Structure deterioration diagnosis system

Abstract

To provide a structure deterioration diagnosis system that performs deterioration diagnosis on a structure on the basis of a feature value that is not dependent on temperature.SOLUTION: A structure deterioration diagnosis system comprises: a sensor which is provided on a structure being diagnosed for deterioration and outputs acceleration information for the structure; and a sensor controller which determines whether deterioration of the structure has occurred on the basis of the acceleration information output from the sensor. The sensor controller calculates, as a feature value, an active load displacement which does not vary depending on temperature as an index value for determining deterioration of the structure on the basis of the acceleration information. The sensor controller also calculates an abnormality level to serve as a deterioration determination index of the structure on the basis of time series data of the feature value, and if the abnormality level exceeds a threshold, it determines that deterioration of the structure has occurred.SELECTED DRAWING: Figure 7

Description

[Technical field]

[0001] The present disclosure relates to a structure deterioration diagnosis system that can identify whether deterioration has occurred in a structure such as a bridge or whether an abnormality has occurred in a sensor when diagnosing deterioration of the structure using acceleration information detected by a sensor. [Background technology]

[0002] For example, structures such as bridges over which vehicles pass will gradually deteriorate due to the passage of vehicles. It is important to detect the deterioration of structures such as bridges before they break.

[0003] Conventional techniques for diagnosing bridge deterioration include diagnosing sensor abnormalities (failures) from changes in inclination or natural frequency calculated based on multiple sensors (see, for example, Patent Document 1). Specifically, the ratio of probability density distributions based on the detection results of multiple sensors deviates from an allowable range, making it possible to diagnose failures in the sensors themselves. As a result, a structure deterioration diagnosis system can be realized that can ensure the reliability of deterioration diagnosis after confirming the soundness of the sensors. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 6081867 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the conventional techniques have the following problems. In Patent Document 1, the detection results of multiple sensors are compared to diagnose sensor abnormalities (failures). That is, in order to identify whether deterioration of a structure or an abnormality in a sensor has occurred, it is necessary to compare the detection results of multiple sensors.

[0006] Furthermore, when comparing the probability density distribution of the slope between multiple sensors, it may not be possible to determine whether there is a structural abnormality or a sensor failure due to the influence of the sensor position or the temperature characteristics of the sensor.

[0007] The present disclosure has been made to solve the above-mentioned problems, and aims to provide a structure deterioration diagnosis system that can identify whether a structure abnormality or sensor failure occurs from the detection results of a single sensor. [Means for solving the problem]

[0008] The structure deterioration diagnosis system of the present disclosure comprises a sensor that is installed in a structure that is to be diagnosed for deterioration and outputs acceleration information of the structure, and a sensor controller that determines whether deterioration of the structure has occurred based on the acceleration information output from the sensor. The sensor controller calculates a live load displacement that does not change depending on temperature as a feature value to use as an index value for diagnosing deterioration of the structure based on the acceleration information, calculates an anomaly degree that serves as an index for determining deterioration of the structure based on the time series data of the feature degree, and determines that deterioration of the structure has occurred if the anomaly degree exceeds a threshold value. Effect of the Invention

[0009] According to the present disclosure, it is possible to obtain a structure deterioration diagnosis system that can distinguish between a structure abnormality and a sensor failure from the detection result of one sensor. [Brief description of the drawings]

