JP2024045515A - Structure diagnosis system, structure diagnosis method, and structure diagnosis program - Google Patents

Abstract

To enable a quantitative abnormality diagnosis of a structure to be performed at high accuracy.SOLUTION: A structure diagnosis system has a diagnosis part for calculating an evaluation index being a ratio between a first periphery likelihood representing a probability distribution of a feature amount in the case of assuming that a structure is in a healthy state and a second periphery likelihood representing a probability distribution of a feature amount in the case of assuming that the structure is not in the healthy state about a feature amount showing a state of the structure generated from a time series of information about the structure to diagnose the state of the structure on the basis of the evaluation index. An autoregressive model represents sensor information acquired from a plurality of sensors attached to the structure at a certain time by a linear combination of time series of sensor information acquired before the certain time. A regression coefficient is a matrix obtained by being multiplied to each of the time series of sensor information linearly combined in the autoregressive model. The feature amount is information based on a coefficient matrix obtained by connecting matrixes.SELECTED DRAWING: Figure 3

Description

The present invention relates to a structure diagnosis system, a structure diagnosis method, and a structure diagnosis program.

Conventionally, features (vibration characteristics) such as the natural frequency of vibration, damping coefficient, and mode shape are estimated based on the acceleration of structures such as bridges, and the structure is determined based on changes in the estimated vibration characteristics. There is technology to diagnose abnormalities in

For example, Patent Document 1 discloses the following technology. In particular, independent component analysis (ICA) is performed on sensor signals indicating vibration obtained from acceleration sensors installed at multiple locations on the bridge, and the natural frequency of the bridge is determined by performing spectral analysis on the independent vibration components obtained by the independent component analysis. Then, based on a comparison between the natural frequency of the bridge in a healthy state obtained in advance and the natural frequency of the bridge in its current state, an abnormality diagnosis is performed on the entire bridge and abnormal locations are identified.

Japanese Patent Application Publication No. 2008-255571

However, the sensitivity to changes in damage varies depending on the location and vibration mode where vibration measurement is performed. Furthermore, in the prior art, estimation variations occur depending on the environmental conditions (external force, temperature, etc.) at the time of vibration measurement. However, it is difficult even for an expert to appropriately preset the installation location and vibration mode of the acceleration sensor. For this reason, when diagnosing an abnormality in a bridge based on natural frequencies, for example, if the vibration mode is incorrectly set, changes in vibration characteristics due to bridge abnormalities and accidental changes in vibration characteristics due to non-bridge abnormalities can be detected. There is a problem in that quantitative abnormality diagnosis cannot be performed with high accuracy, such as inability to discriminate.

The present invention takes the above-mentioned points into consideration and aims to enable quantitative abnormality diagnosis of structures to be performed with high accuracy.

In order to solve the above-mentioned problems, the present invention provides a structure diagnosis system that determines whether the structure is in a healthy state with respect to a feature value indicating the state of the structure generated from a time series of information regarding the structure. a first marginal likelihood representing the probability distribution of the feature amount when it is assumed that the structure is not in the healthy state; and a second marginal likelihood representing the probability distribution of the feature amount when it is assumed that the structure is not in the healthy state. a diagnosis unit that calculates an evaluation index that is a ratio of , and diagnoses the state of the structure based on the evaluation index; and an acquisition unit that acquires sensor information in time series from a plurality of sensors attached to the structure. and an autoregressive model generation unit that generates an autoregressive model that expresses the sensor information acquired at a certain time by the acquisition unit as a linear combination of time series of the sensor information acquired before the certain time. a feature quantity generating unit that generates the feature quantity indicating the state of the structure at the certain time based on the regression coefficient of the autoregressive model; the acquiring unit; and the autoregressive model generating unit; a sensor node configured to include the feature amount generation section and a transmission section that transmits the feature amount generated by the feature amount generation section to the diagnosis section, and at each time in the time series. The sensor information acquired from the plurality of sensors is a sensor information vector at each time whose order is the number of the sensors, and the autoregressive model is based on the sensor information acquired at the certain time by the acquisition unit. The information vector is expressed by a linear combination of time series of the sensor information vectors acquired before the certain time, and the regression coefficient is expressed by the linear combination of the sensor information vectors obtained before the certain time, and the regression coefficient is the linear combination of the sensor information vectors obtained before the certain time. It is a matrix by which each of the series is multiplied, and the feature amount is information based on a coefficient matrix that is a combination of the matrices.

According to the present invention, quantitative abnormality diagnosis of a structure can be performed with high accuracy.

1 is a diagram showing a schematic configuration of a diagnostic system according to a first embodiment; FIG. FIG. 2 is a block diagram showing the configuration of a sensor node according to the first embodiment. 1 is a block diagram showing the configuration of a diagnostic device according to a first embodiment; FIG. FIG. 4 is a sequence diagram showing a reference time process in the diagnostic system of the first embodiment. FIG. 4 is a sequence diagram showing a diagnosis process in the diagnosis system of the first embodiment. FIG. 4 is an explanatory diagram of a diagnosis process in the diagnosis system of the first embodiment. 11A and 11B are diagrams showing a comparison of the data amounts of coefficient matrices and principal component matrices for each number of sensors in the first embodiment and the prior art; 7 is a flowchart showing diagnostic processing in the diagnostic system of Embodiment 2. 13A and 13B are diagrams for explaining a method of calculating a threshold value for determining an abnormality according to the second embodiment. 13A to 13C are diagrams for explaining another example of the abnormality determination method using a plurality of threshold values according to the second embodiment. FIG. 7 is a diagram for explaining experimental results of abnormality determination performed using the diagnostic system of Embodiment 2; 7 is a diagram showing execution conditions of an experiment conducted using the diagnostic system of Embodiment 2. FIG. FIG. 7 is a diagram showing the results of an experiment conducted using the diagnostic system of Embodiment 2. 13 is a diagram showing the change in the natural frequency of a bridge that differs depending on the vibration mode measured under the same conditions as in the experiment conducted using the diagnostic system of embodiment 2. FIG. A graph showing the time-dependent changes in the outside temperature and Bayes factor values for a certain girder of a bridge during a certain period of time. A graph showing the change in deflection and Bayes factor values over time for a bridge over a certain period of time. A diagram showing a comparison of changes over time in Bayes factor values between two spans of a bridge. 13 is an explanatory diagram of sequential updating of a coefficient matrix in the diagnostic system according to the third embodiment. FIG. 7 is a diagram showing convergence of the probability distribution of the coefficient matrix by sequentially updating the coefficient matrix in the diagnosis system of Embodiment 3; FIG. 2 is a block diagram showing the configuration of a sensor node terminal used as a sensor node according to an embodiment.

The following describes embodiments of the present invention with reference to the drawings. The following embodiments are examples for the purpose of fully explaining the present invention, and have been omitted or simplified as appropriate. Not all of the configurations and processes described in the following embodiments are essential for implementing the present invention, and may be omitted as appropriate. In addition to the following embodiments, the present invention can be implemented in various other forms that can achieve the object of the present invention. Furthermore, forms that combine some or all of the embodiments and variations are included in the embodiments of the present invention, so long as they are consistent within the scope of the technical concept of the present invention.

In the following embodiments, configurations and processes described later that are similar to configurations and processes already described may be given the same reference numerals and explanations may be omitted, with only the differences being explained.

In the following embodiment, a bridge is used as an example of a structure to be diagnosed for the presence or absence of abnormalities such as cracks or other damage and deterioration. However, the present invention is not limited to bridges, and can be applied to any civil engineering or architectural structure that serves as social infrastructure, such as tunnels, road structures, river structures, port structures, water supply and sewerage systems, and buildings, to diagnose the presence or absence of abnormalities.

