JP6421033B2 - Method, program and system for estimating damage state of structure - Google Patents

Description

  The present disclosure relates to a method, a program, and a system for estimating a damage state of a structure.

  Large structures such as bridges are gradually damaged from construction and deteriorate over the years. Therefore, it is necessary to repair damaged parts in a timely manner and to maintain and manage the structure.

  The current mainstream method for grasping the damaged part of the structure is a hammering test in which an operator hits the surface of the structure with a hammer and determines the damage state based on the sound. This method requires the skill of an inspection worker to make an appropriate judgment and manually inspects a large number of structures (for example, there are over 700,000 bridges in Japan when it comes to bridges). There is a limit.

  Therefore, Patent Document 1 discloses a system for diagnosing deterioration and / or soundness of a building. The system of Patent Document 1 obtains the natural frequency of the entire structure by using a plurality of wireless sensors arranged at predetermined positions of the structure and information measured by the sensor and transmitted from the sensor. And a determination unit that determines whether or not there is an abnormality in the entire structure by comparing the natural frequency obtained by the acquisition unit and the natural frequency of the structure that is acquired in advance in a healthy state. . Furthermore, in Patent Document 1, when it is determined that the structure is abnormal by the determination unit, a Bode diagram is obtained from the measurement result of the sensor, and it is determined that there is an abnormality with the board diagram of the structure at the time of soundness. Based on the comparison with the Bode diagram of the structure, a specifying means for specifying the crack location is provided.

JP 2008-255571 A

  In the system described in Patent Document 1, previous data is necessary to identify a damaged portion, and it is necessary to acquire data over time or periodically. However, in this case, it takes time to acquire data over time or periodically, and it takes time to accumulate data.

  Therefore, the present disclosure describes a method, a program, and a system that can estimate a damage state of a structure in a timely and simple manner.

  According to one aspect of the present disclosure, a method for estimating a damage state of a structure includes: an acceleration sensor that detects acceleration generated at a predetermined position of the structure when the vehicle travels on the structure where the acceleration sensor is disposed at the predetermined position. A first step of measuring, a second step of acquiring actual displacement data based on the acceleration data obtained from the acceleration sensor in the first step, and a vehicle model imitating a vehicle in a virtual space A virtual displacement that occurs at a virtual position corresponding to a predetermined location in the structure model is acquired as virtual displacement data by virtually running a structure model having a rigidity set to a predetermined value. And a fourth step of determining whether or not the actual displacement data obtained in the second step is approximate to the virtual displacement data obtained in the third step.

  In the method for estimating a damage state of a structure according to one aspect of the present disclosure, acceleration generated at a predetermined position of the structure is measured by an acceleration sensor, and actual displacement data is obtained based on the acceleration data obtained from the acceleration sensor. doing. When actual displacement data is acquired based on acceleration data measured by the acceleration sensor, the vibration state of the structure is detected more accurately than when actual displacement data is directly acquired using the displacement sensor. That is, the displacement (deflection) during traveling of the vehicle at a predetermined location of the structure can be obtained with high accuracy. Therefore, when the virtual displacement data approximates the actual displacement data as a result of the determination in the fourth step, it can be determined that the structure model accurately reproduces the actual structure. Accordingly, the rigidity of an actual structure that cannot be directly measured can be grasped from the rigidity of the structure model. As described above, according to the method for estimating the damage state of the structure according to one aspect of the present disclosure, the rigidity of the structure can be grasped only by the vehicle traveling on the structure without accumulating data over time. Therefore, it is possible to estimate the damage state of the structure in a timely and simple manner.

  If the result of determination in the fourth step is not an approximation, the method further includes a fifth step of changing the stiffness setting value in part or all of the structure model, and after the change in the fifth step The third and fourth steps may be performed again using the stiffness setting value. In this case, the change of the stiffness setting value in the structure model is repeated until the virtual displacement data approximates the actual displacement data. Therefore, the stiffness setting value in the structure model can be made very close to the rigidity of the actual structure. Therefore, it becomes possible to estimate the damage state of the structure more accurately.