[0010] [Figure 1] 1 is a configuration diagram of a structure deterioration diagnosis system according to a first embodiment of the present disclosure. [Diagram 2] 1 is an explanatory diagram showing a state in which sensors are installed in a structure that is a diagnosis target of a structure deterioration diagnosis system according to a first embodiment of the present disclosure. [Diagram 3]11A to 11C are diagrams illustrating a time transition of a slope, a relationship between a slope and a temperature, and a slope anomaly degree in a case where the slope is used as a feature amount in the first embodiment of the present disclosure. [Figure 4] 11A to 11C are diagrams illustrating a time transition of a natural frequency, a relationship between the natural frequency and temperature, and a natural frequency anomaly degree in a case where the natural frequency is used as a feature amount in the first embodiment of the present disclosure. [Diagram 5] 1A to 1C are diagrams showing a time progression of live load displacement, a relationship between live load displacement and temperature, and a live load displacement anomaly degree in a case where live load displacement is used as a feature quantity in embodiment 1 of the present disclosure. [Figure 6] 11 is an explanatory diagram for a case where a tilt, which is one of the feature amounts, is calculated using three-axis acceleration information output from a sensor according to the first embodiment of the present disclosure. FIG. [Figure 7] FIG. 2 is an explanatory diagram showing a case where only the tilt anomaly degree is in an abnormal state, and the natural frequency anomaly degree and the live load displacement anomaly degree are not in an abnormal state, in the first embodiment of the present disclosure. [Figure 8] FIG. 11 is a diagram comparing a histogram of each feature amount when the abnormality degrees for all feature amounts are determined to be in a normal state with a histogram of each feature amount when only the inclination abnormality degree is determined to be in an abnormal state in the first embodiment of the present disclosure. [Figure 9] FIG. 11 is an explanatory diagram for comparing a histogram of the degree of inclination abnormality with a histogram of natural frequencies using a statistical method in the first embodiment of the present disclosure. [Figure 10] 4 is a flowchart showing a series of processes for identifying whether there is a structural abnormality or a sensor failure, which is executed by the structure deterioration diagnosis system according to the first embodiment of the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] Hereinafter, preferred embodiments of the structure deterioration diagnosis system of the present disclosure will be described with reference to the drawings. The present disclosure has a technical feature of having the ability to distinguish between a structural abnormality or a sensor failure by determining multiple feature amounts from the detection results of a single sensor, and further determining an anomaly level corresponding to each of the multiple feature amounts, and performing deterioration diagnosis based on a combination of the multiple anomaly levels.

[0012] Embodiment 1 In the first embodiment, a specific configuration for distinguishing between an abnormality in a structure and a failure of a sensor when performing deterioration diagnosis of a structure based on the measurement result by one sensor will be described.

[0013] 1 is a configuration diagram of a structure deterioration diagnosis system according to a first embodiment of the present disclosure. The structure deterioration diagnosis system in the first embodiment is configured to include one sensor 10 and a sensor controller 20.

[0014] 2 is an explanatory diagram showing a state in which a sensor 10 is installed in a structure that is a deterioration diagnosis target of the structure deterioration diagnosis system according to the first embodiment of the present disclosure. In Fig. 2, a bridge 30 is shown as a specific example of a structure, with Fig. 2(A) being a side view of the bridge 30 and Fig. 2(B) being a rear view of the bridge 30.

[0015] The bridge 30 gradually deteriorates due to aging caused by the passage of the vehicle 1. Therefore, the structure deterioration diagnosis system according to the first embodiment diagnoses the ever-changing deterioration state of the bridge 30 by analyzing, with the sensor controller 20, acceleration information output by the sensor 10 installed at a position suitable for deterioration diagnosis.

[0016] Furthermore, the sensor controller 20 according to the first embodiment has a function of determining, based on the acceleration information outputted from one sensor 10, whether or not the sensor 10 is malfunctioning.

[0017] The sensor 10 is installed on a main girder 31, which is a component of a bridge 30. Here, the main girder 31 corresponds to the structural part of the structure to be diagnosed. As shown in FIG. 2(B), the main girder 31 is configured as three main girders 31a, 31b, and 31c, as an example. The example of FIG. 2(B) illustrates a case where the sensor 10 is installed in the center part of the main girder 31b (i.e., the center part of the span corresponding to the distance between the left and right supports 2). This installation position corresponds to an example of a position suitable for deterioration diagnosis.

[0018] In the following, a specific method will be described for diagnosing deterioration based on the detection result from one sensor 10 and identifying whether there is a structural abnormality or a sensor failure. However, the sensor 10 itself may be installed in a position different from that in the first embodiment, or multiple sensors 10 may be installed in multiple locations on the bridge 30.

[0019] When multiple sensors 10 are installed, the sensor controller 20 can identify for each sensor 10 whether a structural abnormality has been detected or whether a sensor failure has occurred, based on the individual detection results at the installation positions of each sensor 10.

