Structure soundness evaluation method

Abstract

To evaluate soundness of a structure by accurately detecting deterioration and damage in various kinds of concrete structures and steel structures from an early stage.SOLUTION: An attractor trajectory is generated based on acceleration waveform data measured by vibrating a structure to be an evaluation object. A change in an attractor feature amount defined based on the attractor trajectory is grasped so as to evaluate a progress degree of damage and deterioration of the structure. In this case, the attractor trajectory is generated for each fixed process stage where the occurrence and progress of the damage and deterioration are estimated from a time when the structure is not damaged. Then, the attractor feature amount corresponding to the generated attractor trajectory is calculated as a cumulative value Σa b of a trajectory length (a) of the attractor trajectory and a distance (b) between the attractor trajectory and an original point. A ratio of the attractor feature amount between the time without damage and the process stage is obtained, so as to be compared with a ratio in the time without damage or in a time before the process stage. Then, the progress degree of the damage and deterioration of the structure is evaluated from the change of the attractor feature amount.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a method for evaluating the soundness of a structure, and more particularly to a method for evaluating the soundness of a structure using a chaotic response attractor. [Background technology]

[0002] Conventionally, in order to grasp the deterioration or damage of a structure, there are various soundness evaluation methods that evaluate the degree of deterioration or damage of a structure by focusing on, for example, changes in natural frequency based on measurement data obtained by vibrating the structure. In the soundness evaluation method of a structure, it is necessary to grasp the deterioration or damage from an initial minor state and to perform continuous monitoring thereafter. However, with the above-mentioned method of grasping the change in natural frequency, it is difficult to detect relatively minor deterioration or damage, and it is difficult to detect damage in the early stages.

[0003] In response to this, in recent years, an evaluation method based on attractor changes (attractor evaluation method) has been proposed in which an attractor is generated from the response of a structure to which a chaotic signal is input as an external force, and damage detection is performed by comparing the attractor changes (trajectories), for example, before and after damage to the structure. It has been shown that relatively minor damage can be detected using the attractor evaluation method. Since the attractor evaluation method makes it possible to detect relatively minor damage, it is expected that the maintenance and management of structures can be performed efficiently. Therefore, the development of an attractor evaluation method that has a relatively high sensitivity for detecting abnormalities and is relatively easy to apply is highly desirable in the maintenance and management of structures. The applicant has also published a paper on an attractor evaluation method that uses the length ratio of attractor trajectories before and after damage to a concrete structure (Non-Patent Document 1).

[0004] In the attractor evaluation method disclosed in Non-Patent Document 1, as shown in Fig. 7, an attractor trajectory is drawn with the acceleration waveform data A1 obtained by vibrating a test specimen on the horizontal axis and the acceleration waveform data delayed by 0.002 seconds from the acceleration waveform data A1 on the vertical axis, and the cumulative value Σa of the trajectory length at each time step obtained from this attractor trajectory is defined as an attractor feature in the form of the following (Equation 1). In (Equation 1), n ​​represents the final time step. Then, a method for detecting damage to a concrete structure is proposed in which a damage rate is calculated from the ratio of the attractor feature when undamaged to the attractor feature when damaged.

[0005] (Formula 1) TIFF2025029809000002.tif15144 [Prior art documents] [Non-patent literature]

[0006] [Non-Patent Document 1] Shoichiro Kato and 5 others, "Study on Damage Detection of Structures Using Attractors", 77th Annual Lecture of the 2022 JSCE National Convention, JSCE, August 1, 2022, VI-853 Summary of the Invention [Problem to be solved by the invention]

[0007] The applicant has been studying ways to improve the evaluation accuracy of the damage detection method using attractors (see Non-Patent Document 1) described above, but has found the following problems in the process. (1) The attractor features obtained from the attractor trajectory do not have a practical evaluation accuracy, so it is necessary to redefine the attractor features and improve their accuracy. (2) Noise contained in the measured acceleration waveform data causes errors in the generated attractor trajectory length. (3) The peak values ​​of the acceleration waveform data measured in each test vary, making the generated attractor unclear.

