Prestress assessment system and prestress assessment program

Abstract

This invention provides a prestress evaluation system and a prestress evaluation program that can evaluate the tension state of PC steel materials. [Solution] The prestress evaluation system comprises an acquisition means for acquiring waveform data including vibration noise of PC steel materials, and an evaluation means for evaluating the tension state of the PC steel materials based on the waveform data acquired by the acquisition means. The evaluation means evaluates the tension state of the PC steel materials based on previously acquired waveform data and the waveform data acquired by the acquisition means. This makes it possible to evaluate the tension state of the PC steel materials from the waveform data.

Description

[Technical Field]

[0001] The present invention relates to a prestress evaluation system and a prestress evaluation program for evaluating the tension state of PC steel materials. [Background technology]

[0002] Traditionally, PC structures have required re-grouting due to insufficient grouting. In areas with insufficient grouting, the PC steel is exposed to the air, leading to corrosion. As a result, the corroded PC steel, which is already prestressed, may fracture and protrude due to cross-sectional defects or hydrogen embrittlement fracture, raising concerns about potential damage to third parties.

[0003] As a grout refilling method, for example, the PC-Rev method is being used, which allows for grout refilling to be completed with a small diameter of φ15.5 in order to minimize damage to the structure. This method is specialized for grout refilling and has a high refilling rate, as it takes into account factors such as the estimation of voids and the injection of small amounts of grout.

[0004] Furthermore, in the grout refilling method, a perforation is provided that extends to the PC steel. Through this perforation, the PC steel can be visually inspected using an endoscope or similar device. Therefore, by visually inspecting the PC steel, information such as the grout filling status and the degree of corrosion of the PC steel can be obtained.

[0005] However, while it is possible to observe the surface condition of PC steel by visually inspecting it, there is no established method for determining whether the PC steel is fractured or whether prestress has been introduced. If information on the presence or absence of prestress and fracture of the PC steel could be obtained, subsequent repair methods would also change.

[0006] For example, if prestressing has not been introduced, re-tensioning of the PC steel member in question may be considered. If this re-tensioning is difficult, it is desirable to reinforce the external cables of the PC steel member in question to restore the prestressing required in the design. The same applies if the PC steel member in question is fractured. Therefore, if the fracture and prestressing conditions can be detected during the process of simply re-injecting grout, it would be useful information for the service planning of the PC structure. One example of a method for detecting fractures of PC steel members is cited in Patent Document 1.

[0007] Patent Document 1 discloses a stress measurement system for a reinforced concrete member subjected to stress in the x-direction, comprising: a first strain gauge attached to the side surface of the concrete member in the x-direction; a second strain gauge attached in the y-direction, which is perpendicular to the x-direction; strain detection means for detecting strain εx in the x-direction via the first strain gauge and strain εy in the y-direction via the second strain gauge from a concrete piece cut out so as to include the side surface to which the first and second strain gauges are attached; and calculation means for calculating elastic strains Δεx and e in the x-direction from the strains εx and εy detected by the strain detection means, and further calculating effective stresses σx and e in the x-direction. [Prior art documents] [Patent Documents]

[0008] [Patent Document 1] Japanese Patent Publication No. 2008-268123 [Overview of the project] [Problems that the invention aims to solve]

[0009] However, the stress measurement system for concrete members disclosed in Patent Document 1 is not intended to evaluate the prestress state of PC steel. Therefore, the stress measurement system for concrete members disclosed in Patent Document 1 has the problem that it cannot evaluate the tension state of PC steel.

[0010] Therefore, the present invention was devised in view of the above-mentioned problems, and its objective is to provide a prestress evaluation system and a prestress evaluation program that can evaluate the tension state of PC steel materials. [Means for solving the problem]

[0011] The prestress evaluation system according to the first invention is characterized by comprising an acquisition means for acquiring waveform data including vibration noise of PC steel materials, and an evaluation means for evaluating the tension state of the PC steel materials based on the waveform data acquired by the acquisition means.

[0012] The prestress evaluation system according to the second invention is characterized in that, in the first invention, the evaluation means evaluates the tension state of the PC steel material based on waveform data acquired in advance and waveform data acquired by the acquisition means.