[0020] The sensor 10 detects acceleration information occurring on the bridge 30, and outputs the acceleration information to the sensor controller 20. One example of the sensor 10 is a three-axis acceleration sensor that uses a thin-film quartz crystal oscillator, has excellent responsiveness, and has a measurement range capable of measuring acceleration from DC to several tens of Hz.

[0021] In this way, by using a three-axis acceleration sensor as the sensor 10, it is not necessary to perform leveling, and the sensor output can be obtained regardless of the direction of tilt or vibration. Therefore, if the direction of tilt can be specified by performing leveling, a two-axis or one-axis acceleration sensor may be used.

[0022] Furthermore, the sensor 10 can convert analog signals relating to the three-axis acceleration of the structure at the installation location into digital signals at a predetermined sampling rate (for example, a sampling rate of 50 Hz) and transmit the digital signals to the sensor controller 20 as acceleration information.

[0023] The sensor controller 20, which receives the acceleration information output from the sensor 10, is configured to include a feature conversion unit 21, an abnormality degree calculation unit 22, and an abnormality / fault determination unit 23, as shown in FIG.

[0024] The feature conversion unit 21 converts the acceleration information received from the sensor 10 into the inclination, natural frequency, and displacement of the structure at the installation position of the sensor 10. Here, the inclination, natural frequency, and displacement each correspond to a feature that serves as an index value for diagnosing the deterioration of the structure.

[0025] In the present embodiment, "displacement" refers to live load displacement. Here, live load refers to a load whose magnitude is not constant and whose acting position changes. Factors that cause displacement of such live load include the weight of the vehicle 1 passing over the bridge 30, the weight of the bridge 30 itself, and the inertial force acting on the bridge 30 due to an earthquake. In the following explanation, "displacement" and "live load displacement" are treated as synonyms.

[0026] The anomaly degree calculation unit 22 calculates the degree of anomaly for each of the multiple feature quantities, i.e., tilt, natural frequency, and live load displacement, generated by the conversion process by the feature quantity conversion unit 21. A method for calculating the degree of anomaly for each of the tilt, natural frequency, and live load displacement will be described in detail below.

[0027] 3A and 3B are diagrams showing the time transition of the slope, the relationship between the slope and temperature, and the slope abnormality degree when the slope is used as the feature amount in the first embodiment of the present disclosure. FIG. 3A shows the time transition of the slope, with the vertical axis representing the slope and the horizontal axis representing the date. It can be seen that the feature amount related to the slope changes depending on the temperature. Furthermore, FIG. 3A shows that it is difficult to set a threshold value for determining an abnormal state for the slope, which is the feature amount.

[0028] Figure 3(B) shows the relationship between the slope and temperature, with the vertical axis representing the slope and the horizontal axis representing the temperature. Figure 3(B) shows the “prediction distribution”, “normal data distribution”, and “abnormal data distribution”.

[0029] The "predictive distribution" is a distribution that can be created from the relationship between the slope and temperature when temperature is the input and the slope is the output, using time series data showing the transition state of the slope calculated during the "learning period" shown in Figure 3(A).

[0030] Specifically, the anomaly calculation unit 22 uses Gaussian process regression to model the relationship between the input temperature and the output slope from the time series data obtained during the learning period in the form of a response surface or a regression line. Furthermore, the anomaly calculation unit 22 can create, for example, a 3σ range as a "prediction distribution" from the modeled response surface or regression line.

[0031] In other words, the "prediction distribution" corresponds to a confidence interval determined by statistical processing using the covariance matrix of a Gaussian distribution estimated from time series data obtained during the learning period.

[0032] On the other hand, the "normal data distribution" and the "abnormal data distribution" are distributions obtained by plotting the time series data collected during the "assessment period" following the "learning period" shown in Figure 3(A), where the distribution of data that falls within the range of the predictive distribution corresponds to the normal data distribution, and the distribution of data that falls outside the range of the predictive distribution corresponds to the abnormal data distribution.

[0033] One of the important features of using Gaussian process regression is its nonlinearity, which makes it effective when the input-output relationship cannot be fitted well by linear regression. Another important feature is that it uses Bayesian estimation, which means that it is a nonparametric regression model.