[0008] Therefore, in order to solve the problems of the conventional techniques described above, the object of the present invention is to provide a method for evaluating the soundness of a structure with improved evaluation accuracy by redefining a new attractor feature when the attractor trajectory length is used as an evaluation index and by preprocessing the measurement data (acceleration waveform data) of a vibration test used to generate an attractor. [Means for solving the problem]

[0009] The method for evaluating the health of a structure of the present invention generates an attractor trajectory based on acceleration waveform data measured by exciting a structure to be evaluated, and evaluates the degree of progress of damage and deterioration of the structure by grasping changes in attractor features defined based on the attractor trajectory, wherein the attractor trajectory is generated for each predetermined progress stage at which the occurrence and progress of damage and deterioration are expected from a time when the structure is undamaged, the attractor feature corresponding to the generated attractor trajectory is calculated as a cumulative value Σa·b of a trajectory length a of the attractor trajectory and a distance b between the attractor trajectory and an origin, a ratio of the attractor feature at the time when the structure is undamaged to the progress stage is calculated, and this ratio is compared with the ratio at the time when the structure is undamaged or before the progress stage, and the degree of progress of damage and deterioration of the structure is evaluated from the change in the attractor feature.

[0010] The attractor locus is preferably generated on a plane with the acceleration waveform data on the horizontal axis and the acceleration waveform data at a predetermined delay time on the vertical axis.

[0011] It is preferable that the acceleration waveform data is subjected to a high frequency noise removal process and is extracted in a predetermined section in order to generate the attractor locus.

[0012] It is preferable that the waveform data extracted in the predetermined section is averaged to obtain peak values ​​included in the predetermined section.

[0013] The acceleration waveform data is preferably obtained by applying a horizontal impact strike to the structure.

[0014] The acceleration waveform data is preferably obtained by applying vertical vibration to the structure. Effect of the Invention

[0015] As described above, according to the present invention, a highly accurate method for evaluating structural integrity can be realized, which is expected to enable early detection of deterioration and damage to various concrete structures and steel structures. [Brief description of the drawings]

[0016] [Figure 1] 2 is a flowchart showing the flow of a series of processing steps in the method for evaluating the soundness of a structure of the present invention. [Diagram 2] FIG. 1 is an explanatory diagram showing an overview of the test specimen and vibration and measurement equipment for impact vibration testing. [Diagram 3] FIG. 4 is an explanatory diagram showing an attractor trajectory generated based on acceleration waveform data. [Figure 4] FIG. 4 is an explanatory diagram showing attractor features redefined using the attractor trajectory shown in FIG. 3 . [Diagram 5] FIG. 13 is a diagram showing the relationship between the ratio of attractor features and damage rate obtained using the attractor features of the present invention. [Figure 6] 4 is a flowchart showing a pre-processing step of acceleration waveform data in the present invention. [Figure 7] FIG. 1 is an explanatory diagram showing an example of a known attractor feature defined using an attractor trajectory. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0017] An embodiment of the method for evaluating the soundness of a structure according to the present invention will be described below with reference to the accompanying drawings. FIG. 1 shows a flow of a series of processes in the method for evaluating the soundness of a structure according to the present invention. In order to explain this series of processes, in the present invention, as described below, a test specimen 1 (FIG. 2) simulating an actual structure is used, an accelerometer 3 is installed on the test specimen 1, and an artificial impact or vibration similar to that of a vehicle running is applied to measure acceleration waveform data, an attractor (attractor trajectory) is generated based on the acceleration waveform data, and a new attractor feature is redefined based on the attractor. It has been confirmed that the degree of deterioration or damage of the target test specimen 1 can be evaluated from the change between the attractor feature in the initial state and the attractor feature after a certain time (period) has elapsed. First, as technical features of the present invention, the attractor trajectory generated based on waveform data from an impact impact test, the contents of the redefinition of the attractor feature using the generated attractor trajectory, and the soundness evaluation of a structure will be described below, and then the pre-processing process of the waveform data performed to improve the accuracy of the generated attractor will be described.

[0018] [Outline of impact test] (Overview of the test specimen) An outline of an impact impact test using a test specimen 1 for collecting acceleration waveform data in order to generate an attractor trajectory used in the above-mentioned method for evaluating the soundness of a structure will be described with reference to FIG. 2. FIG. 2 is an explanatory diagram showing an outline of an impact impact test using an iron ball 2. As shown in the figure, in this impact impact test, a horizontal impact (horizontal excitation) is applied by impacting a steel-reinforced concrete box culvert as the test specimen 1 with an iron ball 2 suspended in a pendulum shape, and acceleration waveform data propagating within the structure can be collected by an accelerometer 3 installed on the test specimen 1. In addition to the excitation by this horizontal impact, it is also possible to apply continuous vertical vibration to the test specimen 1 by operating a vibration exciter 4 installed on the upper part of the test specimen 1, and collect the vertical acceleration waveform data. In this case, it is expected that the method can be applied to the evaluation of the soundness of a bridge or other structure due to impact vibration or continuous vibration when a vehicle passes through the structure. It is preferable that the frequency range of the input vertical excitation is set assuming the vibration acting on the actual structure. As an example, the test piece 1 can be vibrated for several seconds at a frequency of about 10 to 20 Hz to collect predetermined acceleration waveform data.