[0013] The prestress evaluation system according to the third invention is characterized in that, in the first invention, the evaluation means evaluates the tension of the PC steel material based on two or more waveform data, each containing vibration sounds of PC steel materials with different tension states, which have been acquired in advance, and the waveform data acquired by the acquisition means.

[0014] The prestress evaluation system according to the fourth invention is characterized in that, in the first invention, the acquisition means acquires waveform data including vibration sound of the PC steel material that is not filled with grout and is excited from the outside, through a perforation drilled from the surface of the PC structure to the PC steel material.

[0015] The prestress evaluation system according to the fifth invention, in the first invention, the evaluation means evaluates the tension state of the PC steel material based on waveform data that has been acquired in advance and subjected to short-time Fourier analysis or wavelet analysis, and waveform data that has been acquired by the acquisition means and subjected to short-time Fourier analysis or wavelet analysis.

[0016] The prestress evaluation system according to the sixth invention, in the first invention, the evaluation means evaluates the tension state of the PC steel material based on the waveform data acquired by the acquisition means, by referring to an evaluation model that has been learned using the waveform data for learning acquired in advance as learning data.

[0017] The prestress evaluation program according to the seventh invention, in the first invention, causes a computer to execute an acquisition step of acquiring waveform data including the vibration sound of the PC steel material, and an evaluation step of evaluating the tension state of the PC steel material based on the waveform data acquired in the acquisition step.

Advantages of the Invention

[0018] According to the first to seventh inventions, the tension state of the PC steel material is evaluated based on waveform data. Thereby, even when the grout filling situation is opaque and the degree of contact between the cable sheath and the PC steel material is not uniform, it becomes possible to evaluate the tension state of the PC steel material.

[0019]

[0020] ​In particular, according to the third invention, the evaluation means evaluates the tension of the PC steel material based on two or more waveform data including the vibration sounds of PC steel materials with different pre-acquired tensions and the waveform data acquired by the acquisition means. Therefore, by comparing a plurality of waveform data with different tensions and the waveform data acquired by the acquisition means, it becomes possible to evaluate the tension of the PC steel material.

[0021] In particular, according to the fourth invention, the acquisition means acquires waveform data including the vibration sound of the PC steel material with unfilled grout vibrated from the outside through the perforation part. As a result, for example, it becomes possible to acquire waveform data including the vibration sound of the PC steel material with unfilled grout determined using the perforation part provided for the PC-Rev method. Therefore, it becomes possible to select locations where evaluation of the tension state is necessary, and it becomes possible to efficiently evaluate the tension state of the PC steel material.

[0022] In particular, according to the fifth invention, the evaluation means evaluates the tension state of the PC steel material based on the waveform data acquired in advance and subjected to short-time Fourier analysis or wavelet analysis, and the waveform data acquired by the acquisition means and subjected to short-time Fourier analysis or wavelet analysis. As a result, since it becomes possible to visualize the time change of the intensity for each frequency of the waveform data, it becomes possible to evaluate the tension state of the PC steel material with higher accuracy.

[0023] In particular, according to the sixth invention, the evaluation means refers to an evaluation model and evaluates the tension state of the PC steel material based on the waveform data acquired by the acquisition means. As a result, by using a learned model, it is possible to evaluate the tension state of the PC steel material with higher accuracy.

Brief Description of the Drawings

[0024] [Figure 1] FIG. 1 is a block diagram showing the overall configuration of a prestress evaluation system to which the present invention is applied. [Figure 2] FIG. 2 is a diagram showing an example of a concrete structure. [Figure 3]Figure 3 shows an example of a vibration jig. [Figure 4] Figure 4 shows a specific example of the configuration of the evaluation device. [Figure 5] Figure 5 is a flowchart showing the processing operation of a prestress evaluation system to which the present invention is applied. [Figure 6] Figure 6(a) shows an example of waveform data when the tension force is 0σpy. Figure 6(b) shows an example of waveform data when the tension force is 0.7σpy. [Figure 7] Figure 7(a) shows an example of the short-time Fourier analysis results for waveform data when the tension force is 0σpy. Figure 7(b) shows an example of the short-time Fourier analysis results for waveform data when the tension force is 0.7σpy. [Figure 8] Figure 8(a) shows an example of the short-time Fourier analysis results for waveform data when the tension force is 0.05σpy. Figure 8(b) shows an example of the short-time Fourier analysis results for waveform data when the tension force is 0.25σpy. Figure 8(c) shows an example of the short-time Fourier analysis results for waveform data when the tension force is 0.5σpy. [Modes for carrying out the invention]

[0025] The prestress evaluation system to which the present invention is applied will be described in detail below with reference to the drawings.