In the following description of the embodiment, an expression in which a first symbol is followed by a second symbol, such as "A^", is the same as an expression in which the second symbol is written directly above the first symbol.

<Mathematical background of the present invention>
First, before explaining the embodiments, the mathematical background of the present invention will be explained. In the present invention, the system matrix of the state equation in the state space model of a structure is defined by a matrix of regression coefficients of an autoregressive model that expresses an observed signal at a certain time as a linear combination of the time series of observed signals before a certain time. It uses things that can be approximated. In the present invention, we focus on the fact that the matrix of regression coefficients contains various information regarding the state of the structure included in the system matrix, and use the matrix of regression coefficients to represent the state of the structure. Evaluate the condition of the structure by analyzing the

Although the sensor information acquired by the sensor will be described below as acceleration, it is not limited to acceleration and may be other physical quantities.

Generally, the equation of motion of a structure is expressed as shown in equation (1) below. It is assumed that n sensors (n is a natural number of 1 or more) are installed at each position in a structure, and a time series of n-th observation vectors is acquired using the installed positions as observation points.

The equation of motion of equation (1) above can be converted into an equation of state as shown in equations (2-1) and (2-2) below.

Here, y in the above formula (2-2) represents the observation vector, z in the above formula (2-3) represents the state variable vector, and A _s in the above formula (2-5) represents the system matrix that represents the state of the system (structure to be diagnosed) in state space. C in the above formula (2-2) is a matrix that associates the observation vector y with the state of the system, and becomes a unit matrix when the measurement information at the observation point is regarded as the state of the system as it is.

Now, it is known that the state equations of equations (2-1) and (2-2) above can be expressed by an autoregressive model that is a linear combination of time series of discretized observed signals as shown in equation (3) below. It is being

In the above formula (3), k is the time index, y(k) is the sensor information vector at time k, p is the order of the autoregressive model, _Ai is the matrix of regression coefficients, and e(k) is the error term of the autoregressive model at time k. The matrix _Ai is the regression coefficient multiplied by each of the time series of the linearly combined sensor information vector y(ki) in the autoregressive model.

Each element of the matrix A _i , which is the regression coefficient in the autoregressive model of the above equation (3), is associated with the physical quantity of the structure included in the system matrix A _s of the above equation (2-1), and is associated with the physical quantity of the structure where the sensor is installed. It includes at least one of the following information: displacement, velocity, acceleration, mass matrix m, damping coefficient matrix c, and stiffness matrix k at the observation point. Therefore, by using the matrix A _i , the state of the structure can be expressed so as to include various information regarding vibration. That is, by capturing changes in the matrix A _i , it becomes possible to capture changes in the state of the structure.

However, when there is one sensor (n=1), the matrix A _i in the autoregressive model is a scalar.

Here, the matrix _Ai is an n-th order square matrix as shown in the following formula (4): n in the following formula (4) is the number of sensors installed at each position of the structure.

Then, if the p-th order autoregressive model shown in the above formula (3) generated from the time series of the sensor information vector y(k) measured by n sensors is expressed as a determinant, it becomes as shown in the following formula (5).

Here, in the above formula (5), Y _f on the left side is the predicted state, Y _p on the first term on the right side is the past acceleration of the pth order, and E on the second term on the right side is the uncertainty of the observation of the state of the structure. represents the error caused by In addition, the coefficient matrix A of the first term on the right side in the above equation (5) is an n×(n×p) matrix that combines the matrices A _i for i from 1 to p, and is as shown in the following equation (6). is expressed in

The coefficient matrix A inherits the information of the system matrix A _s . For example, the poles z _k , z _k ^* of the system at time k expressed using the natural frequency ω _k and damping coefficient h _k of the system at time k can be obtained from the above equation (5) as shown in the following equation (7).

The natural frequency ω _k and the damping coefficient h _k obtained by converting the posterior distribution of the coefficient matrix A into the posterior distribution of the poles as shown in the above equation (7) can also be used for diagnosing the structure. The natural frequency ω _k and the damping coefficient h _k obtained in this way have an advantage over the conventional technology in that vibration characteristics such as the natural frequency and the damping coefficient can be identified while taking into account variations. Furthermore, by converting the posterior distribution of the coefficient matrix A into a posterior distribution of poles and finding vibration characteristics from the poles with small variance, it is possible to automate the selection of physically meaningful vibration characteristics. Furthermore, selection of model order p can be automated based on BIC (Bayesian information criterion).

By transmitting the coefficient matrix A obtained as described above to the diagnostic device as a feature value representing the state of the structure, the diagnostic device can reproduce a time series having the vibration characteristics of the structure and determine the state of the structure. Diagnosis becomes possible.

However, the coefficient matrix A contains components of poor quality information that are not related to the diagnosis result. By removing these components of poor quality information and sending the reconstructed matrix A^ to the diagnosis device, the amount of data sent and received can be reduced. This matrix A^ is called the principal component matrix because it is found by principal component analysis of the coefficient matrix A.

For example, the coefficient matrix A estimated at the reference time (e.g., when the structure is healthy) is subjected to singular value decomposition as shown in the following equation (8).

The matrix U and the matrix P ^T (where P ^T is the transposed matrix of the matrix P) in the above equation (8) are orthogonal matrices, and the matrix Λ is the eigenvalue of the coefficient matrix A. Then, the right side of the above equation (8) _{is determined} by the principal component analysis as shown in the following equation (9). It is decomposed into.

This matrix _U1 at the reference time (e.g., healthy time) is a matrix that projects the coefficient matrix A in the probability space onto the principal component space. As shown in the following formula (10), ^the transposed matrix _U1T of the matrix _U1 is applied to the coefficient matrix A estimated at each time after the reference time to generate the principal component matrix A^, which is an N×np matrix from which components of poor quality information have been removed.

As described above, by estimating the coefficient matrix A that inherits information from the system matrix A _s , generating the principal component matrix A^ from the coefficient matrix A, and capturing changes in the principal component matrix A^, it is possible to detect abnormalities in the structure. In the present invention, as shown in the following formula (11), the probability distribution of the elements of the principal component matrix A^ is estimated by Bayesian estimation. Then, by using the estimated probability distribution to diagnose the condition of the structure, it is possible to detect abnormalities in the structure while taking into account the uncertainty included in the above formula (5).

In the present invention, since a probability distribution with a mean value and a variance is used as an evaluation index for evaluating the condition of a structure, it is possible to capture minute changes in the average value of the evaluation index when the structure is damaged compared to when the structure is healthy. Instead, the certainty of the evaluation index can be shown by the variance.

Note that the condition of a structure can also be diagnosed using the coefficient matrix A as the feature instead of the principal component matrix A^. In that case, the above formula (11) becomes a formula in which A^ is replaced with A.

<Embodiment 1>
Embodiment 1 will be described below with reference to FIGS. 1 to 7.

(Configuration of diagnostic system S of Embodiment 1)
FIG. 1 is a diagram showing a schematic configuration of a diagnostic system S according to the first embodiment. The diagnostic system S includes a sensor node 1 and a diagnostic device 2. The sensor node 1 generates a feature amount indicating the state of the bridge 4 based on sensor information acquired from the sensor 3 installed on the bridge 4 via wireless communication (or wired communication). The diagnostic device 2 acquires the feature amount indicating the state of the bridge 4 generated by the sensor node 1 via wireless communication (or wired communication), and diagnoses whether there is an abnormality in the bridge 4 using this feature amount.

Sensor 3 is, for example, an acceleration sensor, and is installed at each location of bridge 4. In this embodiment, sensor 3 is a plurality of sensors installed at multiple locations of bridge 4, but is not limited to this and may be a single sensor installed at one location. Also, in this embodiment, sensor 3 is an acceleration sensor that detects acceleration, but is not limited to this and may be a sensor that can measure other physical quantities (e.g., strain, displacement, speed, etc.).