  The structure is a bridge including a plurality of piers and an upper structure supported on the plurality of piers. In the fifth step, if the result of determination in the fourth step is not approximate, The stiffness setting value in the first portion near the pier in the superstructure of the structure model is set smaller than the stiffness setting value in the second portion different from the first portion in the superstructure of the structure model. May be. The first part in the vicinity of the pier in the superstructure of the actual structure is supported by the pier, and thus receives the load of the superstructure. In addition to this, when the bridge is a continuous girder bridge, a particularly large force is likely to be applied to the first portion when the vehicle passes through the bridge. Therefore, damage is likely to concentrate on the first part of the actual structure because a long period of time elapses after the construction or the vehicle repeatedly passes through the bridge. Therefore, by setting the stiffness setting value in the first portion of the structure model to be smaller than the stiffness setting value in the second portion of the structure model, the stiffness setting value in the structure model is reduced to the actual value. It can be made closer to the rigidity of the structure. As a result, it is possible to estimate the damage state of the actual structure with higher accuracy.

  The structure includes first to fourth piers, a first support beam supported on the first and second piers, and a second support beam supported on the third and fourth piers. A gel bar-structured bridge comprising a suspension girder positioned between the first and second girder and supported on the first and second girder via a gel bar hinge. The girder includes a first portion extending from the first pier close to the suspension girder to the end on the suspension girder side, a second portion extending between the first and second piers, and the first pier A third portion extending to reach the second portion, and the second receiving girder includes a fourth portion extending from the third bridge pier close to the suspension girder to the end on the suspension girder side, A fifth portion extending between the third and fourth piers and a sixth portion extending from the third pier to the fifth portion, in the fifth step If the result of determination in the fourth step is not an approximation, the stiffness setting values in the third and sixth parts of the structure model are set to the rigidity values in the second and fifth parts of the structure model. The stiffness setting value in the first and fourth portions of the structure model may be set smaller than the setting value, and the stiffness setting value in the third and sixth portions of the structure model may be set smaller than the setting value. Of the superstructure of the actual structure, the first, third, fourth and sixth portions in the vicinity of the pier are supported by the pier and thus receive the load of the superstructure. In addition, when the bridge is a continuous girder bridge, a particularly large force is likely to be applied to the first, third, fourth, and sixth portions when the vehicle passes through the bridge. Therefore, damage is likely to concentrate on the first, third, fourth, and sixth portions of the actual structure when a long period of time elapses after construction or when the vehicle repeatedly passes through the bridge. Therefore, by setting the stiffness setting values in the first, third, fourth and sixth portions of the structure model to be smaller than the stiffness setting values in the second and fifth portions of the structure model, The rigidity setting value in the structure model can be made closer to the rigidity of the actual structure. Further, the first and fourth portions of the superstructure of the actual structure receive the load of the hanging girder because they support the hanging girder via the gel bar hinge. In addition, the first and fourth parts of the superstructure of the actual structure are subject to wind and rain after a long period of time since the rain and the like tend to collect from the gap with the hanging girder. It is easy to be eroded by. Therefore, damage is particularly likely to concentrate on the first and fourth portions of the actual structure. Therefore, by setting the stiffness setting values in the first and fourth portions of the structure model to be smaller than the stiffness setting values in the third and sixth portions of the structure model, the rigidity of the structure model is set. The set value can be made closer to the rigidity of the actual structure. As a result, it is possible to estimate the damage state of the actual structure with higher accuracy.

  In the fourth step, it is determined that approximation is made when the extreme value of the virtual displacement data obtained in the third step is within ± 5% of the extreme value of the actual displacement data obtained in the second step. May be. In this case, the virtual displacement data is accurately approximated to the actual displacement data at the extreme value where the absolute value of the displacement tends to be large. Therefore, the rigidity setting value in the structure model can be brought closer to the actual structure rigidity with higher accuracy.