[0019] (Acceleration waveform data collection) In this impact impact test, the test specimen 1 is in an undamaged initial state, and a damaged portion 5 is created by missing a part of the test specimen 1, and the horizontal impact test is performed several times at each damage stage simulating the progression of the damage state, and acceleration waveform data is collected for each test. When the iron ball 2 is collided with the test specimen 1, various conditions such as the swing-up angle of the pendulum are specified so that the impact energy input to the test specimen 1 by the impact is constant every time.

[0020] [About the new attractor features] Below, we explain the content of the new attractor feature for the attractor trajectory generated based on the acceleration waveform data obtained by applying a specified excitation force (horizontal impact, vertical excitation) to the test specimen 1 (Fig. 2), and explain the validity of the method for evaluating the soundness of a structure using the new attractor feature.

[0021] In the present invention, a new attractor feature is defined for the attractor feature used in the evaluation of the attractor trajectory length ratio disclosed in the above-mentioned Non-Patent Document 1, and the degree of deterioration or damage of a structure is grasped from the change in the attractor feature.

[0022] (Generation of attractor loci) FIG. 3 shows an attractor trajectory generated to apply a new attractor feature in the present invention. This attractor trajectory was generated by the same procedure as that adopted in Non-Patent Document 1. Specifically, as shown in FIG. 3, the horizontal axis is the time series data of the waveform, and the vertical axis is the waveform plotted with a delay time d=0.002 seconds to generate an attractor trajectory. The delay time d is set so that the attractor shape spreads in all directions and is easy to visually recognize, and is set to the same value at each stage from undamaged to damaged. The delay time d is preferably set to about 1 / 4 of the period T of the acceleration waveform data. Since the acceleration waveform data assumed in the present invention is an irregular wave, it is also preferable to set the range of the delay time to the following range formula. 0.5×(T / 4) <d<1.5×(T / 4) For example, when the period T of the acceleration waveform data is observed to be 0.008 seconds, the preferable range of the delay time d from the above formula is 0.001 <d<0.003となる。

[0023] In FIG. 3, the dashed line indicates the attractor trajectory when there is no damage, and the solid line indicates the attractor trajectory after a certain amount of damage has occurred. Comparing the solid attractor trajectory with the attractor trajectory before damage (shown by the dashed line), there is a difference (deviation) between the trajectories at each stage. A situation in which the attractor trajectory has changed in this way can be judged to indicate the occurrence of damage between the stages. In other words, by grasping the degree to which the attractor trajectory has changed after a certain period of time from the undamaged state (for example, immediately after completion) or at a certain stage (time) when the occurrence of damage is observed, the damage state of the structure can be qualitatively grasped. The acceleration waveform data for generating the attractor trajectory is pre-processed data obtained by performing a predetermined pre-processing (described later) on the waveform data collected in the horizontal vibration test of the test body 1.

[0024] (Redefinition of attractor features) Figure 4 shows the attractor feature Σa·b, which is newly defined using the attractor trajectory shown in Figure 3, in relation to the attractor trajectory in the first quadrant. In previous studies, the sum (Σa) of the attractor trajectory (trajectory length: a) between each time step (k+1-k) was defined as the attractor feature, as shown in Figure 7. In contrast, in this invention, we focus on the distance of the attractor trajectory from the origin, and redefine and quantify the attractor feature in the attractor trajectory of interest as the cumulative value (Σa·b) of (trajectory length: a) × (distance between origins: b) between each time step (k+1-k), as shown in (Equation 2). Note that the other end of the attractor trajectory with respect to the origin when calculating b was set to the midpoint of the corresponding trajectory length a.