In this embodiment, for ease of understanding, as shown in FIG. 1, a configuration is described as an example in which the sensor node 1 processes sensor information acquired from the sensor 3 and transmits the results to the diagnostic device 2. However, the present invention is not limited to this, and may be configured as a distributed cooperative system in which the sensors 3 form a sensor network and perform distributed cooperative processing to achieve functions equivalent to those of the sensor node 1. When there is only one sensor 3, the single sensor 3 achieves functions equivalent to those of the sensor node 1.

(Configuration of the sensor node 1 of the first embodiment)
2 is a block diagram showing the configuration of the sensor node 1 of the embodiment 1. The sensor node 1 performs edge processing to generate a coefficient matrix A (or a principal component matrix A^) as a feature quantity representing the state of the bridge 4 from sensor information acquired from the sensor 3. The sensor node 1 has a processor 11, a memory 12, a storage 13, and a communication I/F (Inter/Face) unit 14.

The processor 11 is an arithmetic processing device such as a CPU (Central Processing Unit), a PLD (Programmable Logic Device), or a microprocessor. The memory 12 is a main storage device. The storage 13 is an auxiliary storage device. The communication I/F unit 14 is a communication interface that enables the sensor node 1 to perform wireless communication (or wired communication) with the sensor 3 and the diagnostic device 2.

The processor 11 executes a program in cooperation with the memory 12 to realize the following functional units: a sensor information acquisition unit 111, an autoregressive model generation unit 112, a feature generation unit 113, and a feature transmission unit 114.

The sensor information acquisition unit 111 acquires observation signals observed by the sensors 3 at each installation position of the bridge 4 in time series via the communication I/F unit 14, and stores them in the storage 13 as sensor information 131. In this embodiment, the sensor information 131 acquired by the sensor information acquisition unit 111 is assumed to be time series information of continuous time, but may be time series information of discrete time. Further, the sensor information 131 may be acquired all the time, or may be acquired for a certain period of time in response to an acquisition instruction.

The autoregressive model generation unit 112 discretizes the sensor information 131 stored in the storage 13 by time. Then, the autoregressive model generation unit 112 converts the sensor information 131 at a certain time k to p times ki (i=1,... , p) is generated. Details of the processing by the autoregressive model generation unit 112 will be described later with reference to a flowchart.

The feature generator 113 generates a feature representing the state of the bridge 4 at a certain time k from the coefficient matrix A (formula (6) above) that combines the regression coefficients of the linear combination of the autoregressive models generated by the autoregressive model generator 112. The feature may be the coefficient matrix A itself, or a principal component matrix A^ (formula (10) above) that projects the coefficient matrix A onto a principal component space based on principal component analysis. As described below, the surface and internal state of the bridge 4 can be represented as a probability distribution based on such feature. Details of the processing by the feature generator 113 will be described later with reference to a flowchart.

The feature transmission unit 114 transmits the features generated by the feature generation unit 113 to the diagnostic device 2 via the communication I/F unit 14.

(Configuration of diagnostic device 2 of Embodiment 1)
FIG. 3 is a block diagram showing the configuration of the diagnostic device 2 of the first embodiment. The diagnostic device 2 includes a processor 21 , a memory 22 , a storage 23 , a communication I/F section 24 , and an output section 25 . The processor 21 executes programs in cooperation with the memory 22 to realize the feature amount receiving section 211 and the diagnostic section 212.

The processor 21 is an arithmetic processing device such as a CPU, PLD, or microprocessor. Memory 22 is a main storage device. Storage 23 is an auxiliary storage device. The communication I/F section 24 is a communication interface for the diagnostic device 2 to perform wireless communication (or wired communication) with the sensor node 1. The output unit 25 is a monitor, display, etc., and outputs various information.

The feature receiving unit 211 receives the feature 231 indicating the state of the bridge 4 at each time k from the sensor node 1 via the communication I/F unit 24 . The feature amount receiving unit 211 stores the received feature amount 231 in the storage 23.

The diagnosis unit 212 uses Bayesian estimation to obtain a probability distribution (the above formula (11)) of the feature at each time k from the feature 231 stored in the storage 23 as an evaluation index. The diagnosis unit 212 then calculates the Mahalanobis distance MD of the probability distribution obtained as the evaluation index at each time k, as shown in the following formula (12).

In the above equation (12), matrix X is a coefficient matrix A (or principal component matrix A^), and matrix S is a covariance matrix of matrix X. The diagnosis unit 212 determines in advance the Mahalanobis distance MD of the coefficient matrix A (or principal component matrix A^) of the bridge 4 at a reference time (for example, when it is healthy) as a reference evaluation index 232 and stores it in the storage 23 . The diagnosis unit 212 also calculates the Mahalanobis distance MD of the coefficient matrix A (or principal component matrix A^) of the bridge 4 at the time of diagnosis when the presence or absence of damage and the degree of damage are unknown.

Then, the diagnosis unit 212 compares the Mahalanobis distance MD at the time of diagnosis with the standard evaluation index 232, and if the statistical distance such as the difference or ratio between them is greater than a certain value, the diagnosis unit 212 compares it with the standard time and causes damage to the bridge 4. Diagnose that an abnormality has occurred or is progressing. Diagnosis section 212 outputs the diagnosis result to output section 25 such as a display.

The diagnosis unit 212 uses the Z value and other statistical indicators instead of the Mahalanobis distance MD to determine the threshold value of the statistical distance between the statistical indicators at the reference time and the diagnosis time, thereby detecting abnormalities in the bridge 4. It is also possible to diagnose the progress of deterioration.

(Reference time processing in diagnostic system S of Embodiment 1)
FIG. 4 is a sequence diagram showing the reference time processing in the diagnostic system S of the first embodiment. In the reference time processing in the diagnostic system S, a coefficient matrix A and a principal component matrix A^ are generated from the time series of sensor information acquired from the sensor 3 at a reference time (for example, when the bridge 4 is healthy), and the principal component matrix A^ is Based on this, a process is performed to calculate the standard evaluation index 232.

First, in step S101, the sensor information acquisition unit 111 of the sensor node 1 acquires the sensor information 131 from the sensor 3 and stores it in the storage 13. Next, in step S102, the autoregressive model generation unit 112 generates a p-th order autoregressive model of the sensor information vector y(k) at a certain time k, which is obtained by discretizing the sensor information 131 acquired in step S101, based on the above formula (3). The order p of the autoregressive model is determined appropriately, but may be automatically determined from the above formula (5) based on the BIC, for example, as described above.

Next, in step S103, the feature generation unit 113 generates a coefficient matrix A that is a combination of the matrices A _i representing the regression coefficients of the p-th autoregressive model of the sensor information vector y(k), as shown in equation (6) above. generate. Next, in step S104, the feature generation unit 113 projects the coefficient matrix A in the probability space representing the state of the bridge 4 at the reference time onto the principal component space based on the above equations (8) to (9). Generate a component space matrix U ₁ ^T.

Next, in step S106, the feature quantity generation unit 113 operates the principal component space matrix U ₁ ^T generated in step S105 on the coefficient matrix A, as shown in the above equation (10), and generates a principal component matrix as a feature quantity. Generate A^. Next, in step S106, the feature quantity transmitting unit 114 transmits the feature quantity generated in step S105 to the diagnostic device 2.

Next, in step S201, the feature receiving unit 211 of the diagnostic device 2 receives the feature 231 from the sensor node 1 and stores it in the storage 23. Next, in step S202, the diagnostic unit 212 obtains the probability distribution p(A^|Σ, Y) of the feature 231 as in the above formula (11), and calculates the Mahalanobis distance MD of the probability distribution p(A^|Σ, Y) as the reference evaluation index 232. Next, in step S203, the diagnostic unit 212 stores the reference evaluation index 232 calculated in step S202 in the storage 23.