  The method further includes a sixth step of obtaining the natural frequency of the structure based on the acceleration data obtained in the first step. In the third step, the natural frequency of the structure model is also obtained. In the fourth step, It may also be determined whether or not the natural frequencies obtained in the third and sixth steps are approximate to each other. In this case, in addition to the approximation between the virtual displacement data and the actual displacement data, it is also determined whether or not the natural frequencies are approximated in the actual structure and the structure model. Therefore, even if the damage state of the structure is complicated, the damage state of the structure can be estimated with higher accuracy by comparing the actual structure and the structure model from a plurality of viewpoints.

  The acceleration sensor may be a servo type acceleration sensor. Servo-type acceleration sensors have excellent linearity and resolution, have a wide dynamic range, and extremely high acceleration detection accuracy. Therefore, according to the servo acceleration sensor, a response in a low frequency region (for example, 0.1 Hz or less (10 seconds or less)) can be detected with high accuracy. The displacement generated in the structure when the vehicle travels on the structure is generally as small as about 0.001 mm to 1 mm, and the dominant frequency of the displacement when the vehicle travels on the structure is v / 2L (where v is In general, a low-frequency component of about 0.2 Hz to 2 Hz is dominant because L is the vehicle traveling speed and L is the length of the target structure. Therefore, the vibration state of the structure can be detected with extremely high accuracy by using the servo acceleration sensor.

  If the result of determination in the fourth step is an approximation, the rigidity setting value in the structure model at that time is the rigidity value of the structure when it is assumed that the structure is damaged in a predetermined manner. The degree of damage to the structure may be estimated by comparing with the theoretical value. In this case, even when damage that cannot be confirmed by visually inspecting the appearance has occurred inside the structure, it is possible to estimate the degree of damage to the structure.

  A method for estimating a damage state of a structure according to another aspect of the present disclosure is a structure in which first to N-th (N is a natural number of 2 or more) acceleration sensors are arranged in first to N-th locations, respectively. A first step of measuring accelerations generated in the first to Nth locations of the structure when the vehicle travels with the first to Nth acceleration sensors, respectively, and first to first in the first step. A second step of acquiring first to N-th actual displacement data based on first to N-th acceleration data obtained from each of N acceleration sensors, and a vehicle model simulating a vehicle in virtual space By virtually running a structure model imitating a structure and having a rigidity set to a predetermined value, the first to Nth virtual corresponding respectively to the first to Nth locations in the structure model The virtual displacement generated at the position is used as the first to Nth virtual displacement data. The third step to be acquired, the k-th actual displacement data (k is a natural number of 1 to N) obtained in the second step, and the k-th virtual displacement data obtained in the third step. And a fourth step of determining whether or not the approximation is performed for all k.

  In the method for estimating the damage state of the structure according to another aspect of the present disclosure, it is possible to estimate the damage state of the structure in a timely and simple manner as in the above method.

  According to another aspect of the present disclosure, a program for estimating a damage state of a structure is measured by an acceleration sensor disposed at a predetermined position on a computer, when the vehicle travels on the structure. A first process for acquiring actual displacement data at a predetermined location on the basis of the acceleration data obtained by doing the above, and in the virtual space, imitating a vehicle model simulating a vehicle and having a predetermined rigidity By virtually running the structure model set to, a second process of acquiring virtual displacement data generated at a virtual position corresponding to a predetermined location in the structure model as virtual displacement data, and a first process A third process for determining whether or not the obtained actual displacement data and the virtual displacement data obtained by the second process are approximated is executed.

  In the program for estimating the damage state of the structure according to another aspect of the present disclosure, it is possible to estimate the damage state of the structure in a timely and simple manner as in the above method.

