-1 to 21-4 arranged on the concrete surface. At the same time, the sensors 22 arranged at the anchor bolt exposed portions 12 and 15 receive elastic waves generated from the anchor bolt exposed portions 12 and 15 . The analog reception signals received by the sensors 21-1 to 21-4 and the analog reception signal received by the sensor 22 are sent to the AD converter 26. In the AD conversion device 26, for example, the analog reception signals supplied from the plurality of sensors 21-1 to 21-4 are converted into digital signals at a sampling frequency of 100 MHz and a data length of 10 ms, and a first reception signal S26a is converted. While outputting to the information processing device 30 , the analog reception signal supplied from the sensor 22 is converted into a digital signal, and the second reception signal S<b>26 b is output to the information processing device 30 .

(III) Data processing ST3
Data processing including time-domain waveform processing ST3a by the time-domain waveform processing unit 40 and frequency spectrum processing ST3b by the frequency spectrum processing unit 50 in the time-domain waveform processing unit 40 and the frequency spectrum processing unit 50 in the information processing device 30. ST3 is performed.

That is, in the time-axis waveform processing unit 40, the input first received signal S26a and second received signal S26b have high-frequency components removed by the 2 MHz LPF 41, and are converted into a CSV file by the file conversion unit 42. In the converted CSV file, the data is thinned out by the thinning unit 43, the low frequency components are removed by the HPF 44 of 2.5 kHz, and the first time axis waveform S44a and the second time axis waveform S44a with a sampling frequency of 1 MHz and a data length of 10 ms are obtained. A signal of S44b is generated and output to the waveform evaluation index section 60. FIG.

Furthermore, in the frequency spectrum processing section 50, the input first received signal S26a and second received signal S26b are fast Fourier transformed by the FFT section 51 to generate a frequency spectrum signal. The signal of the generated frequency spectrum is converted into a CSV file by the file conversion unit 52, and the signal of the first frequency spectrum S52a and the second frequency spectrum S52b with a frequency range of 0 to 50 kHz, a frequency resolution of 200 Hz, and FFT points of 250 are converted. It is generated and output to spectrum evaluation index section 70 .

(IV) Evaluation index processing ST4
The waveform evaluation index unit 60 and the spectrum evaluation index unit 70 perform the evaluation index processing ST4 as follows, and determine the waveform energy ratio SR between the waveform energy of the first time-axis waveform and the waveform energy of the second time-axis waveform. 1 evaluation index EI1, and a second evaluation index EI2 of each waveform duration TC until attenuation to less than 10% of the maximum amplitude of each wave crest in the first time-axis waveform and second time-axis waveform. , a third evaluation index EI3 of the centroid frequency SC of each spectrum in the first frequency spectrum and the second frequency spectrum, a fourth evaluation index EI4 of the standard deviation SD of the spectrum, and a fifth evaluation index of the number of spectrum peaks SP EI5, and so on. The second evaluation index EI2 of the waveform duration TC of each of the first time-axis waveform and the second time-axis waveform is, for example, the maximum amplitude of the wave crest. It may be the maximum amplitude in the waveform. Also, the waveform continuation time TC until attenuation is not limited to less than 10% and may be set to a predetermined amplitude.

10(a), (b) and FIGS. 11(a), (b) are explanatory diagrams of the waveform energy ratio SR in FIG.
10(a) is a waveform diagram showing the waveform energy of the concrete surface, and FIG. 10(b) is a waveform diagram showing the waveform energy of the anchor bolt exposed portions 12 and 15. FIG. 10A and 10B, the horizontal axis is time (ms) and the vertical axis is voltage (V).
FIG. 11(a) is a schematic diagram showing the magnitude relationship of waveform energy in the case of standard construction, and FIG. 11(b) is a schematic diagram showing the magnitude relationship of waveform energy in the case of defective construction.

但し、Ｅ；エネルギー（Ｖ２）
ｙｉ；サンプリングする各点における振幅（Ｖ）
ｎ；サンプリング数 The waveform energy is represented by the sum of the squares of the amplitude values at each sampling point, and can be calculated by Equation (1).

However, E; energy (V2)
yi; Amplitude (V) at each sampling point
n; number of samples

The waveform energy ratio SR is the ratio between the waveform energy of the concrete surface and the waveform energy of the anchor bolt exposed portions 12 and 15, as shown in Equation (2). That is, the values of the waveform energy in the four directions of the concrete surface are divided by the values of the waveform energy of the anchor bolt exposed portions 12 and 15, respectively.

As shown in FIGS. 11(a) and 11(b), the waveform energy ratio SR has a greater change than the waveform energy alone, so it becomes easy to determine the defective construction. Since the waveform energy ratio SR is an index based on the ratio of the waveform energy of the concrete surface and the waveform energy of the anchor bolt exposed portions 12 and 15 in defective construction, comparison with standard construction is easy, and in some cases comparison with standard construction is possible. It may no longer be necessary.

FIG. 12 is an explanatory diagram of the waveform duration TC in FIG. 2, where the horizontal axis is time and the vertical axis is voltage.
The waveform continuation time TC is the time t2 from the vibration initial arrival time t1 until the vibration is attenuated to, for example, less than 10% of the maximum amplitude (MAX) of the wave crest in the time axis waveform. Therefore, it can be seen that the anchor bolts 10A and 10B are loosely restrained due to poor installation, and the convergence time of the waveform on the time axis is lengthened.

FIG. 13 is an explanatory diagram of the center-of-gravity frequency SC of the spectrum in FIG. 2, where the horizontal axis is the frequency f (kHz) and the vertical axis is the spectrum intensity. A case where the extraction level EL is 30% of the maximum peak Pmax is shown.

FIGS. 14A and 14B are explanatory diagrams of the standard deviation SD of the spectrum in FIG. 2, where the horizontal axis is frequency and the vertical axis is spectrum intensity. In order to reduce the influence of small peaks and broadening of the tail portion, the index calculation range is set to a maximum of ±2.5 kHz. FIG. 14(a) shows the calculation range when the cut width F30 of the 30% line of the maximum peak Pmax is 5 kHz or more (F30≧5 kHz). FIG. 14(b) shows the calculation range when the cut width F30 of the 30% line of the maximum peak Pmax is smaller than 5 kHz (F30<5 kHz).

FIG. 15 is an explanatory diagram of the number SP of spectrum peaks in FIG. 2, where the horizontal axis is the frequency f and the vertical axis is the spectral intensity H(f).

As shown in FIGS. 13 to 15, the center-of-gravity frequency SC of the spectrum, the standard deviation SD of the spectrum, and the number of peaks SP of the spectrum are obtained as shown in (a) to (d) below.