Note that the process is not limited to the illustration in FIG. 4, and in the reference time processing, the diagnostic device 2 may execute the steps that the sensor node 1 executes, or the sensor node 1 may execute the steps that the diagnostic device 2 executes. . That is, it is not limited whether the sensor node 1 or the diagnostic device 2 executes each step shown in FIG.

For example, the processing after any one of steps S102 to S105 may be performed by the diagnostic device 2 instead of the sensor node 1. Furthermore, for example, the process in step S202 may be performed by the sensor node 1 instead of the diagnostic device 2, and the calculated standard evaluation index is transmitted to the diagnostic device 2 and used in the diagnostic device 2 to diagnose the condition of the bridge 4. .

(Diagnosis process in the diagnosis system S of the first embodiment)
Fig. 5 is a sequence diagram showing a diagnosis process in the diagnosis system S of the embodiment 1. The diagnosis process in Fig. 5 differs from the reference time process in Fig. 4 only in that it handles sensor information and feature amounts at the time of diagnosis rather than at the reference time, and steps S111 to S116 are the same as steps S101 to S106 in Fig. 4.

Note that if the principal component space matrix U ₁ ^T is generated in step S104 of the reference time process in FIG. 4, the step of generating the principal component space matrix U ₁ ^T in step S114 in the diagnosis time process of FIG. Optional. If step S114 is omitted, in step S115 following step S113, the feature generation unit 113 converts the principal component space matrix U ₁ ^T generated in step S104 of the reference time process in FIG. By acting on the coefficient matrix A, a principal component matrix A^ is generated as a feature quantity. Next, in step S116, the feature quantity transmitting unit 114 transmits the feature quantity generated in step S115 to the diagnostic device 2.

Next, in step S211, the feature receiving unit 211 of the diagnostic device 2 receives the feature from the sensor node 1. Next, in step S212, the diagnosis unit 212 calculates the probability distribution p(A^|Σ,Y) of the feature using Bayesian estimation, as shown in equation (11) above, based on the feature received in step S211. Calculate. Then, the diagnosis unit 212 calculates the Mahalanobis distance MD at the time of diagnosis as an evaluation index based on the above formula (12), compares the evaluation index with the reference evaluation index 232, and if the difference or ratio between them is greater than a certain value, , it is diagnosed that an abnormality such as damage has occurred or is progressing in the bridge 4 compared to the reference time.

As with FIG. 4, FIG. 5 does not limit whether each step is executed by the sensor node 1 or the diagnostic device 2. For example, the evaluation index may be calculated by the sensor node 1 instead of the diagnostic device 2. The evaluation index calculated in this manner is transmitted to the diagnostic device 2, where it is used to diagnose the condition of the bridge.

In the process shown in FIG. 4 and FIG. 5, the principal component matrix A^ is used as the feature, but the coefficient matrix A may be used as the feature. In this case, A^ is replaced with A in the above formula (11) to calculate the probability distribution p(A|Σ, Y).

Figure 6 is an explanatory diagram of the diagnosis processing in the diagnosis system S of embodiment 1. The illustrated portion 601 in Figure 6 shows an overview of the reference time processing (Figure 4) in which feature amounts are extracted from observation data representing the vibration of the bridge 4 at the reference time, and a statistical index based on these feature amounts is calculated as a reference evaluation index. The illustrated portion 602 shows an overview of the diagnosis time processing (Figure 5) in which feature amounts are extracted from observation data representing the vibration of the bridge 4 at the diagnosis time, and a statistical index based on these feature amounts is calculated as a diagnosis time evaluation index.

Then, in the diagnosis process (Fig. 5), the presence or absence of an abnormality in the bridge 4 and its progression are determined based on the comparison result between the reference evaluation index and the diagnosis evaluation index. That is, as shown in the illustrated portion 603 of Fig. 6, in a probability space, an abnormality in the bridge 4 is diagnosed by comparing a reference evaluation index based on the probability distribution of the reference feature amount with a diagnosis evaluation index based on the probability distribution of the diagnosis feature amount. In comparing the reference evaluation index with the diagnosis evaluation index, a statistical index such as the Mahalanobis distance is used to evaluate how far the diagnosis evaluation index deviates from the reference evaluation index. If the statistical distance between the diagnosis evaluation index and the reference evaluation index is equal to or greater than a certain value, the presence or absence of an abnormality in the bridge 4 can be determined so that it is determined that the state of the bridge 4 at the time of diagnosis has changed from the reference time.

In the present embodiment, the presence or absence of an abnormality in the bridge 4 is determined based on the statistical distance between the evaluation indicators at the reference time and the diagnosis time, but the present invention is not limited to this. For example, when coefficient matrix A (or principal component matrix A^) at each time is clustered, and coefficient matrix A based on new observed data is not classified into an existing cluster but is classified into a new cluster, the feature amount of coefficient matrix A It may be determined that an abnormality has occurred in the bridge 4 based on a change in the pattern.

(Effects of the First Embodiment)
In the above-mentioned first embodiment, the condition evaluation based on the feature quantity representing the state of the structure is performed using an evaluation index based on a probability distribution that includes various physical quantities of the structure and that can take into account the estimation error of the measurement value while absorbing the difference in sensitivity of the change in the measurement value to damage that differs depending on the measurement location. Therefore, according to the first embodiment, it is not necessary to set a vibration mode that is sensitive to damage, which requires a high level of specialized knowledge, and the state of the structure can be evaluated at low cost and with high accuracy without performing numerical analysis by using a probability distribution having a mean and variance based on the feature quantity that is sensitive to damage of the structure. In addition, the validity of the evaluation can be expressed by the variance.

In addition, the probability distribution evaluation index used in embodiment 1 captures minute changes that progress slowly as the structure is damaged and that are easily obscured by noise during measurement and forced vibrations such as live loads, making it possible to perform highly accurate condition diagnosis of the structure.

In addition, in the first embodiment, the coefficient matrix A representing the state of the structure is projected into the principal component space to generate a principal component matrix A^ of lower order in order to remove information of lower quality. Therefore, the feature quantities representing the state of the structure are compressed into feature quantities that can reproduce the state of the structure without reducing the quality of the information, so that the feature quantities can be transferred by edge processing using inexpensive advanced technology such as low power wide area radio, and the sensors and transfer system can be constructed inexpensively, which is expected to reduce management costs significantly.

More specifically, in the past, when evaluating such vibration conditions, it was necessary to use acceleration data for several minutes, which required a large amount of communication. However, in this method, the only information required is the value of the principal component matrix A^, so the principal component matrix A^ is calculated by edge processing in the measuring device, and by making the data small, it becomes possible to use inexpensive advanced technology such as specific low-power radio, and it is considered to have an advantage in terms of management costs as shown in Figure 7. Figure 7 is a diagram showing a comparison of the data amount of the coefficient matrix A and the principal component matrix A^ in the first embodiment and the conventional technology for each number of sensors. Figure 7 shows a case where data measured for 60 seconds at a period of 5 ms is transmitted. It can be seen from Figure 7 that the amount of data transferred from the sensor node 1 can be significantly reduced by the principal component matrix A^ in any number of sensors, whether 8, 4, or 1.

<Embodiment 2>
In the first embodiment, abnormality diagnosis of the bridge 4 is performed using a statistical index such as the Mahalanobis distance of a probability distribution based on the feature amounts of the bridge 4 at the reference time and the diagnosis time, whereas in the second embodiment, abnormality diagnosis of the bridge 4 is performed using, as an evaluation index, a Bayes factor based on the feature amounts of the bridge 4. Note that the configuration of the diagnosis system S in the second embodiment is similar to the configuration of the diagnosis system in the first embodiment, except that a part of the processing of the diagnosis unit 212 of the diagnosis device 2 is different.

