Monitoring device, monitoring program, and monitoring method

Abstract

To make it possible to automatically execute abnormality determination of a structure by recognizing a natural frequency of each structure from a plurality of spectral data and automatically estimating parameters such as tensile force and the like by using the natural frequency.SOLUTION: According to the present invention, a monitoring device includes analysis means for detecting peaks appearing in a frequency spectrum acquired on the basis of a measurement signal from a sensor provided in a structure to acquire information on the peak in each peak, storage means for storing the information on the peak of each peak in each measurement, grouping means for grouping a frequency in which each peak is detected in each frequency band on the basis of the information on the peak of each peak acquired over multiple times of measurement, natural frequency group extraction means for extracting a plurality of frequency bands belonging to a natural frequency group from among the plurality of frequency bands grouped by the grouping means, and natural frequency estimation means for estimating a natural frequency on the basis of the extracted plurality of frequency bands belonging to the natural frequency group.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a monitoring device, a monitoring program, and a monitoring method, and can be applied to, for example, a monitoring device that measures the soundness of a structure based on the vibration of the structure.

Background Art

[0002] Conventionally, various sensors have been installed in factory facilities and infrastructure structures (for example, bridges, columnar structures) to measure the soundness of the object. Many objects have a unique vibration frequency (hereinafter referred to as "natural frequency"), and it is known that the natural frequency changes when an abnormality such as deterioration occurs in the structure.

[0003] On the other hand, in a structure that supports girders with cables such as a cable-stayed bridge, it is necessary to periodically check whether the tension assumed at the time of design is applied to each cable. Conventionally, a method of directly measuring the tension using a jack or the like has been used, but it has required a lot of time and labor. Therefore, in recent years, a measurement method using vibration has attracted attention.

[0004] Non-Patent Document 1 attaches a highly accurate acceleration sensor to a cable and measures the vibration of the cable that vibrates constantly due to wind or the like without applying vibration to the cable. Then, the obtained vibration waveform is subjected to Fourier transform to obtain a frequency spectrum.

[0005] Generally, it is known that vibrations propagating through a structure obtain peaks not only at the natural frequency but also at frequencies that are integer multiples of the natural frequency (hereinafter referred to as "natural frequency group"). The same applies to the vibration of the cable. Basically, peaks are obtained at the natural frequency and frequencies that are integer multiples of the natural frequency.

[0006] It is known that there is a certain relationship between the tension of a cable and its natural frequency, as shown in Equation 1 of Non-Patent Document 1, and the tension can be estimated from the natural frequency. Generally, since the natural frequency is a low frequency of several Hz or less, it is difficult to accurately measure it with a sensor. Also, there is always an error in frequency measurement. Therefore, by using a group of natural frequencies, more accurate measurement becomes possible.

[0007] Also, for example, regarding infrastructure, in an environment where a certain vibration can be obtained because the vibration source is a passing vehicle or wind, etc., in such a case, there may be a situation where no vibration is obtained and no peak can be obtained, or a situation where the peak changes greatly due to an abnormality on the vibration source side. Due to these problems, it is difficult to determine the group of natural frequencies only from the spectrum obtained from a single measurement, and it is necessary to use the spectra obtained from multiple measurements by overlapping them. Even simply extracting the group of natural frequencies is not easy.

[0008] To address these problems, conventionally, a method has been used in which a person visually determines the natural frequency, or normal vibration parameters such as the natural frequency and threshold value are set in advance. Efforts have also been made to automate these settings using machine learning, etc.

Prior Art Documents

Non-Patent Documents

[0009]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0010] As described above, the tension can be estimated from the natural frequency. Generally, however, since the natural frequency is a low frequency of several Hz or less, it is difficult to accurately measure it with a sensor. Also, there is always an error in the measurement of the frequency. Therefore, more accurate measurement is expected by using a group of natural frequencies having the characteristic that peaks appear at frequencies that are integer multiples of the natural frequency.

[0011] Also, for example, regarding infrastructure, it is not an environment where a certain vibration can be obtained because the vibration source is a passing vehicle or wind. In such cases, there may be situations where there is no vibration and no peak can be obtained, or situations where the peak changes greatly due to an abnormality on the vibration source side. Due to these problems, it is difficult to determine the group of natural frequencies only from the spectrum obtained from a single measurement, and it is necessary to use the spectra obtained from multiple measurements by overlapping them. Simply extracting the group of natural frequencies is not easy.

[0012] Regarding the above-described problems, conventionally, a method has been used in which a human visually determines the natural frequency, or normal vibration parameters such as the natural frequency and threshold value are set in advance. Efforts have also been made to automate these settings using machine learning and the like.

[0013] However, as the number of structures to be monitored increases, it is not realistic for a human to set the natural frequency of each individual structure. Even if the setting is automated, it is necessary to confirm the setting result, and it is difficult to confirm that the setting has been made correctly. Also, appropriate setting values may change over time, and it requires a huge amount of labor for a human to perform parameter setting and confirmation regularly.

[0014] Therefore, in view of the above-described problems, the present invention aims to provide a monitoring device, a monitoring program, and a monitoring method that can recognize the natural frequency of each of one or more structures from a plurality of spectrum data without setting the natural frequency for each structure, automatically estimate parameters such as tension using the natural frequency, and automatically perform abnormality determination of the structure.

Means for Solving the Problem

[0015] To solve such a problem, the monitoring device of the first aspect of the present invention includes: (1) analysis means for detecting a peak appearing in a frequency spectrum obtained based on a measurement signal from a sensor provided in a structure, and obtaining information regarding the peak for each peak; (2) storage means for storing the information regarding the peak of each peak for each measurement; (3) grouping means for grouping the frequencies at which each peak is detected into frequency bands based on the information regarding the peak of each peak obtained over a plurality of measurements; (4) natural frequency group extraction means for extracting a plurality of frequency bands belonging to a natural frequency group from among the plurality of frequency bands grouped by the grouping means; and (5) natural frequency estimation means for estimating a natural frequency based on the plurality of frequency bands belonging to the extracted natural frequency group.