(a) As shown in FIG. 13, extract a spectrum having an intensity equal to or greater than a certain ratio (for example, the extraction level EL is 30% of the maximum peak) with respect to the maximum peak of the spectrum.
(b) Extract the lowest frequency spectrum Sfmin from (a).
(c) For the lowest frequency spectrum Sfmin extracted in (b), as shown in FIG. (3) and the spectrum standard deviation SD is calculated by Equation (4). When F30<5 kHz, the center of gravity SC of the spectrum is calculated by Equation (3) and the standard deviation SD of the spectrum is calculated by Equation (4) within the range of F30.
(d) As shown in FIG. 15, count the number of peaks SP of the spectrum extracted in (a). That is, the points at which the slope of the graph shown in FIG. 15 changes from positive to negative are counted.

(V) Evaluation point giving process ST5
The evaluation point imparting unit 80 performs an evaluation point imparting process ST5 for imparting evaluation points for evaluation of poor construction based on a threshold to at least one of the five evaluation indexes EI1 to EI5. In the evaluation point giving process ST5, evaluation points are given to each of the plurality of evaluation indexes EI1 to EI5 based on the threshold values as in (i) and (ii) below.

(i) Set a threshold for bad (NG) evaluation for each evaluation index. At this time, even if the data of the standard construction varies, the setting should be made with a margin so as not to evaluate as defective (NG) as much as possible.
(ii) Give evaluation points based on thresholds for each evaluation index EI1 to EI5.

FIG. 16 is a diagram for explaining how to assign evaluation points to the waveform energy ratio SR (=concrete surface/anchor bolt exposed portion) in FIG.
The threshold value of the waveform energy ratio SR obtained for each of the sensors 21-1 to 21-4 is 0.5 or less (SR≤0.5). Since the waveform energy ratio SR is calculated using measurement data from two locations, the sensors 21-1 to 21-4 on the concrete surface and the sensor 22 on the anchor bolt exposed portions 12 and 15, the four sensors 21- 2 points each for 1 to 21-4. Therefore, the evaluation points total 8 points.

FIG. 17 is a diagram for explaining how to assign evaluation points to the waveform duration time TC in FIG.
The threshold of the waveform duration TC obtained for each of the sensors 21-1 to 21-4 is 8 (ms) or more (TC≧8 (ms)), and the threshold of the waveform duration TC obtained by the sensor 22 is 3 (ms). The above is (TC≧3 (ms)). Since the evaluation points of the anchor bolts 10A and 10B were set to 4 points so as to be the same as the total of the sensors 21-1 to 21-4 on the concrete surface, the total evaluation points are 8 points.

The evaluation points for the center of gravity SC of other spectra, the standard deviation SD of spectra, and the number of peaks SP of spectra are the same as the waveform duration TC.

FIG. 18 is a diagram showing a summary of thresholds and evaluation points.
FIG. 18 shows thresholds for the center of gravity SC of the spectrum, the standard deviation SD of the spectrum, and the number of peaks SP of the spectrum. FIG. 19 shows an example of the processing result of the evaluation point giving processing ST5.

FIG. 19 is a diagram showing an evaluation point graph of examples of defective construction (excluding standard construction). The horizontal axis of FIG. 19 indicates the types (1) to (3) of construction failures and the differences A to O between the anchor bolts 10A and 10B, and the vertical axis indicates the total evaluation points.

The types (1) to (3) of installation defects and the differences A to O between the anchor bolts 10A and 10B are as follows.

Type (1) is the No. 5 metal-based sleeve expandable anchor bolt 10B shown in FIGS. With this No. 5 anchor bolt 10B, the supporting force is insufficient due to the enlarged pilot hole diameter. Among them, the difference A of the anchor bolt 10B is the anchor bolt 10B of No. 5 No. 4 (No. 5_4), and the point is 34. Similarly, the difference B is No. 5 of No. 5 (No. 5_5) and has 30 points, and the difference C is No. 5 and No. 6 (No. 5_6) and has 30 points.
Here, the calculation result of the short-term allowable load of the constructed metal sleeve expandable anchor bolt 10B was 72.4 kN (kilonewtons). The 10B pull-out strength of the anchor bolt was 8.1 to 12.0 kN for No. 5 above. The pullout resistance of No. 5 was found to be extremely low in comparison with the short-term allowable load calculation result (72.4 kN).

The type (2) is the No. 2 adhesive resin encapsulated anchor bolt 10A shown in FIGS. In this No. 2 anchor bolt 10A, since the pilot hole is deep, the tip capsule remains and the adhesive is insufficiently filled. Among them, the difference D of the anchor bolt 10A is the anchor bolt 10A of No. 2 No. 1 (No. 2_1), and the point is 25. Similarly, differences E, F, G, H, and I are No2_2, No2_3, No2_4, No2_5, and No2_6 anchor bolts 10A, respectively, and points are 24, 27, 28, 24, and 25, respectively. Here, the calculation result of the short-term allowable load of the constructed adhesive resin-encapsulated anchor bolt 10A was 36.9 kN. The yield strength when the anchor bolt 10A was actually pulled out was 51.6 to 80.8 kN for No.2 and 0.7 to 3.0 kN for No.3. The pullout resistance of No. 3 was found to be extremely low in comparison with the short-term allowable load calculation result (36.9 kN). In addition, the pull-out yield strength of No. 1 subjected to the standard construction shown in FIG.

The type (3) is the No. 3 adhesive resin encapsulated anchor bolt 10A shown in FIGS. In this No. 3 anchor bolt 10A, the amount of adhesive is insufficient due to the decrease in resin. Among them, the difference J of the anchor bolt 10A is the anchor bolt 10A of No. 3 No. 1 (No.3_1), and the point is 18. Similarly, differences K, L, M, N, and O are No3_2, No3_3, No3_4, No3_5, and No3_6 anchor bolts 10A, respectively, and points are 14, 17, 20, 18, and 17, respectively.

As shown in FIG. 19, if the total evaluation points are displayed as a graph, numerical value, or the like as the processing result of the evaluation point imparting process ST5, the defective installation of the anchor bolts 10A, 10B can be evaluated simply, accurately, and quantitatively. be able to.