[0026] Figure 1 is a block diagram showing the overall configuration of a prestress evaluation system 1 to which the present invention is applied. The prestress evaluation system 1 evaluates the prestress within the cable sheath 71 of a concrete structure 7. As shown in Figure 1, the prestress evaluation system 1 comprises a sound collection unit 8, a converter 9 connected to the sound collection unit 8, an evaluation device 2 connected to the converter 9, and a database 3 connected to the evaluation device 2.

[0027] Figure 2 shows an example of a concrete structure 7. As shown in Figure 2, the concrete structure 7 is a PC structure such as a bridge, elevated bridge, or building in which PC steel members 72 and cable sheaths 71 are arranged inside. The concrete structure 7 may also have perforations 73 drilled during the PC-Rev construction process, as shown in Figure 2.

[0028] The cable sheath 71 is arranged with PC steel members 72 under tension inside, spaced apart from the inner wall surface of the cable sheath 71. Incidentally, in this embodiment, a post-tensioned concrete structure 7 is described as an example. In such a case, the cable sheath 71 is placed in the concrete structure 7, then concrete is filled and hardened, and then the PC steel members 72 are inserted into the cable sheath 71 and tensile stress is applied. In this case, the PC steel members 72 are subjected to a tension force σ that exhibits a constant tensile stress. py This is given. Subsequently, a post-tensioned concrete structure 7 is formed by filling the cable sheath 71 with grout and allowing it to harden. The tension state of the PC steel 72 indicates, for example, the tension force of the PC steel 72. The tension state of the PC steel 72 also indicates whether the tension force of the PC steel 72 is normal or abnormal, or whether it is above or below a threshold.

[0029] PC steel materials 72 are steel materials, including PC steel bars or numerous PC steel wires, used to apply prestress to the concrete structure 7. Furthermore, PC steel materials 72 have the characteristic of vibrating as a chord when struck.

[0030] As shown in Figure 3, the perforated portion 73 is a hole drilled from the surface of the concrete structure 7 to the PC steel material 72. The diameter A of the perforated portion 73 is, for example, 15.5 mm, but is not limited to this and may be any diameter.

[0031] The sound collection unit 8 is, for example, a microphone or the like that excites the PC steel material 72 using an excitation jig 6 and collects sound data including the vibration sound of the PC steel material 72. The sound collection unit 8 transmits the collected sound data to the conversion device 9.

[0032] The vibration jig 6 is a jig capable of vibrating the PC steel material 72 with a constant force. The vibration jig 6 comprises, for example, a spring section 61 that can be pushed out with a constant force, a vibration section 62 that is pushed out by the spring section 61, collides with the object, and vibrates with a constant force, and an opening 63 that is cylindrical in shape and encloses the vibration section 62, and has a hole for releasing the vibration sound of the object vibrated by the vibration section 62 to the outside. Furthermore, the vibration jig 6 is configured so that the vibration section 62 vibrates the center of the perforation section 73 by inserting the opening 63 into the perforation section 73. This makes it possible to vibrate the PC steel material 72 while minimizing the influence of burrs from the cable sheath 71 remaining on the PC steel material 72.

[0033] The conversion device 9 is composed of electronic devices such as a PC (personal computer), smartphone, tablet, or wearable device. This conversion device 9 converts the acquired sound data into waveform data that includes data indicating the intensity and frequency for at least each frequency in order to evaluate the grout filling status in the cable sheath 71, and transmits the converted waveform data to the evaluation device 2.

[0034] The conversion device 9 may transmit the acquired sound data to the evaluation device 2 as waveform data without any special analysis. For example, if the conversion device 9 acquires waveform data showing frequency and intensity for each frequency as sound data, it may transmit the sound data as is as waveform data to the evaluation device 2 without any special analysis.