[0034] Therefore, a specific method for calculating the degree of anomaly, which serves as an index for determining whether or not there is an "abnormal data distribution" state as shown in FIG. 3(B), will be further explained using formulas and FIG. 3(C).

[0035] The procedure for calculating the degree of anomaly using Gaussian process regression when the temperature is the input and the slope is the output is as follows: <Step 1> Given N sets of training data, the input training data is

number

number

number

[0036] <Step 2> Let the output of the response surface (or curve in the case of one dimension) at any coordinate x be f(x). If we assume that the model representing the noise during observation follows a normal distribution with a mean value of 0 and a variance value of σ2, then we obtain the following formula (1).

number

[0037] <Step 3> The distribution of the response surface value f(x) for the input x is calculated from the data D.

number

number

[0038] <Step 4> For the model of smoothness of the response surface, let the values ​​of the response surface at arbitrary inputs x and x' be f(x) and f(x'), respectively. At this time, f(x) and f(x') follow the probability distribution shown in the following formula (3).

number

[0039] Here, K is a kernel function, which intuitively represents "how similar x and x' are." As the kernel function, a RBF (radial basis function) kernel is often used.

[0040] In general, N inputs are given by the following formula (4):

number

number

[0041] This is the basic assumption of the Gaussian process. However, K(x, x') is an N×N matrix if there are N inputs, and a 2×2 matrix if there are two inputs, as shown in equation (3) above.

[0042] <Step 5> The degree of anomaly α(x') is the mean of the predictive distribution of GPR.

number

number

number

[0043] That is, the degree of anomaly corresponding to each feature amount can be calculated based on the prediction distribution (corresponding to the confidence interval) calculated during the learning period.

[0044] The degree of anomaly regarding the slope calculated over the learning period and the evaluation period using the above procedure is the degree of anomaly in slope shown in Fig. 3(C).By setting an appropriate threshold and determining whether or not there is an anomaly that is equal to or greater than the threshold, it is possible to quantitatively determine whether the abnormal data distribution state shown in Fig. 3(B) has occurred.

[0045] By performing such a series of processes, the anomaly degree calculation unit 22 can convert the slope, which is a feature amount, into an anomaly degree, which is an index value for determining the presence or absence of an abnormal state with high accuracy. As a result, when using the slope as a feature amount, it has been difficult to set a threshold value for determining an abnormal state for the slope because the degree of temperature dependency differs for each sensor, but it is now possible to easily set a threshold value for determining an abnormal state for the slope anomaly degree calculated from the slope.

[0046] That is, the sensor controller 20 has a function for making a judgment using the degree of inclination abnormality, and thus can accurately identify an abnormal state based on the inclination while taking into account the influence of temperature fluctuations.

[0047] Next, a case where the natural frequency is used as the feature amount will be described. Fig. 4 is a diagram showing a time transition of the natural frequency, a relationship between the natural frequency and temperature, and a natural frequency anomaly degree in a case where the natural frequency is used as the feature amount in the first embodiment of the present disclosure.

[0048] Figure 4(A) shows the time transition of the natural frequency, with the vertical axis representing the natural frequency and the horizontal axis representing the date. It can be seen that the feature quantity related to the natural frequency, like the inclination, changes depending on the temperature. Therefore, like the inclination, it is also difficult to set a threshold value for determining an abnormal state in the case of the natural frequency.

[0049] Therefore, similarly to the case of the slope described above in FIG. 3, when the natural frequency is used as the feature, the relationship between the natural frequency and temperature shown in FIG. 4(B) and the natural frequency anomaly degree shown in FIG. 4(C) can be obtained.

[0050] Although a detailed description is omitted, by carrying out a series of processes similar to those in the case of the inclination, the anomaly degree calculation unit 22 can convert the natural frequency, which is a feature amount, into an anomaly degree, which is an index value for determining the presence or absence of an abnormal state with high accuracy. As a result, when the natural frequency is used as a feature amount, it has been difficult to set a threshold value for determining an abnormal state for the natural frequency, but it is now possible to easily set a threshold value for determining an abnormal state for the natural frequency anomaly degree calculated from the natural frequency.