[0016] The monitoring program of the second aspect of the present invention causes a computer to function as: (1) analysis means for detecting a peak appearing in a frequency spectrum obtained based on a measurement signal from a sensor provided in a structure, and obtaining information regarding the peak for each peak; (2) storage means for storing the information regarding the peak of each peak for each measurement; (3) grouping means for grouping the frequencies at which each peak is detected into frequency bands based on the information regarding the peak of each peak obtained over a plurality of measurements; (4) natural frequency group extraction means for extracting a plurality of frequency bands belonging to a natural frequency group from among the plurality of frequency bands grouped by the grouping means; and (5) natural frequency estimation means for estimating a natural frequency based on the plurality of frequency bands belonging to the extracted natural frequency group.

[0017] The third monitoring method of the present invention is characterized in that: (1) the analysis means detects peaks appearing in the frequency spectrum obtained based on the measurement signals from sensors provided on the structure, and obtains information regarding the peaks for each peak; (2) the accumulation means accumulates the information regarding the peaks of each peak for each measurement; (3) the grouping means groups the frequencies at which each peak is detected into frequency bands based on the information regarding the peaks of each peak obtained over a plurality of measurements; (4) the natural frequency group extraction means extracts a plurality of frequency bands belonging to the natural frequency group from among the plurality of frequency bands grouped by the grouping means; and (5) the natural frequency estimation means estimates the natural frequency based on the plurality of frequency bands belonging to the extracted natural frequency group.

Effect of the Invention

[0018] According to the present invention, it is possible to recognize the natural frequency of each structure from a plurality of spectrum data, automatically estimate parameters such as tension using the natural frequency, and automatically perform abnormality determination of the structure.

Brief Description of the Drawings

[0019]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Mode for Carrying Out the Invention

[0020] (A) Main Embodiment Hereinafter, embodiments of a monitoring device, a monitoring program, and a monitoring method according to the present invention will be described in detail with reference to the drawings.

[0021] (A-1) Configuration of the Embodiment (A-1-1) Overall Configuration FIG. 2 is an overall configuration diagram showing the overall configuration of a monitoring system according to an embodiment.

[0022] In FIG. 2, the monitoring system 1 includes a master unit 10, a plurality of sensor devices 20 (20-1 to 20-n; n is a positive integer) as slave units, and a monitoring device 30.

[0023] Sensors of the sensor devices 20 are installed on the structure to be monitored, and the sensors measure the vibration of the structure. The sensor device 20 transmits information including the measurement signal (measurement value) measured by the sensor to the master unit 10. The master unit 10 acquires the measurement signals measured by each sensor from each of the plurality of sensor devices 20, and provides the information acquired from each sensor device 20 to the monitoring device 30. The monitoring device 30 collects the measurement signals of the sensors from each sensor device 20, and monitors the state and situation of the object to be monitored using the collected measurement signals.

[0024] Here, there are various "structures" to be monitored, but the structure is an infrastructure structure that is not in an environment where a certain vibration is constantly applied, such as a bridge or the like. Of course, the present invention can also be applied when determining the abnormality of a structure to which a certain vibration is constantly applied. In this embodiment, a case where it is applied to a structure that supports girders with inclined cables or diagonal members, such as a cable-stayed bridge or an extradosed bridge, is exemplified.

[0025] The sensor device 20 mainly includes a sensor, an arithmetic unit for processing data, communication means, a clock, a timer for waking up the entire device from a dormant state, a battery, and the like. The sensor included in the sensor device 20 is an acceleration sensor 203 (see FIG. 3), and the acceleration sensor 203 is fixedly installed on the structure. Note that the acceleration sensor 203 may be installed on the structure when measuring the soundness of the structure. The sensor device 20 transmits a measurement signal (measurement value) intermittently measured by the acceleration sensor 203 to the master device 10.

[0026] The master device 10 is a wireless device that transfers information including the measurement signal received from the sensor device 20 to the monitoring device 30. The master device 10 can communicate with a plurality of sensor devices 20 and transfers the information received from each sensor device 20 to the monitoring device 30.

[0027] Note that in addition to the transfer function, the master device 10 may have the same configuration as the sensor device 20. In that case, the master device 10 may directly transmit the measurement signal measured by its own sensor to the monitoring device 30. As another modification, the master device 10 may have all or part of the functions of the monitoring device 30 described later.

[0028] The monitoring device 30 collects and stores (accumulates) the measurement signals measured by the acceleration sensor 203 from each sensor device 20 via the master device 10. The monitoring device 30 performs a Fourier transform on the measurement signal of the acceleration sensor 203 to derive a frequency spectrum. The monitoring device 30 analyzes the shape of the frequency spectrum, extracts a specific frequency region where a peak appears during normal operation of the structure, and accumulates peak data for each specific frequency region.

[0029] At this time, each time the monitoring device 30 measures using the acceleration sensor 203, it measures and accumulates the peak data for each specific frequency region extracted from the frequency spectrum. For example, when a training period of several days is set and the acceleration sensor 203 measures once a day, the monitoring device 30 measures and accumulates the peak data for each specific frequency region extracted from the frequency spectrum once a day. In this way, by setting a training period, the peak data for each specific frequency region can be accumulated, improving the accuracy.

[0030] Also, the monitoring device 30 groups using a large amount of peak data for specific frequency regions, and obtains and accumulates the peak data for the specific frequency region presumed to be the natural frequency of the structure.

[0031] Furthermore, the monitoring device 30 extracts a group of natural frequencies from among a plurality of specific frequency regions (groups) obtained by multiple measurements, and estimates the natural frequency using the difference value between a certain frequency and the frequency adjacent to it in the group of natural frequencies. The detailed description of the method for extracting the group of natural frequencies and the method for estimating the natural frequency will be given in the section on operations.

[0032] FIG. 2 illustrates an information communication system in which the sensor device 20 as a slave device transmits information to the monitoring device 30 via the master device 10, but the configuration of this information communication system is an example. As long as the sensor device 20 can transmit information to the monitoring device 30, the configuration of the information communication system is not limited to the configuration of FIG. 2.

[0033] A network including the master device 10, the sensor device 20, and the monitoring device 30 may use a wireless connection or a wired connection. For example, the sensor device 20 and the master device 10 may form a sensor network SN. As the communication method of the sensor network SN, a low-speed wireless communication method represented by a specific low-power wireless method or the like can be applied. For example, it may be a wireless network standard represented by IEEE802.11a / b / g / n or the like, or a wireless communication method such as IEEE802.15.4 or Bluetooth (registered trademark). Each of the sensor device 20 and the master device 10 is assigned a unique address (for example, MAC address, short address, IP address, etc.) on the sensor network SN.