  A system for estimating a damage state of a structure according to another aspect of the present disclosure is arranged at a predetermined position of the structure, and measures an acceleration generated at the predetermined position of the structure when the vehicle travels on the structure And first acquisition means for acquiring actual displacement data based on acceleration data obtained from the acceleration sensor, and in the virtual space, imitating a structure similar to a vehicle model simulating a vehicle and having a predetermined rigidity A second acquisition unit that acquires a virtual displacement generated at a virtual position corresponding to a predetermined position in the structure model as virtual displacement data by virtually running the set structure model; and a first acquisition unit. Determining means for determining whether or not the actual displacement data acquired by the second acquisition means approximates the virtual displacement data acquired by the second acquisition means.

  In the system for estimating the damage state of the structure according to another aspect of the present disclosure, it is possible to estimate the damage state of the structure in a timely and simple manner as in the above method.

  According to the method, program, and system according to the present disclosure, it is possible to estimate a damage state of a structure in a timely and simple manner.

FIG. 1 shows an estimation system for estimating a damage state of a bridge according to the present embodiment. FIG. 2 schematically shows an internal configuration of an estimation system for estimating a bridge damage state according to the present embodiment. FIG. 3 is a flowchart for explaining a process for estimating a damage state of a bridge. FIG. 4 is a flowchart for explaining the rigidity estimation process. FIG. 5 is a flowchart for explaining damage degree estimation processing. FIG. 6 shows a state in which a vehicle travels on a bridge with a Gerber structure. FIG. 7 shows a cross section taken along line VII-VII in FIG. FIG. 8 shows experimental values of acceleration generated at the center position of the receiving beam when the bus travels on the bridge. FIG. 9 shows experimental values of the displacement that occurs at the center position of the receiving beam when the bus travels on the bridge. FIG. 10 shows a power spectrum obtained based on the acceleration shown in FIG. FIG. 11 shows the experimental values of FIG. 8 and the analysis value of the displacement that occurs at the center position of the receiving beam when the bus model runs on the bridge model that is set in a healthy state (the state where there is no reduction in rigidity). Show. FIG. 12 shows experimental values of FIG. 8 and analytical values of displacement generated at the center position of the receiving beam when the bus model runs on a bridge model in which the overall rigidity is set to be 30% lower than that in a healthy state. FIG. 13 shows the experimental values of FIG. 8 and the rigidity of the center quarter region of the span of the girder is set in a healthy state, and the rigidity of other regions of the girder is set to be 30% lower than that in the healthy state. The analysis value of the displacement generated at the center position of the receiving girder when the bus model runs through the bridge model is shown. FIG. 14 shows the experimental values of FIG. 8 and the rigidity of the center half area between the spans is set to be in a healthy state and the rigidity of other regions in the support girders is set to be 30% lower than that in the healthy state. The analysis value of the displacement generated at the center position of the receiving girder when the bus model runs through the bridge model is shown. FIG. 15 shows experimental values and analysis values of the natural frequency. FIG. 16 (a) shows a bridge when the rigidity of the center half area between the spans is set in a healthy state and the rigidity of other regions in the bridge girders is set to be 30% lower than that in the healthy state. The measurement position (near the gel bar hinge) in the model is shown, and FIG. 16 (b) shows the experimental value and analysis value of the displacement that occurs at the measurement position in FIG. 16 (a). FIG. 17 (a) shows the bridge when the rigidity of the center half area between the spans is set to a healthy state and the rigidity of the other regions of the bridge girders is set to be 30% lower than the healthy state. FIG. 17B shows experimental values and analysis values of displacement occurring at the measurement position in FIG. 17A. FIG. 18 (a) shows that the rigidity of the span center 1/2 region of the receiving beam is set in a healthy state, and the region of the receiving beam from the bridge pier to the span center 1/2 region is more healthy than the sound state. 18 shows the measurement position (the center part of the receiving beam) in the bridge model when it is set 30% lower and the rigidity of the region from the bridge pier to the end of the receiving beam is set 50% lower than the healthy state. (B) shows the experimental value and analysis value of the displacement which arises in the measurement position of Fig.18 (a). FIG. 19 (a) shows that the rigidity of the span center 1/2 region of the receiving beam is set to a healthy state, and the region of the receiving beam from the bridge pier to the span center 1/2 region is more healthy than the sound state. FIG. 19B shows the measurement position (in the vicinity of the Gerber hinge) in the bridge model when the rigidity of the region from the bridge pier to the end of the girder is set to be 50% lower than the healthy state. ) Shows experimental values and analysis values of displacement occurring at the measurement position in FIG. FIG. 20 (a) shows that the rigidity of the center half area between the spans of the girder is set to a healthy state, and the rigidity of the area from the bridge pier to the center half area of the span of the girder is higher than the sound state. FIG. 20 shows the measurement position in the bridge model (the center part of the suspension girder) when the rigidity of the area from the bridge pier to the end of the bridge girder is set to be 50% lower than the healthy state. b) shows an experimental value and an analysis value of the displacement occurring at the measurement position in FIG.