(Diagnostic processing in diagnostic system S of Embodiment 2)
FIG. 8 is a flowchart showing diagnostic processing in the diagnostic system S of the second embodiment. The flowchart shown in FIG. 8 is another example of the process in step S212 of the diagnosis process in the first embodiment shown in FIG. That is, in the second embodiment, the diagnostic device 2 executes the diagnostic process shown in FIG. 8 following step S211. Note that in the second embodiment, the diagnostic device 2 can execute the diagnosis time process in FIG. 5 without performing the reference time process in FIG. 4 .

First, in step S2121, the diagnosis unit 212 of the diagnosis device 2 calculates a Bayes factor B defined by the following formula (13) as an evaluation index based on the coefficient matrix A (or the principal component matrix A^), which is the feature amount received in step S211. When distinguishing it from a local Bayes factor ^Bj described later, the Bayes factor B is called a global Bayes factor.

In the above equation (13), H ₀ is the null hypothesis (in this embodiment, the healthy state of the bridge 4), H ₁ is the alternative hypothesis (in this embodiment, the abnormal/damaged state of the bridge 4), and Y _t is the hypothesis test. The target feature quantity (in this embodiment, coefficient matrix A or principal component matrix A^), Φ _t is a parameter under the hypothesis of H ₀ or H ₁ (for example, the mean value or standard deviation of the probability distribution of the feature quantity). ).

In the Bayes factor B shown in Equation (13) above, the denominator p (Y _t |Φ _t , H ₀ ) is the same feature as the feature Y _t at the time of diagnosis is the state (healthy state) of the bridge 4 at the reference time. The numerator p(Y _t |Φ _t , H ₁ ) indicates the marginal likelihood in which the feature quantity Y _t at the time of diagnosis is a feature quantity different from the state of the bridge 4 at the reference time. Therefore, when the Bayes factor B, which is the ratio of p(Y _t |Φ _t , H ₁ ) to p(Y _t |Φ _t , H ₀ ), exceeds the threshold, an abnormality or damage occurs to the bridge 4 based on a certain feature amount. It can be determined that this is occurring.

Next, in step S2122, the diagnostic unit 212 determines whether the Bayes factor B calculated in step S2121 is greater than a threshold value. This threshold value is set based on past inspection data and the like for categories that the inspection engineer of the bridge 4 has determined to be abnormal. The diagnosis unit 212 determines that there is an abnormality in the bridge 4 when the Bayes factor B is greater than the threshold (Step S2122 Yes) (Step S2123), and determines that the bridge 4 has an abnormality when the Bayes factor B is less than or equal to the threshold (Step S2122 No). is determined to be normal (step S2124).

Note that the diagnostic processing shown in FIG. 8 may be performed by the sensor node 1 instead of the diagnostic device 2.

(Method of calculating threshold for abnormality determination)
Here, a method for calculating the abnormality determination threshold used in step S2122 of the diagnostic process (FIG. 8) will be described. FIG. 9A is a diagram for explaining a method of calculating a threshold value for abnormality determination according to the second embodiment. In the graph (a) on the left side of FIG. 9A, the horizontal axis is time, the vertical axis is the logarithm of Bayes factor B, and the value of Bayes factor B at each time is plotted. In the graph (b) on the right side of FIG. 9A, the horizontal axis is the frequency, the vertical axis is the logarithm of the Bayes factor B, and the frequency distribution of each value of the Bayes factor B is expressed. The calculation of the threshold value for abnormality determination may be performed by the diagnostic device 2, or may be performed by another information processing device.

On an actual bridge, various situations occur that affect anomaly detection, such as the passing of a truck carrying a heavy load or braking on a bridge. Bayes Factor B is a value with variation that includes specific data under various situations. When calculating the threshold value for Bayes Factor B, by including specific data from multiple measurement data over a specified period of time in the past as the basis for calculating the threshold value rather than eliminating such data, it is possible to calculate a threshold value that takes into account the various situations that occur on an actual bridge.

Among the past measurement data of Bridge 4, the measurement data for a predetermined standard period (for example, one year) (see graph (a) in FIG. 9A) is applied to the above equation (13) to calculate the distribution of Bayes factor B values. (See graph (b) in FIG. 9A). Based on the average value μ and variance σ of the distribution of Bayes factor B values, any value such as μ, μ+σ, μ+2σ, etc. can be determined as the threshold value, for example. In the graph (b) of FIG. 9A, since the value of Bayes factor B is normalized, it has a normal distribution with an average μ=0, and shows the case where the threshold value is the average μ. The threshold value of the Bayes factor B can be determined quantitatively using the average μ and the variance σ based on the characteristics of the bridge 4 and index values such as usage conditions, geographical conditions, and weather conditions. In this way, the accuracy of abnormality determination can be improved by using threshold values that take into account various situations that occur in actual bridges.

(Other examples of abnormality diagnosis thresholds)
Another example of the threshold value for abnormality diagnosis will be explained. FIG. 9B is a diagram for explaining another example of the abnormality determination method using a plurality of threshold values according to the second embodiment. In (graph a) of FIG. 9B, the horizontal axis is the frequency distribution, the vertical axis is the logarithm of the Bayes factor B, and the frequency distribution of each value of the Bayes factor B is expressed. In (graph b) of FIG. 9B, the horizontal axis is the day, the vertical axis is the logarithm of the Bayes factor B, and each value of the Bayes factor B for each day is plotted.

The value of Bayes factor B is determined to be abnormal when it exceeds a threshold value. For this reason, in the distribution of values of Bayes factor B based on the measurement data described above with reference to FIG. 9A, for values greater than the mean μ, multiple threshold values can be defined using the mean μ and standard deviation σ, and abnormality can be determined in stages. Based on multiple threshold values, ranges representing diagnostic levels, such as a "normal" range, a "caution required" range, and an "abnormal" range, can be defined.

For example, multiple values based on the mean μ and variance σ of the distribution of the Bayes Factor B values (such as μ, μ+σ, and μ+2σ in (graph a) of FIG. 9B) are determined as multiple stepped thresholds for abnormality diagnosis of the bridge 4 based on the Bayes Factor B. As shown in (graph a) of FIG. 14, if the Bayes Factor B is B≦μ+σ, it is diagnosed as "normal", if μ+σ<B≦μ+2σ, it is diagnosed as "needs attention", and if μ+2σ<B, it is diagnosed as "abnormal". In the example of FIG. 9B, the number of stepped thresholds for abnormality diagnosis is "2", but is not limited to this. The stepped thresholds and the number of thresholds for the Bayes Factor B can be quantitatively determined using the mean μ and variance σ based on index values such as the characteristics of the bridge 4, the conditions of use, the geographical conditions, and the weather conditions.

In this way, the accuracy of abnormality determination can be improved by using multiple thresholds that take into account various situations that occur on actual bridges. Furthermore, by gradually determining the threshold for abnormality diagnosis of bridge 4 based on Bayes factor B, the manager of bridge 4 can gradually detect abnormalities in bridge 4 that progress gradually over time, and can perform maintenance of bridge 4 in a planned manner.

(Local Bayes Factor B)
Now, the Bayes factor B based on the above formula (13) is an evaluation index for detecting an abnormality by looking at the bridge 4 as a whole. On the other hand, it is also possible to calculate individual Bayes factors based on feature amounts obtained from time series of sensor information of individual sensors 3 installed on the bridge 4. The individual Bayes factors are called local Bayes factors. For example, the local Bayes factor B ^j based on the time-series feature amount of the sensor information of the j-th (j=1,...,n) sensor 3 among the plurality of sensors 3 installed on the bridge 4 is as follows. It is defined by equation (14).