[0025] (Formula 2) TIFF2025029809000003.tif21161

[0026] In the case of the attractor feature proposed in Non-Patent Document 1 (Figure 7), the shape of the attractor trajectory may differ even if the attractor trajectory length is the same. Specifically, the attractor trajectory length may be the same even though the spread of the attractor is slightly different. In that case, it is possible that the attractor feature does not change even though a change is observed in the attractor trajectory. Therefore, in the present invention, i The distance between the origin and the target point (origin distance) b was added as an additional element to redefine the attractor feature, making it possible to capture the attractor trajectory more clearly. As a result, we expect to improve the evaluation accuracy.

[0027] [Calculation of the ratio of attractor features] Furthermore, the ratio of the attractor feature value at each stage of damage progression after damage to the attractor feature value at the time of no damage is calculated, and expressed as the ratio of the value at a predetermined damage stage after damage to the value at the time of no damage. By understanding the relationship between the change in this ratio and the damage rate, the validity of the following soundness evaluation is confirmed.

[0028] [Correlation between attractor feature ratio and damage rate and its application to soundness assessment] FIG. 5 shows the transition of the ratio of the attractor feature amount Σa·b before and after damage. In FIG. 5, a value of 1.0 on the vertical axis means no damage, and the ratio gradually decreases with an increase in the damage rate (sectional loss rate). By knowing the extent to which the ratio of the attractor feature amount Σa·b has decreased compared to 1.0, the degree of damage (soundness) of the structure at a certain stage (time) can be evaluated. As shown in FIG. 5, the correlation coefficient of the ratio of the attractor feature amount using this attractor feature amount and the damage rate relationship is -0.75, and an improvement in the evaluation accuracy of about 1.5 times compared to the applicant's previous research was confirmed. As shown in FIG. 5, the correlation coefficient of -0.75 indicates that the correlation between the ratio of the attractor feature amount and the damage rate relationship is sufficiently strong, and deterioration or damage of the structure can be detected at an early stage, for example, when the damage rate is about 10%. In some cases, it is also necessary to grasp the extent to which the ratio of the attractor feature amount has increased compared to 1.0. In other words, when the same excitation frequency is applied to a structure each time, causing damage to the structure, the natural frequency of the structure generally decreases as the damage progresses. At this time, it is thought that the ratio of the attractor features will tend to decrease or increase compared to 1.0 depending on whether the excitation frequency is closer to the natural frequency of the structure before or after the damage. In this way, it is sometimes necessary to understand how much the ratio of the attractor features has increased compared to 1.0. Note that although the natural frequency is not limited to the first order, if the two frequencies become closer, a resonance effect will occur and the response waveform and the attractor (trajectory) due to that waveform will become larger.

[0029] In addition, in this impact impact test, the damage rate was artificially increased to confirm the validity of the soundness evaluation, but it is believed that the same procedure can be used to evaluate not only visible damage but also material deterioration of the components that make up the structure.

[0030] [Preprocessing of acceleration waveform data] The acceleration waveform data used to calculate the attractor feature quantities described above have problems, such as the noise contained in the measured waveform data causing errors in the attractor locus length, and the variation in the peak values ​​of the measured waveform data making the attractor unclear. To solve these problems, two pre-processing steps are implemented (Fig. 6).

[0031] As shown in FIG. 6, the pre-processing process consists of a process for removing high frequency noise and a process for correcting the waveform amplitude by averaging the peak values ​​of each measurement. (High frequency noise removal processing, creation of extracted waveform data) In the above-mentioned impact impact test, acceleration waveform data is obtained for each impact, but it has been confirmed that the attractor becomes unclear due to the influence of high-frequency noise. Therefore, in the present invention, a noise processing is performed to remove high-frequency noise as a preprocessing of data used for generating attractors. In this case, in order to leave low-frequency components that significantly contribute to the characteristics of the attractor, high-frequency noise that is six times or more the primary natural frequency of the target structure is removed using a known low-pass filter. Waveform data from the first peak value of the acceleration waveform data that has been subjected to the high-frequency noise removal processing and a predetermined section thereafter (for example, a section including three peak values) is extracted to create extracted waveform data. The following waveform amplitude correction is performed on the extracted waveform data to generate an attractor trajectory. The extracted waveform data may also be a single waveform data that continues for a predetermined time.