  Embodiments of the present invention will be described with reference to the drawings. However, the following embodiments are exemplifications for explaining the present invention and are not intended to limit the present invention to the following contents. In the description, the same reference numerals are used for the same elements or elements having the same function, and redundant description is omitted.

[Configuration of estimation system]
First, with reference to FIG.1 and FIG.2, the structure of the estimation system 100 for estimating the damage state of a bridge is demonstrated. The estimation system 100 includes a computer main body (estimation device) 102, a display 104 that displays an image in a display area 104a, and an acceleration sensor AM. The estimation system 100 may further include an operation input device such as a keyboard or a mouse that allows an operator to input an operation to the computer main body 102.

  As shown in FIG. 2, the computer main body 102 includes a CPU (Central Processing Unit) 102a, a RAM (Random Access Memory) 102b, a ROM (Read Only Memory) 102c, an I / O port 102d, and an external storage device (storage means). 102e. The CPU 102a realizes various functions in the computer main body 102 according to a program stored in the ROM 102c, and inputs / outputs signals to / from other components via the I / O port 102d, thereby operating the entire computer main body 102. Take control.

  The RAM 102b stores data and programs used when the CPU 102a operates. The ROM 102c stores a program executed by the CPU 102a (an estimation program for estimating a damage state of the bridge) and permanent data.

  The estimation program can be recorded and distributed on a computer-readable recording medium. In this specification, a computer-readable recording medium includes a non-transitory tangible medium (non-transitory computer recording medium) (for example, various main storage devices or auxiliary storage devices) and a propagation signal (transitory computer recording medium). (E.g., a data signal that can be provided over a network). Examples of the tangible medium that is not temporary include a magnetic medium (hard disk, flexible disk, etc.), an optical medium (CR-ROM, DVD-ROM, BD-ROM, etc.), a magneto-optical medium (MO disk, etc.), or a program instruction. And other hardware devices (RAM, ROM, semiconductor non-volatile memory, etc.) specially arranged to execute or store.

  In order to read or execute the estimation program from the recording medium, a recording medium reading drive (for example, a flexible disk drive) that reads the estimation program from the recording medium is provided via the I / O port 102d as necessary. It may be connected to the CPU 102a. The drive may be built in the computer main body 102 or may be separate from the computer main body 102. When the drive is separate from the computer main body 102, the drive is located outside the computer main body 102, and is connected to the I / O port 102d of the computer main body 102 via a signal line.

  The external storage device 102e is connected to the CPU 102a via the I / O port 102d. The external storage device 102e is used to hold necessary data such as bridge design data, vehicle data, bridge model data, vehicle model data, and acceleration data input from the acceleration sensor AM.

  The display 104 is connected to the CPU 102a via the I / O port 102d, and displays a predetermined image in the display area 104a based on an instruction from the CPU 102a. The display 104 may be integrated with the computer main body 102.