[0034] The network between the master device 10 and the monitoring device 30 may be a backbone network NT. The backbone network NT is, for example, the Internet or Ethernet (registered trademark). The backbone network NT may be a wireless connection or a wired connection.

[0035] (A-1-2) Internal Configuration of Monitoring Device FIG. 1 is an internal configuration diagram showing the internal configuration of the monitoring device 30 according to the embodiment.

[0036] In FIG. 1, the monitoring device 30 includes a communication unit 301, a control unit 302, a peak data analysis unit 303, a grouping processing unit 304, a data storage unit 306, a unique frequency group extraction unit 307, and a unique frequency estimation unit 308.

[0037] The communication unit 301 is a communication interface for communicating with the network connected to the master device 10.

[0038] The control unit 302 is a device or processing unit that controls various functions of the monitoring device 30. The control unit 302 may be configured by a device having, for example, a CPU, a ROM, a RAM, an EEPROM, an input / output interface, etc. The processing may be realized by the CPU executing a processing program (for example, a monitoring program, etc.) stored in the ROM.

[0039] Based on the value of the natural frequency estimated by the natural frequency estimation unit 308, the control unit 302 has a physical value derivation unit that derives a value (physical numerical value) indicating the state of the structure such as tension.

[0040] Further, when the value indicating the state of the structure such as tension deviates from the preset range, or when it deviates from the range of the values obtained in past measurements, the control unit 302 has a warning unit that outputs a warning.

[0041] The data storage unit 306 stores information for determining the soundness of the structure. For example, the data storage unit 306 stores the measurement data of each acceleration sensor 203 (for example, the measurement signal measured by the acceleration sensor 203, the frequency spectrum data obtained by Fourier-transforming the measurement signal, etc.), and the peak data in a specific frequency region obtained using the measurement signal (for example, data including frequency, the value indicating the height of the peak, date and time information, group number, etc.).

[0042] The peak data analysis unit 303 Fourier-transforms the measurement signal measured by the acceleration sensor 203 stored in the data storage unit 306 to obtain a frequency spectrum. Also, the peak data analysis unit 303 uses the concept of the height of the peak to obtain the frequency that becomes the maximum value (hereinafter referred to as the "peak frequency") among many peaks appearing in the frequency spectrum. Then, the peak data analysis unit 303 rearranges the peak data in the order of the height of the peak, extracts the upper peak data to delete unnecessary peaks, and stores it in the data storage unit 306. A detailed description of the peak data analysis unit 303 will be given in the section on operations. Also, when the peak data analysis unit 303 creates peak data, the group number does not have to be determined. The date and time information is the information indicating the measurement date and time of the acceleration sensor 203.

[0043] The grouping processing unit 304 also reads out a plurality of peak data measured in the past from the data storage unit 306, assigns a group number to each specific frequency region, and groups (collects) the peak data for each specific frequency region. Further, the grouping processing unit 304 repeatedly performs grouping using a large amount of peak data in a specific frequency region to obtain the peak characteristics of the specific frequency region estimated to be the natural frequency of the structure. A detailed description of the grouping process will be given in the section on operations.

[0044] The natural frequency group extraction unit 307 extracts frequencies estimated to belong to the natural frequency group from among the plurality of frequencies obtained by grouping. A detailed description of the method for extracting the natural frequency group will be given in the section on operations.

[0045] Here, the "natural frequency group" refers to a plurality of frequencies that are considered to be natural frequencies or frequencies that are integer multiples of natural frequencies among a plurality of specific frequency regions.

[0046] The natural frequency estimation unit 308 estimates the natural frequency based on the plurality of frequencies belonging to the natural frequency group extracted by the natural frequency group extraction unit 307.

[0047] For example, the natural frequency estimation unit 308 utilizes the characteristic that peaks appear at frequencies that are integer multiples of the natural frequency to obtain the frequency difference between each group in the natural frequency group. That is, the difference value of the frequencies between a certain group and the group adjacent to it is obtained. Then, the natural frequency estimation unit 308 obtains the average value of the frequency difference values between the groups (referred to as the average frequency difference value between the groups), and when the average frequency difference value between the groups becomes a certain numerical value (or a value approximated to a certain numerical value), that certain numerical value is estimated as the natural frequency. An example of the method for estimating the natural frequency will be described in detail in the section on operations.

[0048] (A-1-3) Internal Configuration of the Sensor Device FIG. 3 is an internal configuration diagram showing the internal configuration of the sensor device 20 according to the embodiment.

[0049] In FIG. 3, the sensor device 20 includes a communication unit 201, a control unit 202, an acceleration sensor 203, a schedule determination unit 204, a timer unit 205, a clock unit 206, and a data storage unit 207.

[0050] The sensor device 20 may be configured by hardware, and some components of the sensor device 20 may also be configured software-wise.

[0051] The communication unit 201 is a communication interface that communicates with a network connected to the master device 10.

[0052] The acceleration sensor 203 measures the vibration of a structure such as a cable or a bridge. The acceleration sensor 203 measures the instantaneous acceleration and gives a measurement signal to the control unit 202. One sensor device 20 may be provided with a plurality of acceleration sensors 203. Note that the type of sensor is not limited to this. For example, in addition to or instead of the acceleration sensor 203, the sensor device 20 may be provided with various types of sensors such as a temperature sensor, a humidity sensor, a vibration sensor, and an infrared image sensor.

[0053] The control unit 202 is a device and a processing unit that controls various functions in the sensor device 20. The control unit 202 has, for example, a CPU, a ROM, a RAM, an EEPROM, an input / output interface, etc. The processing may be realized by the CPU executing a processing program (for example, a measurement program, etc.) stored in the ROM.

[0054] Also, the control unit 202 receives information regarding the measurement operation from the monitoring device 30 via the communication unit 201, and gives measurement operation schedule information for operating the acceleration sensor 203 to the schedule determination unit 204.

[0055] Based on the measurement operation schedule information received from the monitoring device 30, the schedule determination unit 204 determines the periodic measurement timing for operating the acceleration sensor 203. The schedule determination unit 204 sets the periodic measurement timing of the acceleration sensor 203 in the timer unit 205.