[0035] Furthermore, the conversion device 9 may analyze the acquired sound data, convert it into waveform data, and transmit it to the evaluation device 2. For example, if waveform numerical data showing the relationship between time and amplitude is acquired as sound data, the conversion device 9 may analyze this data, convert it into waveform data showing frequency and intensity for each frequency, and then transmit it to the evaluation device 2. By analyzing the sound data consisting of waveform numerical data, the conversion device 9 will process it appropriately to be optimal for evaluating the degree of grout filling inside the cable sheath 71.

[0036] For example, the conversion device 9 may convert the acquired waveform numerical data showing the relationship between time and amplitude into waveform data showing the relationship between frequency and intensity for each frequency by applying an FFT (Fast Fourier Transform) transform. Alternatively, the conversion device 9 may convert the acquired waveform numerical data showing the relationship between time and amplitude into waveform numerical data showing the relationship between time and frequency by applying a Fourier transform, then remove a specific frequency region, and further apply an inverse Fourier transform to convert it into waveform numerical data showing the relationship between time and amplitude while retaining the specific frequency region. Furthermore, the conversion device 9 may further apply a Fourier transform to this waveform numerical data showing the relationship between time and amplitude while retaining the specific frequency region to convert it into waveform data showing the relationship between frequency and intensity for each frequency while retaining the specific frequency region. Additionally, the conversion device 9 may convert the acquired waveform numerical data showing the relationship between time and amplitude into waveform data showing the relationship between intensity and frequency, such as a power spectrum.

[0037] The conversion device 9 may apply a wavelet transform to the acquired sound data, which is waveform numerical data, to convert it into waveform numerical data that shows the relationship between time and frequency. By applying a wavelet transform, it is possible to retain the time characteristics that would be lost in the Fourier transform. Therefore, by applying a wavelet transform, the data is converted into waveform numerical data that shows the relationship between time and frequency.

[0038] The conversion device 9 may convert the acquired waveform numerical data, which is sound data, into waveform numerical data showing a cepstrum by taking the logarithm of the power spectrum obtained by Fourier transforming the power spectrum and then inverse Fourier transforming it. Furthermore, the conversion device 9 may extract waveform numerical data showing the spectral envelope, which is a low-order cepstrum, and the spectral fine structure, which is a high-order cepstrum, from the waveform numerical data showing the cepstrum. For example, the waveform numerical data showing the spectral envelope may define the cepstrum order. This cepstrum order can take any value, such as 20 or 100. Furthermore, the coefficients of each cepstrum order may be extracted from the waveform numerical data showing the spectral envelope. In addition, the conversion device 9 may convert the acquired waveform image into waveform numerical data showing formants, which are peaks obtained when the amplitude extracted from a certain time domain is converted into the frequency domain. When the peak frequency bands are numbered from lowest to highest as the first formant, second formant, etc., the waveform numerical data may, for example, show the relationship between the first formant and the second formant, or show the relationship between frequencies. The conversion device 9 may also perform an AFTE (Auditory filterbank temporal envelope) transform on the acquired waveform numerical data as sound data. The conversion device 9 may also perform a short-time Fourier analysis on the acquired waveform numerical data as sound data.

[0039] Thus, the conversion device 9 may convert the acquired sound data into two-dimensional waveform data as shown in Figure 6. The two-dimensional waveform data may represent a relationship between two of the following: time, amplitude, frequency, intensity, spectrum, cepstrum, formant, etc. Alternatively, the reciprocals of these may be taken.

[0040] Furthermore, the conversion device 9 may convert the acquired sound data into three-dimensional waveform numerical data, for example, by applying a spectrogram. The three-dimensional waveform numerical data may represent, for example, three relationships among time, amplitude, frequency, intensity, spectrum, cepstrum, formant, etc.

[0041] Incidentally, this conversion device 9 can display each sound data via a display unit, such as a display not shown. The conversion device 9 can also store this data in storage and, based on a command from the user, display this data on the display unit or write this data to a portable memory. The user can detach this portable memory from the conversion device 9 and carry it around freely. Furthermore, the conversion device 9 can also transfer this data to other electronic devices via a public communication network.

[0042] In this invention, the configuration of the conversion device 9 is not essential and may be omitted. In such cases, the sound data output from the sound collection unit 8 will be transmitted directly to the evaluation device 2.

[0043] Database 3 stores various types of data. Database 3 accumulates data sent via the public communication network or data entered by users of this system. Furthermore, based on requests from evaluation device 2, database 3 transmits this accumulated data to evaluation device 2.