By replacing the local Bayes factor B ^j with Bayes factor B and performing the diagnostic processing shown in FIG ^. or damage has occurred.

The local Bayes factor ^Bj can also be used as follows. The type of sensor information that can sensitively detect anomalies using the Bayes factor may differ depending on the structure such as the bridge 4 and the type of anomaly or damage that occurs in the structure. Therefore, sensors 3 of multiple sensor types are installed on the structure such as the bridge 4, and diagnosis is performed based on the local Bayes factor ^Bj based on the feature amount generated for each sensor type.

That is, sensors 3 that detect different physical quantities (different sensor types) are installed on the bridge 4, and a local Bayes factor ^Bj is calculated based on the time-series feature amount of the sensor information for each sensor type. When an abnormality is determined based on the local Bayes factor ^Bj of the jth sensor type, it can be identified from the index j of the local Bayes factor ^Bj that an abnormality or damage has occurred based on the sensor information of the sensor 3 of the jth sensor type. By using the local Bayes factor ^Bj in this way, it is possible to detect an abnormality using any one of multiple physical quantities, thereby improving the sensitivity of anomaly detection.

(Abnormality determination in bridge diagnosis system S of Embodiment 2)
FIG. 10 is a diagram for explaining the experimental results of abnormality determination performed using the diagnostic system S of the second embodiment. FIG. 10 shows the time course of the binary logarithm of ten types of local Bayes factors B ^j , A1 to A10, with the horizontal axis representing time and the vertical axis representing the binary logarithm axis of the local Bayes factor B ^j . In FIG. 10, A1 to A10 indicate the installation positions of the sensors 3 as an example. INT (initial), DMG1, and DMG2 in FIG. 10 represent each period.

In INT, since no damage is caused to the bridge 4, the local Bayes factor ^Bj based on the sensor information of the sensor 3 at any installation position is approximately 0 and does not exceed the threshold value. Conversely, if the local Bayes factor B ^j is a value like INT shown in FIG. 10, it can be estimated that the bridge 4 is in a healthy state with no abnormality or damage.

Following INT, in DMG1 where damage was done to the area near A6 of bridge 4, the value of the local Bayes factor based on the sensor information of sensor 3 of A6, especially among A1 to A10, exceeded the threshold compared to INT. It's getting bigger. Conversely, if the local Bayes factor B ^j is a value like DMG1 compared to INT, it can be estimated that damage has occurred in the area near A6 of the bridge 4.

Furthermore, in DMG2, which follows DMG1 and further inflicts damage on a portion of bridge 4 near A6, the local Bayes factor ^Bj based on the sensor information of sensor 3 at A6 is even larger than that of DMG1. Conversely, if the local Bayes factor ^Bj has a value similar to that of DMG2 compared to DMG1, it can be estimated that the damage that occurred in the portion of bridge 4 near A6 has progressed further. Also, by comparing the sum or average of the local Bayes factors ^Bj of DMG1 with the sum or average of the local Bayes factors ^Bj of DMG2, the degree of progression of damage from DMG1 to DMG2 can be evaluated.

Note that the same evaluation as in FIG. 10 can be performed using the global Bayes factor.

By using Bayes factor B or local Bayes factor B ^j in this way, it is possible to detect that the bridge 4 has changed from a healthy state to an abnormal state, and also to evaluate the progress of damage that has already been found through inspection etc. can.

(Experimental Results of the Second Embodiment)
Fig. 11 is a diagram showing the execution conditions of an experiment performed using the diagnostic system S of the embodiment 2. Fig. 12 is a diagram showing the results of an experiment performed using the diagnostic system S of the embodiment 2.

FIG. 11 shows a pattern of loads applied to a bridge over time, with the horizontal axis representing time and the vertical axis representing load. As shown in FIG. 11, at times t0 to t1 (Stage 1), t2 to t3 (Stage 2), t4 to t5 (Stage 3), t6 to t7 (Stage 4), t7 to t8 (Stage 5), t9 to t10 (Stage 6), From t10 to t11 (Stage 7) and from t12 to (Stage 8), vibrations were applied to the bridge. From time t1 to t2 (Loading 1), a cracking load was applied to the bridge. From time t3 to t4 (Loading 2), a yield load was applied to the bridge. Time t5 to t6 (Loading3) is a period during which the load on the bridge continues. From time t8 to t9 (Loading 4), a design load was applied to the bridge. From time t11 to t12 (Loading 5), the maximum load was applied to the bridge.

As a result of conducting an experiment under such execution conditions, as shown in FIG. 12, it can be said that the Bayes factor generally tends to increase as the stage with loading progresses. Therefore, it is possible to give a clear interpretation to the observation results of changes in the Bayes factor, making it easy to detect abnormalities and damage occurring in bridges.

The advantage of this embodiment, that abnormalities or damage occurring in a bridge can be easily detected regardless of an increase or decrease in the bridge's natural frequency, can be seen from Figure 13. Figure 13 shows, as a comparative example, the change in the natural frequency of a bridge that differs depending on the vibration mode measured under the same conditions as in the experiment conducted using the diagnosis system S of embodiment 2 shown in Figure 12.

Graph 1101 in Figure 13 shows the change in the natural frequency of the primary bending mode in Stage 1 to Stage 8. Graph 1102 in Figure 13 shows the change in the natural frequency of the secondary bending mode in Stage 1 to Stage 8. As shown in graphs 1101 and 1102, the natural frequency increases or decreases depending on the change in the condition of the bridge. Theoretically, the natural frequency decreases due to a decrease in rigidity caused by damage, but the natural frequency may increase if, for example, there is a change in the condition of the bridge's support parts. Also, as shown in graphs 1101 and 1102, the change in the natural frequency differs depending on the mode.

As described above, abnormality diagnosis based on natural frequency is difficult because the manner in which the natural frequency changes varies depending on the location of the damage and the vibration mode, making it difficult to set an appropriate vibration mode and determine the presence or absence of damage and its location. In this regard, abnormality diagnosis using the Bayes factor according to this embodiment does not require the setting of a vibration mode, and can quantitatively detect abnormalities and damage that have occurred in the bridge.

(Change in temperature and Bayes factor value over time)
FIG. 14 is a diagram showing changes over time in the outside temperature and the Bayes factor value during a certain period of time with a certain girder of the bridge 4. (Graph a) in FIG. 14 is a graph showing the change over time in the outside temperature over a certain period, and (graph b) is a graph showing the change over time in the Bayes factor B of the outside temperature. The solid line in FIG. 14 (graph b) indicates the threshold for abnormality determination of Bayes factor B, and when this threshold is exceeded, it is determined to be abnormal.

Figure 14 shows an example of data on changes in outside air temperature of the girders of bridge 4, where it has been confirmed that there are no abnormalities in the structure itself. The outside air temperature fluctuates on an annual cycle, but when this embodiment is applied to the outside air temperature, the abnormality determination determines that the change in outside air temperature during the period for which abnormalities are to be determined is not abnormal. In other words, Figure 14 shows that this embodiment makes it possible to determine abnormalities in a structure by taking into account the presence or absence of the effects of periodic fluctuations that occur in nature, such as temperature changes.

(Changes in the deflection and Bayes factor values of Bridge 4 over time)
Figure 15 shows the change over time in the deflection and Bayes Factor values of a bridge over a certain period of time. Deflection is detected using a connecting pipe. In Figure 15, one year from April to March of the following year is taken as one fiscal year, and (graph a) shows the change over time in deflection over a certain period, while (graph b) shows the change over time in the Bayes Factor B of deflection. The solid line in (graph b) of Figure 15 indicates the threshold for determining an abnormality in Bayes Factor B, and if this threshold is exceeded, it is determined to be abnormal. The threshold for determining an abnormality is determined based on measurement data for one year in the base fiscal year.