That is, in the diagnostic process of FIG. 9, the time domain waveforms S44a and S44b and the frequency spectra S52a and S52b are obtained from the reception signals S26a and S26b received by the sensors 21-1 to 21-4 and 22 by the data processing ST3. Then, the waveform energy ratio SR, the waveform duration TC, the centroid frequency SC of the spectrum, the standard deviation SD, and the number of peaks SP are obtained by the evaluation index process ST4, and the evaluation points are given to them by the evaluation point giving process ST5. The poor installation of the anchor bolts 10A and 10B was evaluated from the total evaluation points. Compared with the above-described pull-out strength, especially in No. 5 of the metal sleeve expandable anchor bolt 10B and No. 3 of the adhesive resin encapsulated anchor bolt 10A, the pull-out strength was remarkably low.
Therefore, in the point graph shown in FIG. 19, for example, if evaluation point 10 is set as an evaluation boundary for detecting defects with extremely low pull-out yield strength, construction defects with pull-out yield strength of 60% or less of the standard construction will be detected. It becomes possible to As a result, a defect with a remarkably low pull-out yield strength could be reliably detected.

(Effect of Example 1)
According to the concrete structure diagnostic method and the diagnostic device of the first embodiment, the received signals of the sensors 21-1 to 21-4 received on the concrete surface, the anchor bolt exposed portion 12 exposed from the concrete surface, A plurality of evaluation indices EI1 to EI4 are converted into points from the received signal of the sensor 22 received at 15, and defective installation of the anchor bolts 10A and 10B is evaluated. Accurate and quantitative evaluation is possible.

(Modification)
The present invention is not limited to the first embodiment described above, and various forms of use and modifications are possible. Examples of usage patterns and modifications include the following (a) to (d).

(a) In the information processing apparatus 30 of FIG. 2, the time domain waveform processing unit 40 and the frequency spectrum processing unit 50 may be changed to other configurations and processing contents.
(b) A portion of the five evaluation indices EI1 to EI5 may be converted into points to evaluate poor installation of the anchor bolts 10A and 10B.
(c) The threshold in FIG. 18 may be changed as appropriate, and if the threshold is brought close to the value of the standard construction product, even anchor bolts with minor construction defects can be strictly detected in consideration of safety. If the threshold value is set apart from the value of the standard construction product, it is possible to detect only anchor bolts with severe construction defects without detecting anchor bolts with minor construction defects.
(d) The conductive rods to be evaluated are not limited to the anchor bolts 10A and 10B, and may be other conductive members such as reinforcing bars, to which the present invention can be applied.

(Modification of Concrete Building Diagnosis Device)
Next, a second embodiment of the diagnostic apparatus for concrete structures will be described.
FIG. 20 is a functional block diagram showing a diagnostic apparatus for concrete structures according to the second embodiment. The information processing apparatus 32 of FIG. 20 differs from the information processing apparatus 30 of FIG. 2 in that an index section for evaluating the correlation coefficient CF(E16) of the spectrum is added to the spectrum evaluation index section. Since other configurations are the same as those of the diagnostic device 30 of the information processing device, description of other configurations is omitted.

FIG. 21 is a flow chart showing a concrete building diagnosis method according to the second embodiment of the present invention. The flowchart showing the diagnostic method for the concrete building 32 in FIG. 21 differs from the flowchart according to the diagnostic method in the first embodiment shown in FIG. E16) is added. Other flows are the same as in FIG.

The correlation coefficient CF(E16) of the frequency spectrum in the diagnostic method of the second embodiment will be explained.
(1) First, find the average value of the frequency spectrum acquired when the standard construction is performed N times.
The frequency spectrum is an FFT waveform obtained by fast Fourier transforming the data of the time domain waveform processing.
FIG. 22 is a diagram showing FFT waveforms when 1 to N standard constructions to be evaluated are measured, and FIG. 23 is an average of FFT waveforms when 1 to N standard constructions to be evaluated are measured. It is a figure which shows. 22 and 23, the vertical axis is amplitude, and the horizontal axis is frequency (kHz). The average value of the frequency spectrum can be obtained by calculating the average amplitude value at each frequency.

(2) Next, the evaluation target is a construction different from the standard construction, such as construction failure, and the average (x) of the FFT waveform of the standard construction and the correlation coefficient (CF) of the evaluation target spectrum amplitude (y) are calculated as follows. It is calculated by Equation (5) or Equation (6).

Here, (n) is the number of data regarding frequency.

Regarding the correlation coefficient of Example 2, the results of measuring the adhesive resin encapsulated anchor bolt 10A and the metal sleeve expandable anchor bolt 10B of Example 1 will be described.
FIG. 24 is a diagram showing a comparison of the FFT waveform between the FFT waveform when standard construction is performed on a new evaluation object and the average (x) of the FFT waveform for standard construction.
The average (x) of the FFT waveform of the standard construction is No. 1, No. 13, No. Twenty-five three evaluation subjects were obtained. The new evaluation target is No. 1_4 (concrete surface 0°). As shown in FIG. 24, the correlation coefficient is 0.94, indicating a high correlation between the FFT waveform of the new evaluation target and the average (x) of the FFT waveform of the standard construction.

FIG. 25 shows the FFT waveform of the construction failure, with the new evaluation object as the construction failure, and the evaluation object No. 1, No. 13, No. 25 is a diagram showing a comparison with the average (x) of the FFT waveform of the standard construction set to 25. FIG. The new evaluation target for poor construction is No. 4_4 (0° concrete surface). As shown in FIG. 25, the correlation coefficient is 0.42, and it can be seen that the average (x) of the FFT waveform of the defective construction, which is the new evaluation target, and the FFT waveform of the standard construction shows a low correlation. .

FIG. 26 is a diagram showing a summary of threshold values and evaluation points in Example 2, and FIG. As shown in FIG. 26, in the adhesive resin-encapsulated anchor bolt 10A, the threshold for concrete (4 directions) is 0.78 or less, and the points are 1 point each. Moreover, the threshold value of the anchor bolt 10A is 0.89 or less, and the points are 4 points.

Furthermore, as shown in FIG. 26, in the metal-based sleeve expandable anchor bolt 10B, the threshold for concrete (4 directions) is 0.66 or less, with 1 point each. The threshold value of the anchor bolt 10B is 0.77 or less, and the points are 4 points.

As shown in FIG. 27, the evaluation targets A, B, and C of the metal sleeve expandable anchor bolt 10B scored 42, 38, and 38, respectively.

As shown in FIG. 27, the evaluation targets D, E, F, G, H, and I of the adhesive resin-encapsulated anchor bolt 10A had points of 33, 32, 35, 36, 32, and 33, respectively. .

In addition, the evaluation targets J, K, L, M, and N of the adhesive resin-encapsulated anchor bolt 10A had points of 26, 22, 25, 28, 26, and 25, respectively.

In Example 2, the construction failure points with the correlation coefficient further added as the sixth evaluation index in FIG. 27 are more than the construction failure points by the first to fifth evaluation indexes shown in FIG. It turned out that it is possible to diagnose construction defects more easily by adding numbers.