[0056] The timer unit 205 manages the measurement timing set by the schedule determination unit 204. The timer unit 205 monitors the current time of the clock unit 206, and when the time reaches the measurement time of the measurement timing, the timer unit 205 activates the control unit 202. As a result, since the control unit 202 is activated, the acceleration sensor 203 becomes active, and the acceleration sensor 203 operates. Note that when it is not the measurement time, the sensor device 20 is in a standby state, and components other than the timer unit 205 and the clock unit 206 are put into standby. Thereby, power consumption can be suppressed.

[0057] The clock unit 206 is a clock that provides the current time.

[0058] (A-2) Operations of the Embodiment First, the overall processing in the monitoring system 1 will be briefly described.

[0059] Based on the measurement operation schedule, the sensor device 20 periodically determines the measurement timing for operating the acceleration sensor 203. In the sensor device 20, when the time of the measurement timing is reached, the acceleration sensor 203 operates for a predetermined time. The operating acceleration sensor 203 measures data indicating characteristics of the structure, such as acceleration due to vibration. The communication unit 201 transmits information including the measurement signal (measurement value) measured by the acceleration sensor 203 to the master device 10, and the master device 10 transfers the information received from the sensor device 20 to the monitoring device 30.

[0060] Next, an example of the operation of the monitoring method in the monitoring device 30 according to the embodiment will be described with reference to the drawings.

[0061] FIG. 4 is a flowchart showing the processing operations in the monitoring device 30 according to the embodiment. Hereinafter, the processing in the monitoring device 30 will be described based on the procedure shown in FIG. 4.

[0062] [S1: Derivation of Peak Height] The monitoring device 30 receives the measurement signal (measurement value) measured by the acceleration sensor 203 of the sensor device 20. In the monitoring device 30, the peak data analysis unit 303 performs a Fourier transform on the measurement signal (time-series vibration waveform) to obtain a frequency spectrum. The monitoring device 30 analyzes the frequency spectrum to identify the natural frequency of the structure, detects changes in the strength (amplitude) of the signal component of the natural frequency, and determines the abnormality of the structure.

[0063] For example, for each measurement by the acceleration sensor 203, the monitoring device 30 extracts peaks from the frequency spectrum obtained by performing a Fourier transform on the measurement signal, and obtains peak data for each specific frequency region.

[0064] Here, there are a large number of peaks in the frequency spectrum. The natural frequency generally exists among a plurality of peak frequencies that appear as maximum values in the frequency spectrum. Even among the maximum values in the frequency spectrum, there are inappropriate maximum values that are not suitable for use in determining the abnormality of the structure.

[0065] For example, due to the characteristics of the Fourier transform, there may be many fine peaks at frequencies before and after a large peak. These fine peaks disappear if the number of data used for the Fourier transform is extremely increased, and it is inappropriate to adopt them as peaks.

[0066] Therefore, in this embodiment, the peak data analysis unit 303 uses the concept of "peak height" to derive the "peak height" of each of the many peaks existing in the frequency spectrum.

[0067] Here, as an example of deriving the "peak height", there is a concept of "prominence". Using FIGS. 5 and 6, the derivation process of prominence will be described.

[0068] FIG. 5(A) is an example of a frequency spectrum. For the sake of convenience in explanation, let the left end point of the frequency spectrum be "a" and the right end point thereof be "g". Each of the five maximum values from "left end a" to "right end g" is set as peak numbers "1" to "5", and each of the five minimum values is set as "b" to "f".

[0069] In the procedure shown in FIG. 6, the peak data analysis unit 303 derives a prominence indicating the height of each peak. First, the peak data analysis unit 303 selects a peak (S101) and moves to the left and right sides of the peak until the value becomes the next state, and derives the end points of the horizontal line (S102). · Crossing with a higher peak and value · The value reaches the left end or the right end.

[0070] Next, the peak data analysis unit 303 obtains the minimum points (minimum values) in the left section and the right section of the peak derived in S102 (S103), and sets the higher value among the two lowest values derived in S103 as the reference level (highest minimum point) (S104). The peak data analysis unit 303 sets the difference between the reference level and the peak as the height (prominence) of the peak (S105).

[0071] The peak data analysis unit 303 determines whether all the peaks have been selected (S106). If not all the peaks have been selected (S106 / No), the peak data analysis unit 303 moves to S101 and repeats the process. On the other hand, if all the peaks have been selected (S106 / Yes), the peak data analysis unit 303 ends the process.

[0072] For example, a method example for deriving the prominence of the peak of "peak number: 1" in FIG. 5(A) will be described.

[0073] First, select the peak with "Peak Number: 1" (S101). When moving the value to the left from this peak, the value reaches "Left End a", so "Left End a" becomes the endpoint of the horizontal line in the left peak section. On the other hand, when moving the value to the right from the peak, the value intersects with the peak of "Peak Number: 2" which is higher than the peak of "Peak Number: 1", so the point where the value intersects with Peak 2 becomes the endpoint of the horizontal line in the right peak section (S102).

[0074] Next, since the minimum point in the left peak section is "Left End a" and the minimum point in the right peak section is "b" (S103), of "Left End a" and "b", the higher one, "b", is set as the reference level (S104). Then, the height "P1" from the reference level "b" to "Peak Number: 1" is set as the height (prominence) of the peak of "Peak Number: 1" (S105). The peak data analysis unit 303 performs the above-described processing for all peaks, and the result is shown in Fig. 5(B).

[0075] [S2: Deletion of Unnecessary Peaks] Next, the peak data analysis unit 303 compares the height of each peak existing in the frequency spectrum with the height of the peaks appearing before and after each peak, and deletes the smaller peak when it is below a certain ratio considering the frequency difference.

[0076] For example, taking Fig. 5(A) as an example, the height of the peak of "Peak Number: 3" is smaller than the height of the adjacent peak of "Peak Number: 4", so the peak of "Peak Number: 3" is a target for deletion.

[0077] More specifically, the peak data analysis unit 303 calculates the ratio of the height P3 of the peak with "peak number: 3" to the height P4 of the peak with "peak number: 4", and when the ratio is equal to or less than the threshold value, it determines that the peak with "peak number: 3" has a small height and designates it as an object to be deleted. As another method, the peak data analysis unit 303 may calculate the difference value between the height P3 of the peak with "peak number: 3" and the height P4 of the peak with "peak number: 4", and when the difference value is equal to or less than the threshold value, it may determine that the peak with "peak number: 3" has a small height. Note that the threshold value used when determining the peak to be adopted based on the peak height may be changed as appropriate.