[0032] (Waveform amplitude correction of extracted waveform data) In the impact impact test, the impact energy applied to the test specimen was set to be constant each time, so the impact impacts (excitation) are about the same each time, but the peak values ​​of the multiple acceleration waveform data obtained at a certain stage are not exactly the same. Therefore, the average value of the multiple peak values ​​(for example, three peak values) in the extracted waveform data described above is calculated, and the waveform amplitude is corrected by multiplying the entire waveform for each time by a magnification to average each peak value. This average calculation is only performed in the first stage (when there is no damage), and for the waveform data at each stage to check the progress of damage and deterioration, waveform amplitude correction is performed by multiplying the entire waveform data by a magnification so that the peak value matches the average value. By using the waveform data that has been subjected to this waveform amplitude correction, a clear attractor can be generated.

[0033] [Application of this invention to the soundness evaluation of real structures and its scalability] As described above, the validity of the method according to the present invention for evaluating the soundness of a structure has been confirmed. In order to utilize the present invention for evaluating the soundness of an actual structure, first, attractor features in an undamaged state at the time of completion of the actual structure are obtained, and then vibration and measurement for obtaining data are performed at regular intervals or when an event that may cause damage occurs or at a stage. This makes it possible to compare the attractor features obtained at the time of completion or at an earlier time or stage, and by observing and examining the changes in the attractor features at that time, it is possible to evaluate the degree of progress of damage or deterioration of the structure.

[0034] In the impact impact test described above, two types of vibration excitation, horizontal impact and vertical excitation, were adopted, and attractor features for each were obtained, but a high correlation coefficient was confirmed from the ratio of attractor features based on acceleration waveform data obtained by both impact impact and continuous vibration excitation. Therefore, it can be said that the evaluation method of the present invention is applicable to both excitation conditions.

[0035] Furthermore, in the course of research, the applicant conducted vibration tests using a reinforced concrete box culvert, a reinforced concrete pier model, a two-story steel structural frame, and the like as test specimens, obtained attractor features, and obtained the relationship between the attractor features before and after damage and the damage rate, and confirmed that a high correlation coefficient was obtained between the ratio of the attractor features and the damage rate for each test specimen. Therefore, it is clear that the present invention can be applied to the main bodies of various reinforced concrete and steel structures. In addition, the present invention is expected to be applied to the soundness evaluation of expansion devices and supports of joints as auxiliary equipment of superstructures of road bridges, etc., during service, such as when vehicles pass through.

[0036] The present invention is not limited to the above-described embodiment, and various modifications are possible within the scope of the claims. In other words, the technical scope of the present invention also includes embodiments obtained by combining technical means that are appropriately modified within the scope of the claims. [Explanation of symbols]

[0037] 1. Test specimen 2. Iron Ball 3 Accelerometer 4. Vibration machine 5 Damage

Claims

1. A method for evaluating the soundness of a structure, the method comprising: generating an attractor trajectory based on acceleration waveform data measured by vibrating a structure to be evaluated; and evaluating a degree of progress of damage and deterioration of the structure by grasping a change in an attractor feature quantity defined based on the attractor trajectory, a trajectory that is generated for each predetermined stage of evolution at which occurrence and progression of damage or deterioration is expected from a time when the structure is undamaged; an attractor feature amount corresponding to the generated attractor trajectory is calculated as a cumulative value Σa·b of a trajectory length a of the attractor trajectory and a distance b between the attractor trajectory and an origin; a ratio of the attractor feature amount at the time when the structure is undamaged to that at the evolution stage is calculated and compared with the ratio at the time when the structure is undamaged or before the evolution stage; and a degree of progression of damage and deterioration of the structure is evaluated from a change in the attractor feature amount.

2. 2. The method for evaluating the soundness of a structure according to claim 1, wherein the attractor locus is generated on a plane whose horizontal axis represents the acceleration waveform data and whose vertical axis represents the acceleration waveform data at a predetermined delay time.

3. 2. The method for evaluating soundness of a structure according to claim 1, wherein the acceleration waveform data is subjected to a high frequency noise removal process and is extracted in a predetermined section for generating the attractor locus.

4. 4. The method for evaluating soundness of a structure according to claim 3, wherein the waveform data extracted in the predetermined section is obtained by averaging peak values ​​included in the predetermined section.

5. 5. The method for evaluating the soundness of a structure according to claim 1, wherein the acceleration waveform data is obtained by applying a horizontal impact strike to the structure.

6. 5. The method for evaluating the soundness of a structure according to claim 1, wherein the acceleration waveform data is obtained by applying vertical vibration to the structure.

Images