Next, Example 3 of the diagnostic apparatus for concrete structures will be described.
28A and 28B are schematic configuration diagrams showing a diagnostic apparatus for concrete structures according to Embodiment 3 of the present invention, in which FIG. ), and FIG. 29 is a functional block diagram showing a configuration example of the information processing device 30 in FIG. is.

The diagnostic apparatus for concrete structures according to the third embodiment has first sensors 21-1 to 21-4 and a second sensor 22, whereas the diagnostic apparatus according to the second embodiment shown in FIGS. , consists of only one sanse. That is, the difference is that only the first sensor 21 described in the second embodiment or only the second sensor 22 is provided. As shown in FIG. 28, in the diagnostic apparatus for concrete structures of the third embodiment, the first sensors 21-1 to 21-4 are omitted, and only the second sensor (that is, the conductor rod sensor) 22 is arranged. However, the other configurations are the same, so description thereof will be omitted. In this case, since there are no first sensors 21-1 to 21-4, there is no first received signal S26a. When using the first sensor instead of the second sensor 22, the number of the first sensors is not limited to four, and may be at least one or more.

FIG. 29 is a functional block diagram showing a diagnostic apparatus for concrete structures according to the third embodiment.
Since the information processing device 32 of FIG. 29 does not have the above-described first received signal S26a and the first time-axis waveform S44a, the processing associated with the first received signal, that is, the waveform energy ratio SR (E11 ) is omitted and there is no signal of the first frequency spectrum S52a derived from the first received signal S26a. , is the same as the information processing device 32 of FIG. 20, so the description thereof is omitted.

Along with this, as shown in FIG. 30, the example of the measurement conditions using the adhesive resin encapsulated anchor bolt 10A is different from the case of FIG. omitted and only the second sensor (conductor bar sensor) 22 is used.

Also, as shown in FIG. 31, in the example of the measurement conditions using the metal-based sleeve expandable anchor bolt 10B, compared to the case of FIG. 4, the first sensors 21-1 to 21-4 are similarly omitted. and only the second sensor (conductor bar sensor) is used.

FIG. 32 is a flow chart showing a concrete building diagnosis method according to the third embodiment of the present invention. 32 differs from the flowchart of the second embodiment shown in FIG. 21 in that the waveform energy ratio SR (E11) of the first evaluation index is omitted in the evaluation index processing. This is the second to sixth evaluation indexes. Other flows are the same as in FIG.

FIG. 33 is a diagram showing a summary of thresholds and evaluation points in Example 3. FIG. As shown in FIG. 33, for example, in the spectral correlation coefficient CF, the threshold value of the anchor bolt is 0.89 or less and the number of points is 4 in the adhesive resin encapsulated anchor bolt 10A. In the anchor bolt 10B, the anchor bolt threshold is 0.77 or less, and the points are 4 points. Evaluation point graphs of construction failures obtained from the second to sixth evaluation indices yielded results similar to those of FIG.
As a result, it is found that the faulty installation of the post-installed anchor bolt can be evaluated by omitting the first sensors 21-1 to 21-4 and detecting the elastic wave only at the bolt head by the second sensor 22. bottom.

Next, a fourth embodiment of the diagnostic apparatus for concrete structures will be described.
The diagnostic apparatus for concrete structures according to the fourth embodiment of the present invention has substantially the same configuration as the diagnostic apparatus for concrete structures according to the third embodiment shown in FIG.
FIG. 34 is a functional block diagram showing the configuration of the information processing device 34 used in the concrete building diagnosis device according to the fourth embodiment, and FIG. 35 is a schematic diagram showing the first configuration example of the information processing device in FIG. is.
The information processing device 34 of FIG. 34 includes a determination processing unit 90 instead of the waveform evaluation index unit 60, the spectrum evaluation index unit 70, and the evaluation point imparting unit 80 in the information processing device 32 according to the third embodiment shown in FIG. ing. Here, the time domain waveform processing unit 40 and the frequency spectrum processing unit 50 have the same configuration as in the case of the information processing device 32 in FIG.

The judgment processing unit 90 is composed of a neural network having an input layer 91, an intermediate layer 92, and an output layer 93, which receives data processing results as input signals, as shown in FIG. 35, for example. It works with programs for neural networks. The intermediate layer 92 is inserted between the input layer 91 and the output layer 93 and receives a teacher signal for discrimination from the input layer 91 . The output layer 93 outputs an output signal. As software for creating such a neural network program, MATLAB (registered trademark) (manufactured by MathWorks, USA), which is numerical analysis software, was used.

In the determination processing section 90, calculations in steps 1 and 2 below are performed.
(Step 1)
The input layer 91 receives the frequency spectrum S52b from the frequency spectrum processing unit 50 as input. The input layer 91 has 250 input units 91a corresponding to each input signal, and each input signal is input to each input unit.

The determination processing unit 90 performs a calculation represented by the following formula ( ₇ ) from the input signals x ₁ , x ₂ , . . . , x _k , . Integers from intermediate values h _i (i is from 1 to n) are obtained. Here, n can be an integer less than 250, for example.

Here, k is an integer from 1 to 250 and i is an integer from 1 to n.
Also, w ¹ _ki is a weighting factor. This weighting factor is determined by the machine learning result of the neural network, which will be described later, that is, the learning result.

The intermediate layer 92 has n intermediate units 92a corresponding to respective intermediate values, and each intermediate value h _i described above is input to each intermediate unit 92a.
The determination processing unit 90 calculates the output signal y _m (m is 1 or 2) from the intermediate value h _i input to each intermediate unit 92 a of the intermediate layer 92 by performing the calculation of the following formula (8).
Also, w ² _km is a weighting factor, which is similarly set by machine learning of a neural network, which will be described later. Since the output signal is m=1 or 2, specifically, the output signals are y ₁ and y ₂ and are written as [y ₁ y ₂ ].

FIG. 36 is a schematic diagram showing an example of a teacher signal used in the evaluation analysis apparatus of FIG. 35, and FIG. 37 is a schematic diagram showing a discrimination result in the evaluation analysis apparatus of FIG. _{[ 1} 0], and [0] in the case of construction failure 1].

(Step 2)
As a result of such neural network machine learning, when the frequency spectrum S52b based on the signal received by the conductor rod sensor 22 due to an unknown post-installed anchor bolt is input to the input layer 91, the determination processing unit 90 generates an intermediate signal and outputs By calculating the signal in two steps, the output signal [y ₁ y ₂ ] is output as shown in FIG. That is, as an output signal, [1 0] is output in the case of standard construction, and [0 1] is output in the case of defective construction. An output based on the learning result of such a neural network is also called a judgment.
Therefore, by checking the output signal [y ₁ y ₂ ] from the output layer 93 of the determination processing unit 90, it is possible to determine whether the installation status of the post-installed anchor bolts is standard installation, or the installation is performed by adhesive back filling or front filling. Whether it is defective or not can be easily determined, and whether it is standard construction or construction defect can be easily identified.