[0051] In other words, the sensor controller 20 has a function for making a judgment using the natural frequency abnormality degree, and thus can accurately identify an abnormal state based on the natural frequency while taking into account the influence of temperature fluctuations.

[0052] Next, a case where live load displacement is used as a feature quantity will be described. Fig. 5 is a diagram showing a time transition of live load displacement, a relationship between live load displacement and temperature, and a live load displacement anomaly degree in a case where live load displacement is used as a feature quantity in the first embodiment of the present disclosure.

[0053] Figure 5(A) shows the time transition of live load displacement, with the vertical axis representing live load displacement and the horizontal axis representing date. Unlike the inclination and natural frequency, the characteristic quantities related to live load displacement do not change depending on temperature. The reason that live load displacement does not change depending on temperature is that the most frequent value in the maximum displacement amount is more influenced by the amount of travel, speed, and weight of the vehicle, and is less influenced by temperature.

[0054] As shown in Figure 5(B), even when using the live load displacement, which is a feature that does not depend on temperature, the “prediction distribution” can be calculated as a distribution equivalent to the confidence interval using the covariance matrix of the Gaussian distribution estimated from the time series data obtained during the learning period.

[0055] As an example, when using live load displacement, which is a feature independent of temperature, the anomaly calculation unit 22 may use, for example, Hotelling's T2 method to calculate the anomaly α(x') according to the following equation (7).

number

[0056] In the above formula (7), x' represents data, μ with a ^ represents the average value, and σ with a ^ represents the variance value.

[0057] When the live load displacement is used as the feature quantity, a threshold value for determining an abnormal state for the live load displacement can be set as shown in FIG. 5(A).

[0058] Furthermore, as shown in Figure 5(C), in the same way as in the case where the inclination and natural frequency that change depending on temperature are used as feature quantities, even when the live load displacement is used as the feature quantity, by calculating the live load displacement abnormality degree using the above equation (7) or the like, an abnormal state can be easily identified using a threshold value.

[0059] The thresholds for the inclination anomaly, natural frequency anomaly, and live load displacement anomaly can be set in advance, or they can be set statistically based on the results of time-series data acquired during the learning period. For example, the thresholds can be set to 3σ, where σ is the standard deviation, or the value of the following formula (8).

number

[0060] In this way, by comparing the degree of abnormality calculated based on the time series data of the feature with a threshold value for the degree of abnormality, it is possible to easily quantitatively identify whether the feature indicates an abnormal state, regardless of whether the feature changes depending on the temperature.

[0061] Next, a method for identifying whether a structure is abnormal or a sensor is broken will be described, which is a technical feature of the structure deterioration diagnosis system according to the first embodiment. As described above, by using an index called an abnormality degree, it is possible to easily identify whether a feature is in an abnormal state or not. However, such an abnormal state can occur both when a structure becomes abnormal and when a sensor is broken.

[0062] Therefore, in the structure deterioration diagnosis system according to the first embodiment, the inclination anomaly degree, the natural frequency anomaly degree, and the live load displacement anomaly degree calculated based on the three feature quantities of inclination, natural frequency, and live load displacement are used as deterioration diagnosis indices to determine whether the structure is abnormal or whether the sensor is broken. This identification process will now be described in detail.

[0063] In the structure deterioration diagnosis system according to the first embodiment, a series of processes from [Process 1] to [Process 7] described below is performed for three abnormality levels to distinguish between a structure abnormality and a sensor failure. [Process 1] Based on the acceleration information output from the sensor 10, the feature conversion unit 21 calculates three feature amounts, namely, the inclination, natural frequency, and live load displacement of the structure.

[0064] [Process 2] The anomaly calculation unit 22 calculates the inclination anomaly degree, the natural frequency anomaly degree, and the live load displacement anomaly degree as an indicator for determining deterioration of the structure for each of the three feature quantities, that is, the inclination, the natural frequency, and the live load displacement.