Although the temperature fluctuates throughout the year, as shown in (graph a) in Figure 15, the determination results according to this embodiment indicate that an abnormality occurs during the period when the amount of deflection increases (for example, from September of the fourth year to January of the following year). You can see that it has been judged. Furthermore, it can be seen that from the 5th year onwards, it is clearly determined to be abnormal.

(Comparison of changes in Bayes factor values over time for two spans of a bridge)
Figure 16 is a diagram showing a comparison of the change over time in the Bayes factor values of two spans of a bridge. In Figure 16, the fiscal year is one year from May to April of the following year. Bridge 4 whose data is shown in Figure 16 is the same as Bridge 4 whose data is shown in Figure 15.

In (graph a) and (graph b) of Fig. 16, an abnormality is determined based on the acceleration measurement results at two different points of the same bridge 4 (the damaged first span and the second span with relatively little damage). The results are calculated. In FIG. 16, the threshold value is set based on one year's worth of measurement data, which is the base year. However, the base year, 1st year, 2nd year, and 3rd year in Figure 16 are different from those in Figure 15, so they are distinguished by notation, for example, 1st year in Figure 15 and 1st year in Figure 16. .

In the abnormality judgment according to this embodiment, an abnormality in the damaged first span was detected in September of the first year in Figure 16, and deterioration over time was able to be detected thereafter. On the other hand, in the second span, which was relatively less damaged, it was judged to be normal except for temporary abnormal values, even in the year in which an abnormality was detected in the first span in Figure 16. This result for the first span is the same as the period when the fluctuation in the amount of deflection became large as shown in Figure 15 (from the fourth and fifth years in Figure 15), so it is believed that appropriate abnormal values can be judged.

(Modification of Embodiment 2)
In this embodiment, the diagnostic device 2 uses the coefficient matrix A or the principal component matrix A^ of the bridge 4 as a feature amount representing the state of the bridge 4, and uses the Bayes factor B or the local Bayes factor ^Bj based on the feature amount to evaluate the bridge condition. The aim is to diagnose abnormalities. However, this is not limited to this, and features can be extracted from time-series data obtained by continuously measuring one or more physical quantities of each part of a structure, and a Bayes factor based on this feature can be used to diagnose abnormalities in a structure. It's okay. Physical quantities include displacement, velocity, acceleration, external force, strain, and temperature. For example, abnormality diagnosis of the structure may be performed using a Bayes factor based on a feature quantity extracted from data for a predetermined period of time series data obtained by observing the temperature of the bridge 4 at a fixed point.

The diagnosis unit may also correct the Bayes factor using the results of an on-site inspection of the bridge 4 by an inspector, and diagnose the condition of the bridge 4 based on the corrected Bayes factor. For example, if an abnormality is determined based on the Bayes factor of a certain feature, but no abnormality is found as a result of an on-site inspection, the Bayes factor is corrected so that the same feature is not subsequently determined to be abnormal. Alternatively, if an abnormality is found as a result of an on-site inspection, the threshold may be set or corrected by referring to the Bayes factor based on a sensor installed near the location where the abnormality was detected.

(Effects of Embodiment 2)
In Embodiment 2, the Bayes factor, which is the ratio of the probability of a healthy state to an abnormal state calculated from the time series of observation data of a structure in each time domain, and an inspection engineer with highly specialized knowledge The state of the structure is diagnosed based on the threshold value set based on the damaged state. Therefore, even those without highly specialized knowledge can use quantitative evaluation indicators to assess structural integrity in the same way as inspection engineers with highly specialized knowledge, without the need for high-cost processing such as numerical analysis. It is possible to evaluate the health of objects with high precision.

<Embodiment 3>
In Embodiment 3, in Embodiment 1 or 2, the coefficient matrix A, which is a feature amount representing the state of bridge 4, is sequentially learned using Bayesian estimation, thereby improving the accuracy of detecting abnormalities in bridge 4. The sequential learning of the coefficient matrix A may be performed by the sensor node 1 (for example, the feature value generation unit 113) or by the diagnostic device 2 (for example, the diagnosis unit 212). Note that the same effect can be obtained even if the same processing is performed on the principal component matrix A^ instead of the coefficient matrix A.

FIG. 17 is an explanatory diagram of sequential updating of the coefficient matrix A in the diagnostic system S of the third embodiment. In FIG. 17, the acceleration detected by the sensor 3 is shown on the vertical axis, and the time is shown on the horizontal axis. The posterior probability p(A,Σ|Y) _i of the coefficient matrix A at time i shown in FIG. 17 is obtained using Bayesian estimation as shown in the following formula (15-1). However, the prior probability p(A,Σ) _i at time i is the posterior probability p(A,Σ|Y) _i-1 at time (i-1) as shown in the following formula (15-2). By replacing i with i+1 in the following formulas (15-1) and (15-2), the posterior probability p(A,Σ|Y) _i+1 of the coefficient matrix A at time (i+1) can be obtained in the same manner.

(Effects of the Third Embodiment)
FIG. 18 is a diagram showing the convergence of the probability distribution of the coefficient matrix by sequentially updating the coefficient matrix in the diagnosis system of the third embodiment. In FIG. 18, the horizontal axis represents the values that the elements of the feature (for example, the coefficient matrix A) can take, and the vertical axis represents the posterior probability at each value of the element. In the third embodiment, by sequentially learning the probability distribution of the coefficient matrix A, as shown in FIG. 18, the average value of the probability distribution approaches the true value and converges in a direction in which the variance becomes smaller (the probability distribution shown in FIG. 18 converges from the thin line to the dashed line and then to the thick line), so that the coefficient matrix A represents the state of the structure with higher accuracy. Then, the diagnosis unit 212 diagnoses the state of the structure using the coefficient matrix A or the principal component matrix A^ after the sequential learning as a feature. Therefore, it is possible to improve the accuracy of the abnormality estimation of the state of the structure. Since the coefficient matrix A is a feature including various features of the vibration data of the structure, it is possible to capture a more minute change in the state of the structure and perform an abnormality determination than the conventional technology that performs an abnormality determination using the natural frequency or the like.

(Other embodiments)
In the embodiment described above, the diagnostic system S was described as having the sensor node 1 and the diagnostic device 2, but the embodiment is not limited to this. For example, the diagnostic system S may be configured without the diagnostic device 2. In such a case, the diagnostic system S includes a sensor node 1 having an acquisition section, an autoregressive model generation section, a feature amount generation section, and a diagnosis section. That is, the sensor node 1 includes an acquisition unit that acquires sensor information in time series from sensors attached to a structure, and a sensor information acquired at a certain time by the acquisition unit and a sensor information acquired before a certain time. An autoregressive model generator that generates an autoregressive model expressed as a linear combination of time series information, and a feature that generates a feature that indicates the state of a structure at a certain time based on the regression coefficient of the autoregressive model. a generation unit, and a first periphery representing a probability distribution of the feature amount indicative of the state of the structure generated from the time series of information regarding the structure, assuming that the structure is in a healthy state. An evaluation index, which is the ratio between the likelihood and a second marginal likelihood representing the probability distribution of the feature amount assuming that the structure is not in a healthy state, is calculated, and the state of the structure is determined based on the evaluation index. It has a diagnostic section that diagnoses.

In the embodiment described above, a case has been described in which a feature value indicating the state of a structure at a certain time is generated based on the regression coefficient of an autoregressive model, but the embodiment is not limited to this. For example, the feature quantity indicating the state of the structure may be generated from other than the regression coefficient of the autoregressive model.