Here, in step 1 described above, specifically, the two weighting factors w ¹ _ki and w ² _km are set by neural network learning as follows.
38 is a graph showing input data during known standard construction that is input to the input layer during learning in the evaluation analysis apparatus of FIG. 35, and FIG. 40 is a graph showing input data during known back filling construction input to , and FIG. 40 shows input data during known front filling construction input to the input layer during learning in the evaluation analysis device of FIG. graph. Known data used for these learnings are also called teacher data.
That is, as known data for learning, for example, the frequency spectrum of anchor bolts of standard construction shown in FIGS _. The determination processing unit 90 calculates and determines weighting coefficients w ¹ _ki and w ² _km so that 1, y ₂ =0, that is, [1 0] is obtained.

Similarly, for example, the frequency spectrum of the poorly installed (adhesive back filling) anchor bolt shown in FIGS. is input to the input layer ₉₁ , _and the determination processing unit 90 sets the weighting coefficients Calculate and determine w ¹ _ki and w ² _km . At this time, only one combination of the weighting coefficients w ¹ _ki and w ² _km is determined so as to simultaneously satisfy all of the two patterns of construction failures of standard construction, adhesive back filling, and adhesive front filling.
Here, the greater the number of input known data, the higher the accuracy of the determined weighting factors w ¹ _ki and w ² _km .

Through such neural network machine learning, the determination processing unit 90 learns the difference in the shape of the frequency spectrum between standard construction and defective construction.
Specifically, in the case of installation failure due to back filling of the adhesive, the peak frequency of the frequency spectrum described above is low and the peak is sharp. It learns that there are more than one, that the peak frequency is lower, that the peak is sharper, and so on.
Then, based on such learning results, when unknown, that is, unlearned data is input, the determination processing unit 90 performs calculations using the two-stage weighting coefficients w ¹ _ki and w ² _km obtained from the learning results. Then, output signals y ₁ and y ₂ [y ₁ y ₂ ] are output from the output layer 93 according to the construction state.

Next, a specific example of step 2 will be described.
41 is a graph showing input data at the time of unknown standard construction input to the input layer during identification in the evaluation analysis apparatus of FIG. 35, and FIG. FIG. 43 is a graph showing input data during unknown back filling construction input to , and FIG. 43 shows input data during unknown front filling construction input to the input layer during identification in the evaluation analysis device of FIG. graph.
That is, as unknown data for evaluation, for example, the frequency spectrum of anchor bolts of standard construction shown in FIGS. 41 (A) and (B), FIGS. B) and the frequency spectrum of the poorly installed anchor bolt are input to the input layer 91 .
As a result, the determination processing unit 90 performs the calculations of the above-described formulas (7) and (8) for each frequency spectrum using the two-stage weighting coefficients w ¹ _ki and w ² _km obtained from the learning results. and output the output signal [y ₁ y ₂ ].
In this case, the output signal [1 0] is obtained in the frequency spectra of different constructions in FIGS. 41(A) and (B), indicating the standard construction.
Also, in the frequency spectrums of different constructions shown in FIGS. 42A and 42B and FIGS.
In this way, regarding unknown data, the output signal [y ₁ y ₂ ] output from the output layer 93 of the determination processing unit 90 is a correct construction evaluation regarding the construction status of post-installed anchor bolts for all data. It was confirmed that

(Modification)
FIG. 44 shows a modification of the determination processing section 90 shown in FIG.
44, the determination processing unit 90 is configured differently from the determination processing unit 90 shown in FIG. 34 only in that the output layer 93 outputs three output signals [y ₁ y ₂ y ₃ ]. . By using three output signals, the judgment processing unit 90 can distinguish between three installation states, that is, standard installation, installation failure due to back filling of adhesive, and installation failure due to front filling of adhesive. In this case, the determination processing unit 90 calculates the output signal _ym ( _m is from 1 to integer of 3).

FIG. 45 is a schematic diagram showing an example of a teacher signal used in the evaluation analysis device of FIG. 44, and FIG. 46 is a schematic diagram showing a discrimination result in the evaluation analysis device of FIG. As shown in FIG. 45, the output signal is 1 for _y1 , 0 for _{y2 , and} 0 for _y3 , that is, [1 0 0].
In addition, in the case of a construction failure due to back filling of the adhesive, the teacher signal is set to 0 for _y1 , 1 for _{y2 , and} 0 for _y3 , that is, [0 1 0].
Furthermore, in the case of an installation failure due to prefilling of the adhesive, the teacher signal is set to 0 for _y1 , 0 for _{y2 , and} 1 for _y3 , that is, [0 0 1].
Here, the determination processing unit 90 performs the calculations of the above-described formulas (7) and (9) for each frequency spectrum, and outputs an output signal [y ₁ y ₂ y ₃ ]V corresponding to the teacher signal. Two-stage weighting factors w ¹ _ki and w ² _km are determined as follows.
As a result of such neural network machine learning, when the frequency spectrum S52b based on the signal received by the conductor rod sensor 22 due to an unknown post-installed anchor bolt is input to the input layer 91, the determination processing unit 90 generates an intermediate signal and outputs By calculating the signal in two steps, the output signals _y1 , _y2 , _y3 , ie, [ _{y1 y2 y3} ] are output as shown in FIG.

That is, as shown in FIG. 46, when the unknown post-installed anchor bolt is standard installed, the output signal is [1 0 0].
If the unknown post-installed anchor bolt is installed incorrectly due to adhesive backfilling, the output signal becomes [0 1 0].
Furthermore, if the unknown post-installed anchor bolt is defective due to pre-filling, the output signal becomes [0 0 1].
Therefore, by checking the output signal [y ₁ y ₂ y ₃ ] from the output layer 93 of the determination processing unit 90, the installation status of the post-installed anchor bolts, that is, standard installation, or installation failure due to back filling of adhesive, can be determined. Alternatively, it can be easily determined whether the adhesive is improperly filled in front of the adhesive.