[0065] [Process 3] The abnormality / failure determination unit 23 determines that an abnormal tilt has occurred if the degree of abnormal tilt exceeds the first threshold value. [Process 4] The abnormality / failure determination unit 23 determines that an abnormality in the natural frequency has occurred if the degree of abnormality in the natural frequency exceeds the second threshold value. [Process 5] The abnormality / fault determination unit 23 determines that a live load displacement abnormality has occurred if the live load displacement exceeds the third threshold value.

[0066] [Process 6] If the anomaly / failure determination unit 23 determines that a tilt anomaly has occurred and that neither a natural frequency anomaly nor a live load displacement anomaly has occurred, it determines that a sensor failure has occurred. [Process 7] The anomaly / failure judgment unit 23 judges that deterioration of the structure has occurred if it determines that at least one of a natural frequency anomaly and a live load displacement anomaly has occurred, regardless of whether an inclination anomaly has occurred or not.

[0067] The reason why process 6 and process 7 make it possible to distinguish between a structural abnormality and a sensor failure will be described with reference to Fig. 6. Fig. 6 is an explanatory diagram of a case where a tilt, which is one of the feature quantities, is calculated using triaxial acceleration information output from the sensor 10 in the first embodiment of the present disclosure.

[0068] ax, ay, az are the acceleration values ​​for each axis. However, to calculate the static inclination excluding the effects of vibration, low-pass filtering is performed for each axis using a digital filter.

[0069] Therefore, the inclination is calculated from the direct current (DC) component of each axis excluding the vibration component. This DC component may change due to the deterioration of electronic components such as the power supply inside the sensor 10. On the other hand, the natural frequency and live load displacement are calculated from the vibration and not the DC component, so they are less susceptible to changes in the DC component due to sensor failure, etc.

[0070] Normally, if there is an abnormality in the inclination, it is possible that there is a decrease in rigidity, such as damage to the girder or support, and that the decrease in rigidity affects the live load displacement and natural frequency in the same way as the inclination. Therefore, if only the inclination shows an abnormal state, it is assumed that the electronic components inside the sensor have deteriorated or been damaged, that is, the sensor 10 has failed.

[0071] On the other hand, it is unlikely that a sensor failure would cause an abnormality only in the natural frequency. The reason for this is that when the natural frequency is used as an index for determining the failure state, it is less susceptible to the effects of DC components because it deals with vibrations as described above.

[0072] In addition, the natural frequency is greatly influenced by the moisture content of the entire bridge, the temperature of the entire bridge, the condition of the bearings, etc., and so varies greatly to begin with. Therefore, if an abnormality occurs in the natural frequency, it is assumed that the cause is a decrease in rigidity.

[0073] Similarly, it is unlikely that a sensor failure would cause an abnormality only in the live load displacement. The reason for this is that when the live load displacement is used as an index for judging the failure state, the live load displacement is also calculated from the vibration component of the acceleration value (for example, around 1 Hz, which does not include DC components) as mentioned above.

[0074] Therefore, the live load displacement is not easily affected by changes in the DC component due to deterioration of the internal electronic components, such as deterioration around the power supply. Therefore, if an abnormality occurs in the live load displacement, it is assumed that the cause is a decrease in the rigidity of the structure.

[0075] 7 is an explanatory diagram showing a case where only the tilt anomaly degree is in an abnormal state, and the natural frequency anomaly degree and the live load displacement anomaly degree are not in an abnormal state, in the first embodiment of the present disclosure. This case corresponds to the above-mentioned process 6, and the anomaly / failure determination unit 23 can determine that a failure has occurred in the sensor 10 because a tilt anomaly has occurred, and neither the natural frequency anomaly nor the live load displacement anomaly has occurred.

[0076] In addition, when determining whether or not tilt abnormality has occurred and neither natural frequency abnormality nor live load displacement abnormality has occurred, the abnormality / failure determination unit 23 can use a statistical determination method using a histogram or the like in addition to a determination method based on whether or not each abnormality degree exceeds a threshold value. Therefore, the statistical determination method using a histogram will be described with reference to Figs. 8 and 9.