[0044] The evaluation device 2 is composed of electronic devices such as a personal computer (PC), but it may also be implemented using any other electronic devices such as mobile phones, smartphones, tablet devices, and wearable devices. The user can obtain evaluation results of the tension state as a search solution using this evaluation device 2.

[0045] Figure 4 shows a specific configuration example of the evaluation device 2. This evaluation device 2 has a control unit 24 for controlling the entire evaluation device 2, an operation unit 25 for inputting various control commands via operation buttons, a keyboard, etc., a communication unit 26 for wired or wireless communication, a search unit 27 for searching for optimal design conditions, and a storage unit 28, represented by a hard disk, for storing programs for performing searches to be executed, all connected to an internal bus 21. Furthermore, a display unit 23, which acts as a monitor for actually displaying data, is connected to this internal bus 21.

[0046] The control unit 24 is a so-called central control unit for controlling each component implemented in the evaluation device 2 by transmitting control signals via the internal bus 21. Furthermore, the control unit 24 transmits various control commands via the internal bus 21 in response to operations performed via the operation unit 25.

[0047] The operation unit 25 is implemented via a keyboard or touch panel, and the user inputs execution commands to run the program. When an execution command is input by the user, the operation unit 25 notifies the control unit 24. Upon receiving this notification, the control unit 24 coordinates with the search unit 27 and other components to execute the desired processing operation. The operation unit 25 may also receive various types of data from the user.

[0048] The search unit 27 searches for the results of the tension state evaluation. In order to perform the search operation, the search unit 27 reads various data stored in the memory unit 28 and various data stored in the database 3 as necessary data. This search unit 27 may be controlled by artificial intelligence. This artificial intelligence may be based on any well-known artificial intelligence technology.

[0049] The display unit 23 is comprised of a graphics controller that generates a display image based on control from the control unit 24. This display unit 23 is implemented, for example, by a liquid crystal display (LCD).

[0050] When the storage unit 28 is configured as a hard disk, predetermined data is written to each address based on control by the control unit 24, and this data is read out as needed. Furthermore, the storage unit 28 stores a program for executing the present invention. This program is read and executed by the control unit 24.

[0051] The operation of the prestress evaluation system 1, which has the configuration described above, will now be explained. In this case, the processing operation flow of the prestress evaluation system 1 is shown in Figure 4. The detailed processing at each step in Figure 4 will be explained below.

[0052] First, in step S11, the evaluation device 2 acquires waveform data including vibration noise from the PC steel material 72. For example, the evaluation device 2 acquires waveform data including vibration noise from the PC steel material 72 of a concrete structure 7 whose tension state is to be evaluated by performing vibration using the vibration jig 6. In this case, the vibration jig 6 is inserted into the drilled section 73, and the vibration section 62 is made to collide with the PC steel material 72 using the inserted vibration jig 6, thereby vibrating the PC steel material 72. After this, the vibration noise from the PC steel material 72 emitted from the hole in the opening 63 is collected by the sound collection section 8, and the sound data is transmitted to the conversion device 9. Since this vibration noise differs depending on the tension state of the PC steel material 72, it is possible to evaluate the tension state from the vibration noise. For example, if the tension state of the PC steel material 72 is normal, this vibration noise remains as an afterglow after the collision sound between the vibration section 62 and the PC steel material 72. However, if the tension of the PC steel member 72 is abnormal, no vibration noise will be generated, and only collision noise will be generated. Also, in step S11, the evaluation device 2 may acquire waveform data including vibration noise of the PC steel member 72 generated by, for example, applying vibration using the vibration jig 6 to the PC steel member 72 in which the grout of the concrete structure 7 is not filled. Also, in step S11, the evaluation device 2 may acquire chord length information indicating the length of the chord of the PC steel member 72. In this case, the evaluation device 2 may acquire chord length information indicating the length of the chord of the PC steel member 72 between cavities confirmed from multiple drilled holes 73. Also, if the cavity section is wide, or if there is a limit to the number of drilled holes 73, and the length of the chord of the PC steel member 72 between cavities cannot be confirmed from the drilled holes 73, the chord length may be fixed by fixing both ends of the PC steel member 72 between cavities within the range that can be confirmed. In this case, the evaluation device 2 acquires chord length information indicating the length of the chord of the fixed PC steel member 72.