Also in this case, the two weighting factors w ¹ _ki and w ² _km described above are specifically set by neural network learning as follows.
That is, as known data for learning, for example, the frequency spectrum of anchor bolts of standard construction shown in FIGS. ], the determination processing unit 90 calculates and determines the weighting coefficients w ¹ _ki and w ² _km .
Similarly, for example, the frequency spectrum of anchor bolts with installation defects (adhesive back filling) shown in FIGS. As obtained, the determination processing unit 90 calculates and determines the weighting factors w ¹ _ki and w ² _km .
43A and 43B, for example, the frequency spectrum of anchor bolts with poor installation (adhesive front filling) is input to the input layer 91, and an output signal [0 0 1] is obtained as a teacher signal. The determination processing unit 90 calculates and determines the weighting coefficients w ¹ _ki and w ² _km so that At this time, only one combination of the weighting coefficients w ¹ _ki and w ² _km is determined so as to satisfy all three patterns of construction at the same time: standard construction, defective construction due to back filling of adhesive, and defective construction due to front filling of adhesive.
Thereby, the determination processing unit 90 can calculate and determine the weighting factors w ¹ _ki and w ² _km based on these known data, that is, the teacher data and teacher signal.
Here, the greater the number of input known data, the higher the accuracy of the determined weighting factors w ¹ _ki and w ² _km .

By substituting the weighting coefficients w ¹ _ki and w ² _km obtained in this way into the above-mentioned formulas (7) and (9), it is possible to evaluate the installation failure of the unknown post-installed anchor bolt.
That is, as unknown data for evaluation, for example, the frequency spectrum of anchor bolts of standard construction shown in FIGS. The frequency spectrum of the defective anchor bolt and the frequency spectrum of the poorly installed anchor bolt due to adhesive pre-filling shown in FIGS.
As a result, the determination processing unit 90 performs the calculations of the above-described formulas (7) and (9) for each frequency spectrum, and outputs output signals [y ₁ y ₂ y ₃ ], respectively.

In this case, the output signal is [1 0 0] in the frequency spectra of FIGS. 41A and 41B, and this value indicates the standard construction.
In addition, in the frequency spectra of FIGS. 42A and 42B, the output signal is [0 1 0], and from this value, it can be seen that the installation failure is due to the back filling of the adhesive.
Furthermore, in the frequency spectra of FIGS. 43A and 43B, the output signal is [0 0 1], and from this value, it can be seen that the installation failure was due to the prefilling of the adhesive.
As described above, regarding unknown data, the output signal [y ₁ y ₂ y ₃ ] output from the output layer 93 of the determination processing unit 90 is obtained as It was confirmed that the correct construction evaluation was performed.

In this way, according to the diagnosis device according to the fourth embodiment, as in the case of the third embodiment, construction evaluation is performed only by one sensor (conductor rod sensor) arranged at the bolt head of the post-installed anchor bolt. As a result, the cost can be reduced, the burden of signal processing can be reduced, and construction evaluation can be performed in a shorter period of time. In Example 4, it was determined whether there were three construction situations, that is, standard construction, construction failure due to adhesive back filling, and construction failure due to adhesive front filling. The total number of layers may be appropriately selected.

In addition, in Example 4, it is possible to perform installation evaluation without processing the frequency spectrum and without setting a threshold value, and the type of installation failure such as adhesive back filling or adhesive front filling can be determined. can also be identified.

Furthermore, in Examples 2 and 3, by using the correlation coefficient, similar shapes with respect to the shape of the frequency spectrum are judged as the same spectrum, but in Example 4, by machine learning of the neural network Since construction evaluation is performed including the size of the frequency spectrum, construction evaluation can be performed with higher accuracy.
The frequency spectra of Examples 2 to 4, for example, the training data shown in FIGS. 38 to 43 and the data from unknown construction were frequency spectra in the frequency range up to 50 kHz. A predetermined frequency range that can be extracted may be used. For example, frequencies may range up to 20 kHz.

In the fourth embodiment described above, based on the frequency spectrum S52b, the determination processing unit 90 evaluates the construction of the post-installed anchor bolt based on the calculation results of the formulas (7) to (9). Not limited to this, based on the time-axis waveform S44b, the determination processing unit 9
It is also possible to evaluate the installation of post-installed anchor bolts.
Also in this case, similarly, by determining the weighting factor by machine learning of the neural network so that the teacher signal is output according to the teacher data, the evaluation of the post-installed anchor bolt by the determination processing unit 90 can be performed with high accuracy. can be done.
In Examples 1 to 4, the adhesive resin encapsulated anchor bolt 10A and the metal sleeve expandable anchor bolt 10B are shown as examples, but it goes without saying that other anchor bolts can also be applied.

Next, Example 5 of the diagnostic apparatus for concrete structures will be described.
FIG. 47 is a functional block diagram showing a diagnostic apparatus for concrete structures according to the fifth embodiment. The information processing apparatus 36 of FIG. 47 differs from the information processing apparatus 30 of FIG. 29 in that an index section for evaluating the cross-correlation function XF(E17) of the time-axis waveform is added to the waveform evaluation index section 60. . Since other configurations are the same as those of the information processing apparatus 30, description of the other configurations is omitted.

FIG. 48 is a flow chart showing a concrete building diagnostic method according to the fifth embodiment of the present invention. 48 differs from the flowchart of the third embodiment shown in FIG. 32 in that in the evaluation index process ST4, in addition to the second to sixth evaluation indexes, The point is that the cross-correlation function XF (E17) of the time domain waveform is added. Other flows are the same as in FIG.

The cross-correlation function XF(E17) of the time-domain waveform in the diagnostic method of the fifth embodiment will be described.
(1) First, find the average value of the time-axis waveforms acquired when the standard construction is performed N times.
The time domain waveform is data for time domain waveform processing.
FIG. 49 shows the time axis waveform when measuring 1 to N standard constructions to be evaluated, and FIG. 50 shows the average of the time axis waveforms when measuring 1 to N standard constructions to be evaluated. show. 49 and 50, the vertical axis is amplitude and the horizontal axis is time (msec). The average value of the time domain waveform can be obtained by calculating the average value of the amplitudes at the same time in each time domain waveform.

(2) Next, assuming that the object of evaluation is a construction different from the standard construction, such as a defective construction, the average (f(i)) of the time-axis waveform of the standard construction and the evaluation target time-axis waveform (g(i)) A correlation function (R) and a normalized cross-correlation function (Rnorm) are calculated by the following equations (10) and (11), respectively. The cross-correlation function (R) and normalized cross-correlation function (Rnorm) are so-called discrete systems and can be calculated by digital operations.

ただし、ｎは時間軸波形における１０^－６秒毎の ｉ のデータ数、ｉは１からｎの整数であり、時間軸波形における所定間隔毎のポイントを順次に数字で示している。τはｉに対するずれであり、位相を示す変数である。

However, n is the number of i data for every 10 ⁻⁶ seconds in the time axis waveform, i is an integer from 1 to n, and points at predetermined intervals in the time axis waveform are sequentially indicated by numbers. τ is the shift with respect to i and is a variable indicating the phase.