[0077] 8 is a diagram comparing (A) histograms of each feature amount when the anomaly degrees for all feature amounts are determined to be normal, and (B) histograms of each feature amount when only the inclination anomaly degree is determined to be abnormal, in the first embodiment of the present disclosure. Comparing FIG. 8(A) with FIG. 8(B), the frequency distribution of the inclination anomaly degree is clearly different. On the other hand, there is no significant difference between FIG. 8(A) and FIG. 8(B) in the distribution of the natural frequency anomaly degree and the live load displacement anomaly degree.

[0078] In this way, if only the anomaly of the slope increases, it is possible to determine that only the anomaly calculated from the slope has increased by comparing the anomaly of the slope with the anomaly calculated from other physical quantities using a statistical method.Specific statistical methods that can be used include the T value from Welch's t-test, or the density ratio.

[0079] 9 is an explanatory diagram of a case where a histogram of the inclination anomaly degree and a histogram of the natural frequency are compared using a statistical method in the first embodiment of the present disclosure. (A) to (D) in FIG. 9 show the following diagrams. Figure 9(A): Histogram of each feature when the degree of anomaly for the two feature quantities, tilt and natural frequency, is judged to be in the normal state. Figure 9(B): Histogram of each feature when, of the anomalies related to the two features, tilt and natural frequency, only the tilt anomaly is judged to be in an abnormal state and the natural frequency anomaly is judged to be normal. FIG. 9(C): Time series data of T values ​​from Welch's t-test calculated based on the histograms in FIG. 9(A) and FIG. 9(B). Figure 9(D): Time series data of density ratio calculated based on the histograms in Figures 9(A) and 9(B).

[0080] The T-value of the Welch's t-test can be calculated using the average value, variance, and number of samples for the slope anomaly and natural frequency anomaly, respectively, according to the following formula (9):

number

[0081] In addition, the density ratio can be calculated using the probability density of the tilt anomaly and the probability density of the natural frequency anomaly according to the following formula (10).

number

[0082] 9(C) and 9(D), the T value and the density ratio are both greater in the transition state during the judgment period than in the transition state during the learning period. Therefore, the abnormality / failure judgment unit 23 can judge that only the inclination abnormality degree has become abnormal based on the comparison result between the transition state during the judgment period and a threshold value obtained from the transition state during the learning period or a preset threshold value.

[0083] In the example of FIG. 9 described above, a comparison is made using the inclination and the natural frequency as the feature quantities. However, a similar statistical method can be applied to a comparison using the inclination and the live load displacement as the feature quantities, and a detailed description thereof will be omitted.

[0084] Next, a series of processes for identifying whether a structure abnormality or a sensor failure is caused from a detection result by one sensor by the structure deterioration diagnosis system in the present embodiment 1 will be described with reference to a flowchart. Fig. 10 is a flowchart showing a series of processes for identifying whether a structure abnormality or a sensor failure is caused, which is executed by the structure deterioration diagnosis system according to the embodiment 1 of the present disclosure.

[0085] First, in step S1001, the feature conversion unit 21 acquires acceleration information output from the sensor 10, and also sequentially acquires temperature information from the outside, which corresponds to the air temperature when the acceleration information is generated.

[0086] Next, in step S1002, the feature amount conversion unit 21 converts the acceleration information into physical amounts (tilt, natural frequency, and live load displacement) corresponding to the three feature amounts.

[0087] Next, in step S1003, the anomaly degree calculation unit 22 calculates the tilt anomaly degree, the natural frequency anomaly degree, and the live load displacement anomaly degree based on each converted physical quantity. Note that, when calculating the tilt anomaly degree and the natural frequency anomaly degree, the anomaly degree calculation unit 22 takes temperature information into consideration, thereby suppressing the influence of temperature-dependent changes in the determination process using the tilt anomaly degree and the natural frequency anomaly degree, and realizing stable determination process.

[0088] Next, in step S1004, the abnormality / failure determination unit 23 determines whether the tilt abnormality degree has become abnormal. If the abnormality / failure determination unit 23 determines that the tilt abnormality degree has become abnormal, the process proceeds to step S1007, and if the abnormality / failure determination unit 23 determines that the tilt abnormality degree has not become abnormal, the process proceeds to step S1005.