(Example of sensor node 1 implementation)
19 is a block diagram showing the configuration of a sensor node terminal 1B used as the sensor node 1 of the embodiment. The sensor node terminal 1B implemented as the sensor node 1 further includes an acceleration sensor 15, in comparison with the configuration of the sensor node 1 (FIG. 2). Such a sensor node terminal 1B exerts the same functions as the sensor node 1 and the sensor 3 by executing a predetermined application. That is, the sensor node terminal 1B is installed on the bridge 4 in the same way as the sensor 3, generates feature quantities of the bridge 4 from the acquired acceleration data, and transmits them to the diagnostic device 2.

The diagnostic device 2 is constructed, for example, on a cloud server, and provides diagnostic results based on the features as a cloud service. The diagnostic device 2 transmits the diagnostic results to the sensor node terminal 1B or other terminal device. The user checks the diagnostic results by viewing the screen output of the sensor node terminal 1B or other terminal device.

In addition, multiple sensor node terminals 1B may be implemented as multiple sensor nodes 1 and installed at multiple locations on the bridge 4, similar to the sensors 3. In this case, a feature of the bridge 4 is generated from the acceleration data acquired by all of the multiple sensor node terminals 1B by independent processing of one sensor node terminal 1B representing the multiple sensor node terminals 1B or by cooperative processing of a predetermined number of sensor nodes 1, and is transmitted to the diagnostic device 2.

The present invention is not limited to the embodiments described above, and the configurations of each embodiment can be added, deleted, replaced, integrated, or distributed. Furthermore, the configurations and processes shown in the embodiments can be distributed, integrated, or replaced as appropriate based on processing or implementation efficiency. A program that executes each process of the diagnostic system described in the above embodiments is installed in one or more computers via a recording medium or a transmission medium, or is provided as an embedded program.

S: diagnostic system, 1: sensor node, 2: diagnostic device, 3: sensor, 4: bridge, 111: sensor information acquisition unit, 112: autoregressive model generation unit, 113: feature generation unit, 114: feature transmission unit, 131: sensor information, 211: feature reception unit, 212: diagnostic unit, 231: feature, 232: reference evaluation index

Claims (9)

A first marginal likelihood representing a probability distribution of the feature when the structure is assumed to be in a healthy state, with respect to a feature indicating the state of the structure generated from a time series of information regarding the structure. , and a second marginal likelihood representing the probability distribution of the feature amount when it is assumed that the structure is not in a healthy state, and based on the evaluation index, calculate the evaluation index of the structure. A diagnostic section that diagnoses the condition;
an acquisition unit that acquires sensor information in time series from a plurality of sensors attached to the structure;
an autoregressive model generation unit that generates an autoregressive model that expresses the sensor information acquired at a certain time by the acquisition unit as a linear combination of time series of the sensor information acquired before the certain time;
a feature amount generation unit that generates the feature amount indicating the state of the structure at the certain time based on the regression coefficient of the autoregressive model;
A sensor configured to include the acquisition section, the autoregressive model generation section, the feature amount generation section, and a transmission section that transmits the feature amount generated by the feature amount generation section to the diagnosis section. has a node;
The sensor information acquired from the plurality of sensors at each time in the time series is a sensor information vector at each time whose order is the number of the sensors,
The autoregressive model expresses the sensor information vector acquired by the acquisition unit at the certain time as a time-series linear combination of the sensor information vectors acquired before the certain time,
The regression coefficient is a matrix that is multiplied by each time series of the linearly combined sensor information vectors in the autoregressive model,
The structure diagnosis system, wherein the feature amount is information based on a coefficient matrix that is a combination of the matrices. The structure diagnosis system according to claim 1, wherein the feature amount generation unit generates, as the feature amount, a principal component matrix of the coefficient matrix based on singular value decomposition of the coefficient matrix. The feature value generation unit is
A claim characterized in that the feature quantity is generated by sequential learning that repeats a process in which a posterior distribution obtained by Bayesian estimation using a probability distribution of the previously generated feature quantity as a prior distribution is used as a probability distribution of the currently generated feature quantity. Item 2. Structure diagnosis system according to item 1 or 2. Claims 1 to 3, wherein the feature amount includes at least one of information on displacement, velocity, acceleration, mass, stiffness, and damping coefficient of the structure at an observation point of the feature amount. The structure diagnosis system according to any one of the above. The structure diagnosis system according to any one of claims 1 to 4, characterized in that the diagnosis unit corrects the evaluation index using inspection results from an on-site inspection of the structure, and diagnoses the condition of the structure based on the corrected evaluation index. The diagnosis unit diagnoses a state of the structure based on a comparison between the evaluation index and a threshold value;
The structure diagnosis system according to any one of claims 1 to 5, characterized in that the threshold value is determined based on the average and variance of multiple measurement data of the structure over a predetermined period of time in the past. A plurality of the threshold values are determined based on the average and the variance,
The structure diagnosis system according to claim 6, wherein the diagnosis unit diagnoses the state of the structure based on a comparison between the evaluation index and a range based on the plurality of threshold values. A structure diagnosis method executed by a structure diagnosis system, comprising:
The structure diagnosis system includes:
A sensor node including an acquisition unit, an autoregressive model generation unit, a feature generation unit, a transmission unit, and a diagnosis unit,
The acquisition unit acquires sensor information in time series from a plurality of sensors attached to a structure,
The autoregressive model generation unit generates an autoregressive model that expresses the sensor information acquired at a certain time by the acquisition unit as a linear combination of time series of the sensor information acquired before the certain time. death,
The feature amount generation unit generates a feature amount indicating the state of the structure at the certain time based on the regression coefficient of the autoregressive model,
The transmitting unit transmits the feature amount generated by the feature amount generating unit to the diagnostic unit,
The diagnosis unit calculates, for the feature amount, a first marginal likelihood representing a probability distribution of the feature amount when the structure is assumed to be in a healthy state, and a first marginal likelihood representing a probability distribution of the feature amount when it is assumed that the structure is not in the healthy state. and a second marginal likelihood representing the probability distribution of the feature amount in the case, and diagnosing the state of the structure based on the evaluation index,
The sensor information acquired from the plurality of sensors at each time in the time series is a sensor information vector at each time whose order is the number of the sensors,
The autoregressive model expresses the sensor information vector acquired at the certain time by the acquisition unit as a time-series linear combination of the sensor information vectors acquired before the certain time,
The regression coefficient is a matrix that is multiplied by each time series of the linearly combined sensor information vectors in the autoregressive model,
A structure diagnosis method, wherein the feature amount is information based on a coefficient matrix that is a combination of the matrices. Computer,
a diagnosis unit that calculates an evaluation index, which is a ratio between a first marginal likelihood representing a probability distribution of a feature quantity indicating a state of a structure, generated from a time series of information about the structure, when it is assumed that the structure is in a healthy state, and a second marginal likelihood representing a probability distribution of the feature quantity when it is assumed that the structure is not in the healthy state, and diagnoses the state of the structure based on the evaluation index;
an acquisition unit that acquires sensor information in time series from a plurality of sensors attached to the structure;
an autoregressive model generation unit that generates an autoregressive model that expresses the sensor information acquired by the acquisition unit at a certain time as a linear combination of a time series of the sensor information acquired before the certain time;
a feature generating unit that generates the feature indicating a state of the structure at the certain time based on a regression coefficient of the autoregressive model;
a sensor node including the acquisition unit, the autoregressive model generation unit, the feature generation unit, and a transmission unit that transmits the feature generated by the feature generation unit to the diagnosis unit,
The sensor information acquired from the plurality of sensors at each time point in the time series is a sensor information vector at each time point, the vector having an order equal to the number of the sensors;
the autoregressive model expresses the sensor information vector acquired by the acquisition unit at the certain time as a linear combination of a time series of the sensor information vector acquired before the certain time,
the regression coefficient is a matrix multiplied by each of the time series of the sensor information vectors linearly combined in the autoregressive model;
the feature amount is information based on a coefficient matrix obtained by combining the matrices.

Images