ただし、ｎは時間軸波形における１０^－６秒毎の ｉ のデータ数、ｉは１からｎの整数、τはｉに対するずれ、ｆバー，ｇバーは、それぞれｆ（ｉ），ｇ（ｉ）の平均値である。
ここで、数式（１１）は、数式（１０）に対して、関数ｆ（ｉ）及びｇ（ｉ－τ）の大きさで正規化した相互相関関数（Ｒｎｏｒｍ）を求めるものである。正規化した相互相関関数（Ｒｎｏｒｍ）は、０～１の値となる。

where n is the number of i data in every 10 ^-6 seconds in the time axis waveform, i is an integer from 1 to n, τ is the deviation from i, f bar and g bar are f(i) and g(i), respectively is the average value of
Here, formula (11) obtains a cross-correlation function (Rnorm) normalized by the magnitude of functions f(i) and g(i−τ) with respect to formula (10). The normalized cross-correlation function (Rnorm) has a value between 0 and 1.

Regarding the cross-correlation function of Example 5, FIG. 51(B) shows the average ( f) shows the cross-correlation function R according to equation (10) between The horizontal axis of FIG. 51A is time (msec). i is a score, and its interval is 1×10 ⁻⁶ seconds, and in the case of the drawing, it shows from 1×10 ⁻⁶ seconds to 10000×10 ⁻⁶ seconds. The horizontal axis of FIG. 51(B) indicates the number of data points for τ, from -10000 points to +10000 points.
For example, in FIG. 51 and FIG. 52A described later, n corresponds to time 10 msec. On the other hand, τ represents a deviation with respect to i, and is, for example, a value every 10 ⁻⁶ seconds (horizontal axis in FIGS. 51 and 52 (B) and (C) described later).
The maximum absolute value of the cross-correlation function R shown in FIG. 51(B) is 5.40. It can be seen that there is a correlation.
Further, FIG. 51(C) shows the relationship between the time-axis waveform of the standard construction and the average (f) of the time-axis waveform of the standard construction shown in FIG. ) shows the normalized cross-correlation function Rnorm, the maximum absolute value of which is 0.84. It can be seen that a high correlation is shown.

On the other hand, FIG. 52B shows the average (f) of the time-axis waveform of the new construction failure and the time-axis waveform of the standard construction shown in FIG. The maximum absolute value is 0.68, and the time-axis waveform of the defective construction and the time-axis waveform of the standard construction, which are new evaluation targets, are shown. It can be seen that it shows a low cross-correlation with the average.
Further, FIG. 52(C) shows the relationship between the time-axis waveform of the defective construction shown in FIG. ) shows the normalized cross-correlation function Rn, and the maximum absolute value is 0.22. It can be seen that the cross-correlation is low.

In this case, since the normalized cross-correlation function Rnorm takes a value between 0 and 1, the level of the cross-correlation function can be easily grasped.
Also, if there is a large difference in amplitude between the time-axis waveform in the case of defective construction, which is the target of cross-correlation evaluation, and the time-axis waveform in standard construction, the cross-correlation function R becomes large, and if the time-axis waveforms are similar, Although there is a possibility that it may be misjudged, in the case of the normalized cross-correlation function Rnorm, a more accurate correlation function is obtained without being affected by the amplitude level of the time-axis waveform in the case of construction defects to be evaluated. As a result, the determination of construction failure can be made more accurately.

FIG. 53 shows a summary of threshold values and evaluation points in Example 5, and FIG.
As shown in FIG. 53, for example, in the cross-correlation function XF of the waveform on the time axis, the anchor bolt threshold value is 0.98 or less, the points are 4 points, and the expansion of the metal sleeve is In the anchor bolt 10B of the formula, the anchor bolt threshold is 0.83 or less, and the points are 4 points. Evaluation point graphs of construction failures obtained from the second to seventh evaluation indices shown in FIG. 54 yielded results similar to those of FIG.
As a result, it was found that the defective installation of the post-installed anchor bolts can be evaluated not by the spectrum evaluation by the FFT waveform of the frequency spectrum in Examples 1 to 3, but by the waveform evaluation by the time-axis waveform.
Furthermore, when calculating the cross-correlation function between the average (f(i)) of the time-axis waveform of the standard construction and the evaluation target time-axis waveform (g(i)), the difference time i is calculated while shifting by τ, so even if there is a phase shift, the correlation can be confirmed without being greatly affected by this phase shift, and construction defects of concrete structures can be evaluated more accurately. Is possible.
Although the adhesive resin capsule type anchor bolt 10A and the metal sleeve expandable type anchor bolt 10B are shown as examples in the fifth embodiment, it is needless to say that the present invention can be applied to other anchor bolts.

1, 1A, 1B: Concrete 2, 2A, 2B: Pilot hole 3: Resin 4: Resin capsule 5A, 5B: Specimen 6: Air reservoir 10, 10A, 10B: Anchor bolt 11, 14: Fixed part 12, 15: Exposed portion 13: Metal sleeve 21-1 to 21-4: First sensor 22: Second sensor (conductor bar sensor)
23: Ring 24: Coil unit 25: Electromagnetic pulse power supply 27: Magnet 30, 32, 34, 36: Information processing device 40: Time axis waveform processing unit 50: Frequency spectrum processing unit 60: Waveform evaluation index unit 80: Evaluation point assignment Section 90: Judgment Evaluation Section ST1: Elastic Wave Generation Processing ST2: Reception Processing ST3: Data Processing ST3a: Time-axis Waveform Processing ST3b: Frequency Spectrum Processing ST4: Evaluation Index Processing ST5: Evaluation Point Assignment Processing

Claims (9)