[0089] When the process proceeds to step S1005, the abnormality / failure determination unit 23 determines whether or not at least one of the live load displacement abnormality degree and the natural frequency abnormality degree has become abnormal. If the abnormality / failure determination unit 23 determines that at least one of the abnormality degrees has become abnormal, the process proceeds to step S1006, where it determines that a structural abnormality has occurred.

[0090] Furthermore, if the abnormality / failure judgment unit 23 judges in step S1005 that both the live load displacement abnormality degree and the natural frequency abnormality degree are in a normal state, all abnormality degrees are in a normal state, so the process returns to step S1001 and repeats the processes from step S1001 onwards.

[0091] Furthermore, when the process proceeds to step S1007, the abnormality / failure judgment unit 23 judges whether or not both the live load displacement abnormality degree and the natural frequency abnormality degree are in a normal state. Then, when the abnormality / failure judgment unit 23 judges that at least one of the abnormality degrees has become abnormal, the process proceeds to step S1006, and it is determined that a structural abnormality has occurred.

[0092] On the other hand, if the abnormality / failure judgment unit 23 judges in step S1007 that both the live load displacement abnormality degree and the natural frequency abnormality degree are in a normal state, it proceeds to the processing of step S1008, and determines that a sensor failure has occurred since only the tilt abnormality degree is in an abnormal state.

[0093] In addition, when the abnormality / failure judgment unit 23 is configured with multiple sensors 10 and judges that only the degree of tilt abnormality is in an abnormal state based on the respective acceleration information from two or more sensors 10, it is also possible to determine that a minor structural abnormality has occurred.

[0094] As described above, according to the first embodiment, the system has a function of calculating multiple feature amounts from acceleration information detected by one sensor, determining the tilt anomaly degree, natural frequency anomaly degree, and live load displacement anomaly degree corresponding to each feature amount, and identifying whether there is a structural abnormality or a sensor failure from the combination of the states of these anomaly degrees. In this way, by performing the identification process using multiple anomaly degrees calculated based on acceleration information from one sensor, it is possible to identify with high accuracy whether there is a structural abnormality or a sensor failure from the detection result of one sensor.

[0095] In the above-mentioned first embodiment, the case where the three feature quantities, the inclination, the natural frequency, and the live load displacement, are used has been described, but the present invention is not limited to this. By using two feature quantities, that is, the inclination as the first feature quantity and a feature quantity other than the inclination as the second feature quantity, it is possible to obtain a structure deterioration diagnosis system that can distinguish between a structure abnormality and a sensor failure from the detection result of one sensor.

[0096] As the second feature, the natural frequency or live load displacement described in the first embodiment can be used, and any other feature can also be used as the second feature as long as it is an index value for diagnosing the deterioration of a structure.

[0097] In this case, if only an inclination abnormality occurs, it is possible to determine that a sensor failure has occurred, and if it is determined that a natural frequency abnormality or a live load displacement abnormality has occurred based on the second feature regardless of whether an inclination abnormality has occurred or not, it is possible to determine that deterioration of the structure has occurred. [Explanation of symbols]

[0098] 10 sensor, 20 sensor controller, 21 feature conversion unit, 22 anomaly degree calculation unit, 23 anomaly / fault determination unit.

Claims

1. a sensor that is installed in a structure that is a deterioration diagnosis target and outputs acceleration information of the structure; a sensor controller that determines whether or not deterioration of the structure has occurred based on the acceleration information output from the sensor; Equipped with The sensor controller includes: calculating a live load displacement that does not change depending on temperature as a feature value, as an index value for diagnosing deterioration of the structure based on the acceleration information; Calculating an abnormality degree serving as an index for determining deterioration of the structure based on the time-series data of the feature amount; If the degree of abnormality exceeds a threshold, it is determined that deterioration of the structure has occurred. Structural deterioration diagnosis system.

2. The sensor controller calculates the degree of anomaly by using Hotelling's T2 method based on the time series data of the live load displacement, which is the feature amount. The structure deterioration diagnosis system according to claim 1 .

3. The sensor controller uses a preset value as the threshold value, or uses a value that is statistically set based on the acquisition result of the time series data of the feature amount over a preset learning period. The structure deterioration diagnosis system according to claim 1 or 2.

Images