[0078] Furthermore, as another method, the peak data analysis unit 303 may sort the peak data in descending order of peak height and adopt only the data of a fixed ratio from the top as peaks.

[0079] FIG. 7 shows an example of a frequency spectrum. The horizontal axis represents frequency (Hz), and the vertical axis represents the amplitude spectrum (gal×s). For example, high peaks appear around 8 Hz and around 11 Hz, respectively. On the other hand, although peaks also appear around 7 Hz and around 10 Hz, the height of the peak around 7 Hz is lower than that of the peak around 8 Hz, and similarly, the height of the peak around 10 Hz is lower than that of the peak around 11 Hz. Therefore, the peaks around 7 Hz and around 11 Hz are not adopted as the peaks of this embodiment. That is, the peak data analysis unit 303 deletes the peaks around 7 Hz and around 11 Hz.

[0080] [S3: Creation of Peak Data] The peak data analysis unit 303 creates peak data for a plurality of remaining peaks after deleting unnecessary peaks among the peaks existing in the frequency spectrum. For example, the peak data analysis unit 303 extracts peaks from the frequency spectrum based on the measured values of one day and creates peak data for each peak. Then, the peak data analysis unit 303 accumulates the peak data of each peak in the data storage unit 306.

[0081] Here, the peak data is information for each peak, including the frequency at which the peak appears (peak frequency), the value of the peak height, date information, group number, and the like. Note that at this stage, the group number for each peak has not been determined and no group number has been assigned.

[0082] [S4; Grouping Process] The grouping processing unit 304 reads peak data for multiple days from the data storage unit 306 and sorts all the peak data in ascending order of frequency. Then, the grouping processing unit 304 assigns a group number to all the peak data. For example, in the initial state, a unique group number is sequentially assigned to each of the peak data sorted in ascending order of frequency.

[0083] Next, the grouping processing unit 304 focuses on the two groups with the smallest difference in frequency among the peak data with group numbers assigned. Then, the grouping processing unit 304 examines all the peak data of the two groups to determine that there is no data on the same day in the two groups. If there is no data on the same day, the grouping processing unit 304 combines the two groups. Note that if there is data on the same day, the groups are not combined.

[0084] The grouping processing unit 304 performs the above-described group combination process for all pairs of groups. The combined group is regarded as a new single group, and it is determined whether there is data on the same day with other groups, and the grouping is repeated. By repeating the grouping process, the peak data with close frequencies will belong to the same group in order, and a frequency band is naturally formed.

[0085] For a group formed as a single group, the grouping processing unit 304 also compares the peak data of two groups whose group numbers (or frequencies) are adjacent, and if there is no data on the same day, combines the groups.

[0086] For example, being in different groups means that there is peak data on the same day in both groups for at least one day or more. Here, the existence of peak data on the same day in both groups is denoted as "duplication". The grouping processing unit 304 examines the ratio of how much "duplication" exists in the group (for example, the number of duplications relative to the total number). Also, the grouping processing unit 304 compares the number of peak data belonging to two groups.

[0087] For example, the grouping processing unit 304 compares the number of peak data belonging to each of the two groups. For example, assume that the number of peak data belonging to one group is large and the number of peak data belonging to the other group is small. In that case, if the number of duplications in the two groups is also small (when the number of duplications is below the threshold), the grouping processing unit 304 combines the two groups.

[0088] Also, for example, when the "peak height" of the peak data overlapping in the two groups is clearly low (when the peak height is below the threshold), considering it not an important peak, the grouping processing unit 304 combines the two groups.

[0089] In this way, the grouping processing unit 304 groups while considering the peak data overlapping in two groups even for a group formed as one group.

[0090] [S5: Selection of Group] Through the above-described grouping process, multiple groups are often obtained. Next, the grouping processing unit 304 selects the group to be used as the natural frequency of the structure.

[0091] An example of the method for selecting groups by the grouping processing unit 304 will be described. The method for selecting groups is not limited to selecting by one criterion (or method), but may be selected using a plurality of criteria (or methods). For example, the grouping processing unit 304 selects a group in which peak data exists at a certain ratio or more with respect to the number of measurements. Also, for example, the grouping processing unit 304 selects a group to which peak data having a peak height equal to or greater than a threshold value belongs.

[0092] Here, setting it as "a certain ratio or more" is a particularly important item when monitoring infrastructure without a vibration source. For example, when there is no vehicle passage due to a road closure or the like, or when there is no wind blowing at all, even if there is no abnormality in a structure such as a bridge, a peak may not appear. Therefore, it cannot be simply determined that it is abnormal when there is no peak.

[0093] Therefore, in order to select not only frequencies (groups) with many prominent peaks and frequent peaks, but also frequencies (groups) where peaks do not appear during normal operation of the structure, it is set as "a certain ratio or more".

[0094] Of course, the frequencies of prominent peaks and frequently appearing peaks are also important frequencies (groups), so it is necessary to select these groups as well. Also, the value of a certain ratio is not particularly limited, and may be appropriately determined according to the situation and environment of the structure to be monitored.

[0095] FIG. 8 is an example of the data results measured by the monitoring device 30 for consecutive days. The horizontal axis represents the date, and the vertical axis represents the frequency (Hz). In FIG. 8, the size of the circle indicates the "peak height", and the larger the diameter of the circle, the higher the "peak height".

[0096] Figure 9 shows an example of peak data belonging to some groups among the data results in Figure 8. For example, the "f column" indicates the frequency, the "p column" indicates the peak height, and the numbers of "f" and "p" indicate the group numbers. For example, the vicinity of 6 Hz in Figure 8 is the 9th group, the vicinity of 8 Hz is the 10th group, and the vicinity of 11 Hz is the 11th group.

[0097] For example, the 11th group near 11 Hz is a group to which peak data having a peak height equal to or higher than the threshold value belongs. Also, for example, the 10th group near 8 Hz is a group in which peak data exists at a certain ratio or more with respect to the number of measurement times. In this way, it is possible to select a group in which no peak appears (or a peak hardly appears) during normal times.

[0098] [S6: Extraction of natural frequency groups] The natural frequency group extraction unit 307 extracts a natural frequency group from among a plurality of frequencies obtained by grouping.