[0053] Next, in step S12, the conversion device 9 performs various analyses on the sound data collected by the sound collection unit 8 as needed, and processes the waveform data as needed to facilitate searching by the subsequent search device. Incidentally, if no analysis or processing is performed, this step S12 can be omitted. The conversion device 9 outputs the converted waveform data to the evaluation device 2.

[0054] FIG. 6(a) is a diagram showing an example of waveform data when the tensile force is 0σ py FIG. 6(b) is a diagram showing an example of waveform data when the tensile force is 0.7σ py FIG. 7(a) is a diagram showing an example of the short-time Fourier analysis result of waveform data when the tensile force is 0σ py FIG. 7(b) is a diagram showing an example of the short-time Fourier analysis result of waveform data when the tensile force is 0.7σ py FIG. 8(a) is a diagram showing an example of the short-time Fourier analysis result of waveform data when the tensile force is 0.05σ py FIG. 8(b) is a diagram showing an example of the short-time Fourier analysis result of waveform data when the tensile force is 0.25σ py FIG. 8(c) is a diagram showing an example of the short-time Fourier analysis result of waveform data when the tensile force is 0.5σ py As shown in FIG. 7, when comparing the case where the tensile force is 0σ py with the case where the tensile force is 0.7σ py it can be seen that the vibration sound at a specific frequency remains longer in time in the case where the tensile force is 0.7σ py In step S12, the conversion device 9 may convert the sound data into waveform data as shown in FIG. 6. Also, in step S12, the conversion device 9 may convert the sound data into waveform data obtained by short-time Fourier analysis or wavelet analysis as shown in FIGS. 7 and 8. Thereby, it becomes possible to show the time change of the intensity for each frequency, and it becomes possible to visualize the intensity for each frequency having a length of a certain time or more.

[0055] Next, in step S13, the evaluation device 2 evaluates the tension state of the PC steel material 72 based on the waveform data converted in step S12. The evaluation device 2 may evaluate the tension state of the PC steel material 72 as abnormal if the waveform data converted in step S12 does not contain an intensity of a specific frequency for a certain period of time or longer. For example, the evaluation device 2 evaluates the tension state of the PC steel material 72 based on, for example, previously acquired waveform data and the waveform data converted in step S12. In such a case, the evaluation device 2 may calculate the similarity between, for example, waveform data acquired in advance when the tension state of the PC steel material 72 is normal and the waveform data converted in step S12, and if the similarity is above a threshold, it may evaluate the tension state of the PC steel material 72 as normal, and if the similarity is below the threshold, it may evaluate the tension state of the PC steel material 72 as abnormal.

[0056] Furthermore, in step S13, the evaluation device 2 may evaluate the tension of the PC steel members 72 based on two or more waveform data sets, each containing vibration sounds of PC steel members 72 with different tensions or the same tension but different support points, and the waveform data converted in step S12. In such a case, as shown in Figures 7 and 8, the tension of the PC steel members 72 is set to 0σ in advance. py , 0.05σ py , 0.25σ py , 0.5σ py , 0.7σ pyIn this case, the similarity between the acquired waveform data and the waveform data converted in step S12 is calculated, and the tension of the waveform data with the highest similarity is evaluated as the tension of the PC steel member 72. Furthermore, even when the tension is the same, the vibration sound differs depending on the chord length of the PC steel member 72. Therefore, the tension of the PC steel member 72 may be evaluated based on two or more waveform data containing the vibration sound of PC steel members 72 with the same tension but different support points, and the waveform data converted in step S12. In such a case, based on the tension information acquired in step S11, waveform data with the same chord length may be selected from previously acquired waveform data, and the tension of the PC steel member 72 may be evaluated based on the selected waveform data and the waveform data converted in step S12.

[0057] Furthermore, in step S13, the evaluation device 2 may refer to an evaluation model that has been trained using previously acquired training waveform data as training data, and evaluate the tension state of the PC steel material 72. In this case, the evaluation device 2 may, for example, refer to an evaluation model that compresses the input waveform data and outputs restored data obtained by restoring the compressed waveform data, output restored data based on the waveform data converted in step S12, and evaluate the tension state of the PC steel material 72 based on the waveform data and the restored data.