In a method for diagnosing a concrete structure in which a conductor rod is partially embedded in concrete,
placing a coil on an exposed portion of the conductor rod that is exposed from the concrete surface;
one sensor placed on the concrete surface around the conductor rod or on the exposed portion of the conductor rod;
An elastic wave generation process in which a pulse current is passed through the coil to generate a pulse magnetic field from the coil, and the pulse magnetic field is applied to the conductor rod to generate an elastic wave;
a receiving process for receiving the elastic wave with the sensor and generating a received signal;
data processing for obtaining a time domain waveform by filtering the received signal and a frequency spectrum by Fourier transforming the received signal and outputting a data processing result;
Judgment processing for judging a defective construction of the conductor rod on the concrete based on the data processing result;
has
The determination process is
Evaluation index processing for obtaining a plurality of evaluation indexes based on the data processing result;
an evaluation point assignment process of assigning evaluation points for evaluation of construction defects to the plurality of evaluation indicators based on a threshold value;
has
In the evaluation index process,
an evaluation index of the waveform duration until the maximum amplitude of the time-domain waveform is attenuated to a predetermined amplitude;
an evaluation index of the center-of-gravity frequency of the spectrum in the frequency spectrum;
an evaluation index of the standard deviation of the spectrum;
an evaluation index for the number of peaks in the spectrum;
an evaluation index of the correlation coefficient between the spectrum of defective construction and the average spectrum of a plurality of standard constructions;
seeking
In the evaluation point granting process,
An evaluation index of the waveform duration, an evaluation index of the center-of-gravity frequency of the spectrum, an evaluation index of the standard deviation of the spectrum, an evaluation index of the number of peaks of the spectrum, and a ratio between the spectrum of the defective construction and the average spectrum of a plurality of standard constructions. A method for diagnosing a concrete structure , comprising: diagnosing an evaluation index of a correlation coefficient by assigning the evaluation points based on the threshold value and summing the evaluation points . The evaluation index of the waveform duration time expresses an event in which the fixation of the conductor rod is loosened due to the construction failure, and the convergence time of the time-axis waveform is prolonged,
The evaluation index of the center-of-gravity frequency of the spectrum expresses an event in which the frequency spectrum shifts to the low frequency side due to the construction failure,
The evaluation index of the standard deviation of the spectrum expresses the degree of concentration of the frequency spectrum that appears due to the construction failure,
The evaluation index of the number of spectral peaks expresses a plurality of frequency spectra that appear due to the defective construction when the conductor rod has a complicated structure,
The evaluation index of the correlation coefficient between the spectrum of the defective construction and the average spectrum of a plurality of standard constructions is the correlation coefficient between the frequency spectrum appearing due to the defective construction and the FFT waveform obtained by averaging the spectrum of a plurality of standard constructions. which expresses
The diagnostic method for a concrete structure according to claim 1, characterized in that: In the evaluation index process,
Furthermore, the evaluation index of the cross-correlation function between the time-axis waveform of defective construction and the average time-axis waveform of multiple standard constructions is obtained ,
In the evaluation point granting process,
Further, the evaluation point is given based on the threshold value to the evaluation index of the cross-correlation function between the time-axis waveform of the defective construction and the average time-axis waveform of a plurality of standard constructions. Item 2. The method for diagnosing a concrete structure according to Item 1. The evaluation index of the cross-correlation function between the time-axis waveform of the defective construction and the average time-axis waveform of the plurality of standard constructions is normalized by the magnitude of the two functions that constitute the cross-correlation function. 4. The method for diagnosing a concrete structure according to claim 3 . The evaluation index of the cross-correlation function between the time-axis waveform of the construction defect and the average time-axis waveform of a plurality of standard constructions is that the greater the degree of the construction defect, the more the difference in the shape of the time-axis waveform, and the smaller the cross-correlation function. It expresses the event that
5. The method for diagnosing a concrete structure according to claim 3 or 4 , characterized in that: 6. The method for diagnosing a concrete structure according to claim 1 , wherein said conductive rod is a conductive member including a post-installed anchor bolt. In a diagnostic device for a concrete structure in which a conductor rod is partially embedded in concrete,
a coil disposed on an exposed portion of the conductor rod that is exposed from the concrete surface;
a sensor placed on the concrete surface around the conductor rod or on the exposed portion of the conductor rod;
elastic wave generating means for generating a pulse magnetic field from the coil by applying a pulse current to the coil, and applying the pulse magnetic field to the conductor rod to generate an elastic wave;
a waveform receiving unit that receives the elastic wave with the sensor and generates a received signal;
a time domain waveform processing unit that filters the received signal to obtain a time domain waveform;
a frequency spectrum processing unit that Fourier transforms the received signal to obtain a frequency spectrum;
a waveform evaluation index unit that obtains an evaluation index of waveform duration based on the processing result of the time domain waveform processing unit;
a spectrum evaluation index unit that obtains an evaluation index of the centroid frequency of the spectrum, an evaluation index of the standard deviation of the spectrum, and an evaluation index of the number of peaks of the spectrum based on the processing result of the frequency spectrum processing unit;
An evaluation index of the waveform duration, an evaluation index of the center-of-gravity frequency of the spectrum, an evaluation index of the standard deviation of the spectrum, an evaluation index of the number of peaks of the spectrum, and a correlation between the spectrum of defective construction and the average spectrum of a plurality of standard constructions. a concrete structure diagnosis apparatus, comprising: an evaluation point assigning unit that assigns evaluation points for evaluation of construction defects based on a threshold value to the evaluation index of the relational coefficient, and totals the evaluation points. In a diagnostic device for a concrete structure in which a conductor rod is partially embedded in concrete,
a coil disposed on an exposed portion of the conductor rod that is exposed from the concrete surface;
a sensor placed on the concrete surface around the conductor rod or on the exposed portion of the conductor rod;
elastic wave generating means for generating a pulse magnetic field from the coil by applying a pulse current to the coil, and applying the pulse magnetic field to the conductor rod to generate an elastic wave;
a waveform receiving unit that receives the elastic wave with the sensor and generates a received signal;
a time domain waveform processing unit that filters the received signal to obtain a time domain waveform;
a frequency spectrum processing unit that Fourier transforms the received signal to obtain a frequency spectrum;
Based on the processing result of the time-axis waveform processing unit, the waveform evaluation obtains the evaluation index of the waveform duration and the evaluation index of the cross-correlation function between the time-axis waveform of the defective construction and the average time-axis waveform of the plurality of standard constructions. an index part;
a spectrum evaluation index unit that obtains an evaluation index of the centroid frequency of the spectrum, an evaluation index of the standard deviation of the spectrum, and an evaluation index of the number of peaks of the spectrum based on the processing result of the frequency spectrum processing unit;
An evaluation index of the waveform duration, an evaluation index of the cross-correlation function between the time-axis waveform of the defective construction and the average time-axis waveform of a plurality of standard constructions, an evaluation index of the centroid frequency of the spectrum, and an evaluation of the standard deviation of the spectrum. Evaluation points for evaluating construction defects are given based on a threshold value to the index, the evaluation index of the number of peaks of the spectrum, and the evaluation index of the correlation coefficient between the spectrum of construction defects and the average spectrum of a plurality of standard constructions. and an evaluation point giving unit that totals the
A diagnosis device for a concrete structure, comprising: 9. The concrete structure diagnosis apparatus according to claim 7 , wherein the conductor rod is a conductive member including a post-installed anchor bolt.

Images