[0099] Various methods can be applied as the method for extracting the natural frequency group. The extraction of the natural frequency group is based on the premise that, in principle, peaks appear at frequencies that are integer multiples of the natural frequency due to the vibration of the structure. Although it is ideal for peaks to always appear at integer multiples of the natural frequency, there are other factors and they may not be equally spaced.

[0100] For example, the following formula (1) is Formula 1 described in Non-Patent Document 1, but also in this example, fi 2 is a component proportional to i 4 (that is, a component in which the square of the natural frequency is proportional to the fourth power of an integer value (the value of the order)). However, in many cases, it is considered that the influence of these terms is small. When the number of parameters increases in this way, it is also possible to adopt a configuration in which the most appropriate parameters are estimated using the distribution of peaks obtained from a plurality of spectra. Also, a configuration in which a plurality of parameters including the natural frequency are simultaneously estimated using a method such as the least squares method can be adopted.

[0101]

Number

[0102] [S7: Estimation of Natural Frequency] First, the natural frequency group extraction unit 307 obtains the average frequency of a plurality of grouped specific frequency regions (frequency bands). That is, the natural frequency group extraction unit 307 obtains the average value of the frequencies of each group.

[0103] Next, the natural frequency group extraction unit 307 obtains the difference in the average frequency between a certain group and an adjacent group.

[0104] The natural frequency estimation unit 308 averages the difference values of the average frequencies between each group obtained by the natural frequency group extraction unit 307 to obtain an average difference value. That is, the natural frequency estimation unit 308 obtains the interval between the frequencies at which peaks appear (hereinafter also referred to as the "estimation interval").

[0105] The natural frequency estimation unit 308 obtains the distribution of the average difference values between each group and assumes the average difference value with the highest occurrence frequency as the natural frequency. Utilizing the characteristic that peaks appear at integer multiples of the natural frequency, if the intervals between the groups are approximately equal, those intervals can be regarded as the natural frequency. Therefore, here, the average difference value with the highest occurrence frequency among the average difference values between the groups is derived and assumed as the natural frequency.

[0106] Next, the natural frequency estimation unit 308 divides the average frequency of each group by the assumed natural frequency and checks that the quotient is close to an integer value. It should be noted that since peaks may not be detected at specific multiples, the average frequencies may not always be equally spaced.

[0107] If it is not possible to confirm that the quotient is close to an integer value, since the assumed natural frequency is incorrect, the natural frequency estimation unit 308 excludes abnormal data and starts over from the assumption of the natural frequency.

[0108] If it is possible to confirm that the quotient is close to an integer value, the natural frequency estimation unit 308 excludes abnormal values (= peaks that are not part of the natural frequency group) within the group and determines the natural frequency.

[0109] Here, an explanation will be given regarding the confirmation that the quotient is close to an integer value. For example, due to sensor characteristics such as being unable to measure frequency components below a certain frequency, sensors often cannot measure the natural frequency. Even in such cases, if the interval between the frequencies at which peaks appear is constant, the natural frequency can be estimated.

[0110] For example, assume that the natural frequency is 0.8 [Hz]. At this time, assume that peaks only appear at 3.2 [Hz], 4 [Hz], and 4.8 [Hz]. In such a case, since the difference value between the frequencies at which each peak appears is 0.8 [Hz], the natural frequency estimation unit 308 divides the frequencies 3.2 [Hz], 4 [Hz], and 4.8 [Hz] by the difference value 0.8 [Hz] to obtain integer values of 4 (times), 5 (times), and 6 (times) as multiples. In that case, the natural frequency can be determined to be 0.8 Hz.

[0111] However, in reality, the frequencies at which peaks appear are often shifted, such as 3.3 [Hz], 4.1 [Hz], 4.7 [Hz], etc., so it is difficult to mechanically estimate the natural frequency because clean integer values (multiple values) cannot be obtained.

[0112] Therefore, for example, it is decided in advance to confirm that the quotient becomes a value close to an integer value (multiple value) by a pre-determined fractional part processing (such as rounding, truncation, ceiling, etc.). In that case, since the quotient is assumed to be a value close to the planned integer value (multiple value), a certain range may be set in advance, and if it is within that range, it may be recognized that the quotient is an integer value.

[0113] Alternatively, as a different configuration, it is also possible to adopt a configuration in which the estimated natural frequency is specified in advance, and a group having an average frequency close to an integer multiple of the specified value is adopted as the natural frequency group.

[0114] [S8: Estimation of Tension] The control unit 302 estimates physical values such as tension using the natural frequency.

[0115] Thereby, for example, from the value of the natural frequency, it is possible to estimate the tension acting on the cable, etc., and it is possible to perform an abnormality determination such as whether the tension, etc. is as designed. Note that, for example, when the value of the natural frequency itself changes, it may be considered that there is an abnormality in the cable.

[0116] Also, in the embodiment, the case of obtaining the value of the cable tension is exemplified, but as long as a physical numerical value for determining the soundness of the structure can be derived based on the natural frequency, it is not limited to the tension.

[0117] For example, as exemplified in the above-described formula (1), it is known that there is a certain relationship between the natural frequency and the tension acting on the cable. When the structure to be subjected to the abnormality determination is a cable or the like, it may be necessary to consider other elements other than the natural frequency, and it is desirable to derive the tension including other elements and perform the abnormality determination. For example, in formula (1) including other elements, the tension can be derived using parameters that do not change such as the cable length.

[0118] For example, the control unit 302 may estimate the parameters using a method such as the least squares method from the distribution of the natural frequency group, and obtain the tension, etc. using the estimated parameters.

[0119] Incidentally, as a modification, when the tension changes beyond a determined range, the control unit 302 has a function of issuing a warning, so that the monitoring work can be reduced. At this time, a certain error is included in the estimation of the tension. Therefore, instead of issuing a warning based on only one measurement result, when the change in tension is detected, the acceleration sensor 203 is measured again to confirm that the abnormality (change in tension) continues. Then, when the change in tension continues, it is necessary to devise such that the control unit 302 issues a warning.

[0120] Also, when a plurality of acceleration sensors 203 are attached to a structure such as a cable, it is also possible to provide a function of making a determination by combining the estimation results of other sensors. For example, assume that two acceleration sensors 203 are attached to a cable, and the natural frequency is estimated using the measured values of each acceleration sensor 203 to obtain the tension. In that case, when both of the tensions estimated using the measured values of each acceleration sensor 203 change beyond the range, it may be determined that an abnormality has occurred in the cable.