[0058] The evaluation model is, for example, a model that compresses input waveform data and outputs restored data by restoring the compressed waveform data. The evaluation model has an encoder that compresses the input waveform and a decoder that restores the compressed waveform data. Alternatively, the evaluation model may be a pre-trained model generated by machine learning using previously acquired past training waveform data as training data. In this case, the evaluation model is generated by compressing the training waveform data, extracting the features of the compressed training waveform data, and accumulating the features of the training waveform data. This makes it possible for the evaluation model to compress input waveform data and restore the compressed waveform data based on the features of the training data. The evaluation model may be generated by machine learning, such as an AutoEncoder. The features of the training waveform data are data that can reproduce the input training waveform data. Furthermore, waveform data compression is the process of converting waveform data into lower-dimensional data. For example, waveform data may be represented by frequency and intensity for each frequency, in which case waveform data compression is the process of limiting the frequencies that have features. Furthermore, waveform data restoration is the process of converting waveform data into higher-dimensional data. For example, waveform data may be represented by frequency and frequency intensity, in which case waveform data reconstruction means converting the data from a specific frequency to the original frequency.

[0059] The training waveform data used for learning may, for example, be previously acquired historical waveform data. This training waveform data may also be, for example, artificially generated waveform data.

[0060] Furthermore, the training data may be classified into normal data and abnormal data based on the tension state of the PC steel 72. For example, the training data may be classified as normal data if the tension of the PC steel 72 is above a threshold, and abnormal data if it is below the threshold. Specifically, the threshold can be set to 0.5σ. py The tension is 0.7σ py The training waveform data obtained in this case is considered normal data, and the tension force is 0.25σ.py If this is the case, the training waveform data obtained may be treated as abnormal data. Furthermore, the training data may be normalized training waveform data.

[0061] In the evaluation model, for example, the number of nodes in the input layer and the output layer are the same, and the number of nodes in the hidden layer is less than the number of nodes in the input layer. For example, the number of nodes in the input layer and the output layer may be 4750, and the number of nodes in the hidden layer may be 8. In the evaluation model, each node in the input layer is connected to each node in the hidden layer by three or more levels of weighting, and each node in the hidden layer is connected to each node in the output layer by three or more levels of weighting. In addition, multiple hidden layers may be provided between the input layer and the hidden layer, or between the hidden layer and the output layer. Each node in the hidden layer is connected to each node in the adjacent layer by three or more levels of weighting.

[0062] The weights indicate the degree of accuracy between each node. For example, for a node in the input layer that takes training waveform data A as input, the weights are connected to two nodes in the hidden layer with correlation degrees of 30% and 60%, respectively. In such a case, a node with a higher weight is closer to making an accurate judgment than a node with a lower weight. These weights may be expressed, for example, on a 5-point scale or as a percentage.

[0063] The prestress evaluation system 1 may generate the evaluation model described above in advance, store it in the database 3, and output the evaluation model to the evaluation device 2 upon request from the evaluation device 2. In step S13, the evaluation device 2 refers to the evaluation model described above and outputs reconstructed data based on the waveform data converted in step S12.

[0064] Furthermore, in step S13, the evaluation device 2 may refer to an evaluation model generated using only training waveform data associated with specific condition data as training data, and output reconstructed data. Alternatively, in step S13, the evaluation device 2 may refer to an evaluation model generated using, for example, only normal data as training data, and output reconstructed data. As a result, the reconstructed data is reconstructed based on the characteristics of the normal data, and thus becomes evaluation waveform data based on the characteristics of the normal data. This makes it possible to compare the evaluation waveform data with the normal data, and to evaluate the tension state with high accuracy.

[0065] Furthermore, in step S13, the evaluation device 2 may refer to an evaluation model generated using only abnormal data as training data and output reconstructed data. As a result, the reconstructed data is reconstructed based on the characteristics of the abnormal data, and thus becomes evaluation waveform data based on the characteristics of the abnormal data. This makes it possible to compare the evaluation waveform data with the abnormal data, and to evaluate the tension state with high accuracy.