[0121] Also, by repeatedly measuring with the acceleration sensor 203 over a long period of time and estimating the natural frequency using a large amount of measurement data, a highly accurate natural frequency can be estimated.

[0122] [Embodiment] FIG. 10 is a diagram showing the frequency for each group in the embodiment.

[0123] In FIG. 10, for example, the vicinity of 0.3 Hz is set as the 0th group, and the frequency at which a peak appears in the column of "f0" is shown. FIG. 10 shows the frequencies of the groups from f0 to f7.

[0124] First, the natural frequency group extraction unit 307 excludes f1 and f4 with a small number of data from the groups of f0 to f7 in FIG. 10, and sets six groups of f0, f2 to f3, and f5 to f7 as candidates for the natural frequency group.

[0125] For example, in this example, a case is illustrated where groups (f1, f4) with the number of data less than a threshold value (for example, 6) are excluded. However, the determination of extracting candidate groups of natural frequencies is not limited to the determination based on the number of data. For example, it may be determined based on the ratio of the number of data when peaks are detected with respect to the total number of samples (that is, the comparison between the value indicating the peak detection ratio and the threshold value). Also, since the selection of candidate groups of natural frequencies is very difficult, it is desirable to be able to appropriately set the threshold value according to the operation.

[0126] Next, the natural frequency group extraction unit 307 obtains the average frequency of each group of candidate groups of natural frequencies for each group. The values shown in the column of "(A) Average frequency value" in FIG. 10 correspond to this.

[0127] Furthermore, the natural frequency group extraction unit 307 obtains the difference value of the average frequencies between a certain group and the group adjacent to it. The values shown in the column of "(B) Difference value" in FIG. 10 correspond to this. Note that since there is no group existing before the group of f0, the value in the column of "(B) Difference value" of the group of f0 is set as the average frequency value.

[0128] Next, the natural frequency group extraction unit 307 obtains the average value of the difference values of each group. In the case of this example, the average difference value D obtained by averaging the difference values of each group is 0.757558 [Hz].

[0129] Here, there are values such as "1.381701" and "1.217227" that significantly exceed the average difference value D (0.75755 Hz). The natural frequency estimation unit 308 divides these values by the average difference value, and the numerical value obtained by rounding off the quotient thus obtained (for example, rounding, rounding up, rounding down, etc.) is "2".

[0130] For example, when "1.381701" is divided by the average difference value D "0.75755", the quotient is approximately 1.82391, and when rounded to the first decimal place, it becomes "2".

[0131] In that case, it can be inferred that no peak has been measured for the difference value "1.381701" between f0 and f2. The same applies to "1.217227".

[0132] Therefore, the natural frequency estimation unit 308 divides each of "1.381701" and "1.217227" by "2" to obtain the difference value of the average frequency of each group again. The values shown in the column of "(C) Estimation interval" in FIG. 10 correspond to this.

[0133] Then, when the natural frequency estimation unit 308 calculates the average value of the values of "(C) Estimation interval" for each group, it becomes "0.54098". This value is the estimated value C of the natural frequency.

[0134] In the above example, the case where the value exceeding the average difference value D is such that the numerical value obtained by dividing the average difference value D is "2" is shown, but there may be cases where the rounded value is an integer such as 3 or 4. For example, if the integer value obtained by rounding is "3", it can be determined that two peaks have not been measured between the difference values. In that case, the value obtained by dividing the difference value by "3" may be calculated, and the average value of the difference values of each group may be used as the estimated value of the natural frequency.

[0135] Furthermore, the natural frequency estimation unit 308 calculates for each group how many times the average frequency value (A) of each group is of the estimated value C of the natural frequency. The values in the column of "(D) How many times of C" in FIG. 10 correspond to this.

[0136] For example, the value obtained by dividing the average frequency value "0.304026" of the f0 group by the estimated value C of the natural frequency "0.54098" is "0.561991". The other groups are calculated in the same way.

[0137] The fundamental frequency estimation unit 308 performs rounding processing or the like, and if the value in the column of "How many times of (D) C" in FIG. 10 is close to an integer value, it can be determined that the estimated fundamental frequency value is correct. For example, to determine whether it is an integer value, the difference between the expected integer value and the actual value may be taken, and how much it deviates may be determined using a threshold value or the like. Another example is that rounding at the second decimal place or the like may be appropriately used for rounding processing.

[0138] In this example, values that are hard to say are close to integer values appear for f0, f5, f6, and f7. This means that the estimation of the fundamental frequency has failed.

[0139] Therefore, the fundamental frequency group extraction unit 307 and the fundamental frequency estimation unit 308 re-do the exclusion (extraction of the fundamental frequency group) of the group that is considered to be the cause of the failure in the estimation of the fundamental frequency.

[0140] For re-doing the method of estimating the fundamental frequency, various methods can be applied according to the situation. For example, in the above-described example, f0, which deviates the most from the expected integer value, is excluded, and the estimated value of the fundamental frequency is obtained again.

[0141] That is, again, the fundamental frequency estimation unit 308 obtains the difference value between each group (the "estimated interval (C2)" in FIG. 10) and obtains the estimated value C2 of the fundamental frequency. Here, since it is the second estimation of the fundamental frequency, it is expressed as the estimated value C2, and the estimated value C2 of the fundamental frequency is "0.562583".

[0142] Then, the fundamental frequency estimation unit 308 again obtains, for each group, how many times the average frequency value (A) of each group is the estimated value C2 of the fundamental frequency. For example, it is assumed that the values in the column of "How many times of (D2) C2" in FIG. 10 are generally integer values, and f2 can be regarded as 3 times, f3 as 4 times, f5 as 6 times, f6 as 7 times, and f7 as 8 times. In this example, the fundamental frequency is the estimated value C2 "0.562583", and f2, f3, f5, f6, and f7 are determined as the fundamental frequency group.

[0143] In the above-described example, the natural frequency estimation unit 308 uses the estimated value C2 “0.562583” as the natural frequency. However, the present invention is not limited to this. The natural frequency estimation unit 308 may alternatively use, as the natural frequency, a value obtained based on the average difference value between each group belonging to the determined natural frequency group. Thereby, a more accurate natural frequency can be obtained.

[0144] Then, the control unit 302 applies the above-described method and estimates the tension applied to the cable based on the obtained natural frequency.