[0066] In step S13, the evaluation model is referenced, and the tension state of the PC steel 72 is evaluated based on the outputted reconstructed data and waveform data. The evaluation device 2 may calculate anomalies based on, for example, the difference between the waveform data converted in step S12 and the reconstructed data output in step S13. In this case, for example, the cumulative sum of the squared differences in intensity for each frequency between the waveform data and the reconstructed data may be used as the anomaly value. Alternatively, the evaluation device 2 may calculate anomalies based on, for example, the ratio between the waveform data converted in step S12 and the reconstructed data output in step S13. In this case, for example, the cumulative sum of the squared values ​​of the ratio of intensity for each frequency between the evaluation waveform data and the reconstructed data plus -1 may be used as the anomaly value. Since this anomaly value increases in proportion to the difference between the waveform data and the reconstructed data, it serves as an indicator of the difference between the waveform data and the reconstructed data. Since the reconstructed data is waveform data that reflects the characteristics of the training data of the evaluation model referenced in step S13, a large anomaly value indicates that the waveform data does not contain the characteristics of the training data. For example, when an evaluation model generated using normal data as training data is referenced, the outlier value will be larger when there is an abnormality in the tension state of the PC steel material 72. This makes it possible to evaluate the similarity between waveform data and normal data by using normal data as training data, and to evaluate the similarity between waveform data and abnormal data by using abnormal data as training data.

[0067] Furthermore, the evaluation model may be generated using machine learning, for example, a neural network model. The evaluation model may be trained using machine learning, for example, a neural network model such as a CNN (Convolutional Neural Network), or any arbitrary model may be used. In addition, the evaluation model may be generated using methods such as Retrieval-Augmented Generation (RAG), Seq2Seq (Sequence To Sequence) linear discriminant analysis, support vector machines, k-nearest neighbors, random forests, or deep learning.

[0068] In such cases, the evaluation model stores a correlation with weights between, for example, the input data, which is the training waveform data, and the output data, which is the evaluation result. This allows the evaluation device 2 to input the waveform data converted in step S12 into the evaluation model in step S13 and output the evaluation result.

[0069] The operation of the prestress evaluation system to which the embodiment of the present invention is applied is completed by the steps described above. This allows for the evaluation of the tension state of the PC steel based on the waveform data. Therefore, it is possible to evaluate the tension state of the PC steel from the waveform data.

[0070] While embodiments of the present invention have been described, these embodiments are presented as examples only and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims of the invention and its equivalents. [Explanation of Symbols]

[0071] 1. Prestress Assessment System 2. Evaluation device 3 Databases 6. Vibration jig 7. Concrete structures 8 Sound collection section 9. Conversion device 21 Internal bus 23 Display section 24 Control Unit 25 Control section 26 Communications Department 27 Exploration Department 28 Memory section 61 Spring section 62 Vibration section 63 Opening 71 Cable Sheath 72 PC steel material 73 Perforation part

Claims

1. A means for acquiring waveform data including vibration noise of PC steel materials, The system includes an evaluation means for evaluating the tension state of the PC steel material based on the waveform data acquired by the acquisition means. A prestress evaluation system characterized by the following.

2. The evaluation means evaluates the tension state of the PC steel material based on waveform data acquired in advance and waveform data acquired by the acquisition means. The prestress evaluation system according to claim 1, characterized by the following:

3. The evaluation means evaluates the tension of the PC steel material based on two or more waveform data, each containing vibration sounds of PC steel materials with different tensions, which have been acquired in advance, and the waveform data acquired by the acquisition means. The prestress evaluation system according to claim 1, characterized by the following:

4. The acquisition means acquires waveform data including vibration noise of the PC steel material, which is not filled with grout, through a perforation drilled from the surface of the PC structure to the PC steel material. The prestress evaluation system according to claim 1, characterized by the following:

5. The evaluation means evaluates the tension state of the PC steel based on waveform data acquired in advance and subjected to short-time Fourier analysis or wavelet analysis, and waveform data acquired by the acquisition means and subjected to short-time Fourier analysis or wavelet analysis. The prestress evaluation system according to claim 1, characterized by the following:

6. The evaluation means refers to an evaluation model that has been trained using previously acquired training waveform data as training data, and evaluates the tension state of the PC steel material based on the waveform data acquired by the acquisition means. The prestress evaluation system according to claim 1, characterized by the following:

7. Acquisition step to acquire waveform data including vibration noise of PC steel material, The computer is instructed to perform an evaluation step, which evaluates the tension state of the PC steel material based on the waveform data acquired in the acquisition step. A prestress assessment program characterized by [feature].

Images