[0145] (A-3) Effects of the Embodiment As described above, according to this embodiment, the natural frequency can be estimated from the frequency spectrum obtained by performing Fourier transform on the measurement value of the sensor. Furthermore, based on the natural frequency, physical values such as tension can be estimated, and the abnormality determination of the structure can be automatically performed.

[0146] Also, according to the embodiment, by using the estimated physical values such as tension to determine the abnormality of the structure and automatically issuing a warning using the result, it is possible to reduce the monitoring work.

[0147] (B) Other Embodiments Although various modified embodiments have been mentioned in the above-described embodiment, the present invention can also be applied to the following modified embodiments.

[0148] (B-1) FIGS. 8 and 9 show data obtained from one axis of a single sensor. However, an acceleration sensor can generally acquire data of three axes, x, y, and z, simultaneously. In which direction of the three axes the natural vibration of the structure occurs depends on the installation situation. Also, not all peaks may occur in a single direction. For example, there may be a situation where the peak at four times the natural frequency appears on the X axis and the peak at five times appears on the Y axis. Therefore, a configuration can also be adopted in which data from a plurality of axes and sensors are integrated and used.

[0149] (B-2) In the above-described embodiment, first, the natural frequency group extraction unit 307 extracts candidates for the natural frequency group, and then the natural frequency estimation unit 308 estimates the natural frequency. However, the extraction of candidates for the natural frequency group may be omitted. For example, without extracting candidates for the natural frequency group, the natural frequency estimation unit 308 may obtain the average frequency value of each group and estimate an integer multiple value in a similar manner to obtain the natural frequency. At that time, if necessary, since there is a process of repeating the estimation of the natural frequency, groups that are considered unnecessary at that time may be excluded.

[0150] (B-3) In the above-described embodiment, the case where the monitoring target is one identical structure is exemplified. However, the present invention can also be used as it is when monitoring a plurality of structures implemented simultaneously as one group.

[0151] (B-4) The various functions of the above-described monitoring device may be performed by physically the same device, or may be distributed and processed by different devices.

Explanation of Reference Numerals

[0152] 1... monitoring system, 10... master unit, 20... sensor device, 30... monitoring device, 201... communication unit, 202... control unit, 203... acceleration sensor, 204... schedule determination unit, 205... timer unit, 206... clock unit, 207... data storage unit, 301... communication unit, 302... control unit, 303... peak analysis unit, 304... grouping processing unit, 306... data accumulation unit, 307... natural frequency group extraction unit, 308... natural frequency estimation unit.

Claims

1. Analysis means for detecting peaks appearing in the frequency spectrum obtained based on the measurement signal from a sensor provided in the structure, and obtaining information regarding the peaks for each peak; Accumulation means for accumulating the information regarding the peaks of each of the peaks for each measurement; Grouping means for grouping the frequencies at which each of the peaks was detected into frequency bands based on the information regarding the peaks of each of the peaks obtained over a plurality of measurements; Natural frequency group extraction means for extracting a plurality of frequency bands belonging to the natural frequency group from among the plurality of frequency bands grouped by the grouping means; Natural frequency estimation means for estimating the natural frequency based on the plurality of frequency bands belonging to the extracted natural frequency group A monitoring device characterized by comprising the same.

2. The monitoring device according to claim 1, wherein the natural frequency estimation means estimates the natural frequency based on the estimated interval between peak frequencies adjacent to each other in the plurality of frequency bands belonging to the natural frequency group.

3. The natural frequency group extraction means obtains a difference value between the adjacent frequency bands using the average frequency values of the grouped frequency bands, and the natural frequency estimation means estimates the natural frequency based on an average difference value obtained by averaging the difference values between the adjacent frequency bands. The monitoring device according to claim 1 or 2, characterized by the above.

4. The natural frequency estimation means is such that for each of the grouped frequency bands, when the value obtained by dividing the difference value between the adjacent frequency bands by the average difference value approximates a value corresponding to an integer, the difference value is divided by the approximating integer to obtain a value of the estimated interval, and further, an average value of the estimated intervals for each of the grouped frequency bands is averaged to estimate the natural frequency. The monitoring device according to claim 3, characterized by the above.

5. The natural frequency estimation means is such that for each of the grouped frequency bands, when the value obtained by dividing the difference value between the adjacent frequency bands by the average difference value does not approximate a value corresponding to an integer, any one of the grouped frequency bands is excluded, and the estimation process of the natural frequency is performed again. The monitoring device according to claim 3 or 4, characterized by the above.

6. The monitoring device according to any one of claims 1 to 5, further comprising control means for deriving a numerical value indicating the state of the structure based on the natural frequency.

7. The monitoring device according to claim 6, wherein the control means estimates a parameter related to the derivation of a numerical value indicating the state of the structure based on the distribution of the natural frequency group, and derives a numerical value indicating the state of the structure using the parameter.

8. The monitoring device according to claim 6 or 7, wherein the control means issues a warning when the derived numerical value indicating the state of the structure deviates from a predetermined range or when it deviates from the range of values obtained in past measurements.

9. A computer, analysis means for detecting peaks appearing in a frequency spectrum obtained based on a measurement signal from a sensor provided in a structure, and obtaining information regarding the peaks for each peak; accumulation means for accumulating the information regarding the peaks for each measurement for each of the peaks; grouping means for grouping the frequencies at which the respective peaks are detected into frequency bands based on the information regarding the peaks for each of the peaks obtained over a plurality of measurements; natural frequency group extraction means for extracting a plurality of frequency bands belonging to a natural frequency group from among the plurality of frequency bands grouped by the grouping means; natural frequency estimation means for estimating a natural frequency based on the plurality of frequency bands belonging to the extracted natural frequency group characterized in that it functions as such.

10. The analysis means detects peaks appearing in a frequency spectrum obtained based on a measurement signal from a sensor provided in a structure, and obtains information regarding the peaks for each peak. The accumulation means accumulates the information regarding the peaks for each measurement for each of the peaks. The grouping means groups the frequencies at which the respective peaks are detected into frequency bands based on the information regarding the peaks for each of the peaks obtained over a plurality of measurements. The natural frequency group extraction means extracts a plurality of frequency bands belonging to a natural frequency group from among the plurality of frequency bands grouped by the grouping means. The natural frequency estimation means estimates a natural frequency based on the plurality of frequency bands belonging to the extracted natural frequency group. characterized by the above.

Images