  The acceleration sensor AM is a sensor that detects the acceleration of the measurement object by being attached to the measurement object. As shown in FIG. 1, the acceleration sensor AM transmits the measured acceleration to the computer main body 102 as acceleration data. A method for transmitting the acceleration data from the acceleration sensor AM to the computer main body 102 may be wired communication or wireless communication. In this case, the acceleration data may be transmitted immediately after the measurement, or the acceleration data may be transmitted regularly or irregularly after a predetermined period has elapsed or after a predetermined amount of acceleration data has been accumulated. Alternatively, a recording medium capable of recording acceleration data measured by the acceleration sensor AM is built in the acceleration sensor, and when the acceleration sensor AM is directly or indirectly connected to the computer main body 102, the recording medium is recorded on the recording medium. Acceleration data may be transmitted to the computer main body 102.

In this embodiment, various types of acceleration sensors AM can be used, but it is particularly preferable to use a servo acceleration sensor as the acceleration sensor AM. The servo type acceleration sensor includes, for example, a movable part (pendulum), a detection part for detecting the position of the movable part, a drive part (electromagnetic coil) for driving the movable part, and a housing that accommodates these. With body. When acceleration is generated by applying an external force to the servo-type acceleration sensor, the movable unit is displaced from a predetermined position, but the drive unit drives the movable unit so as to cancel out the shift amount of the movable unit detected by the detector. For this reason, the movable portion is kept stationary in the housing. In the servo type acceleration sensor, the acceleration generated in the servo type acceleration sensor is acquired based on the force applied to the movable part by the driving unit at this time. Servo-type acceleration sensors have excellent linearity and resolution, have a wide dynamic range, and extremely high acceleration detection accuracy. Therefore, according to the servo acceleration sensor, a response in a low frequency region (for example, 0.1 Hz or less (10 seconds or less)) can be detected with high accuracy. When the vehicle V travels on the bridge 1 that is much larger than itself, the displacement generated in the bridge 1 is very small, and therefore the low frequency component is dominant in the acceleration generated in the bridge 1. Therefore, the vibration state of the bridge 1 can be detected with extremely high accuracy by using the servo acceleration sensor. The noise density of the acceleration sensor AM may be 500 ng _rms / √Hz or less, 400 ng _rms / √Hz or less, or 300 ng _rms / √Hz or less.

[Estimation process]
Next, a bridge damage state estimation process executed by the estimation program in the estimation system 100 will be described with reference to FIGS. As shown in FIG. 3, the damage state estimation process includes a rigidity estimation process (step S1) and a damage degree estimation process (step S2).

  Next, the rigidity estimation process will be described with reference to FIG. First, a vibration measurement experiment is performed (step S10). As shown in FIG. 6, the vibration measurement experiment is performed when the vehicle V travels on the actual bridge 1. The bridge 1 is a structure that extends over rivers, seas, valleys, other structures, etc., and enables traffic between the ends. The bridge 1 includes a plurality of piers 2 and an upper structure 3 supported by the piers 2. In the present embodiment, the bridge 1 has a gel bar structure. FIG. 6 shows a part of the bridge 1.

  The plurality of bridge piers 2 are arranged in a line along the direction in which the bridge 1 extends (the left-right direction in FIG. 6). FIG. 6 shows six of the plurality of piers 2. Hereinafter, the six piers 2 are referred to as piers 2A to 2F in order from the left side of FIG.

  The upper structure 3 includes a plurality of receiving girders 4, a plurality of hanging girders 5, and a support (not shown). The plurality of receiving girders 4 and the plurality of hanging girders 5 are arranged alternately along the direction in which the bridge 1 extends. Adjacent receiving girders 4 and hanging girders 5 are not integrated. As shown in FIG. 7, each of the receiving girders 4 and the hanging girders 5 is provided with floor plates 4 a and 5 a, main girders 4 b and 5 b that are provided on the lower surfaces of the floor plates 4 a and 5 a and extend in the extending direction of the bridge 1, There are rails 4c, 5c provided on the upper surfaces of the floor plates 4a, 5a and extending in the direction in which the bridge 1 extends. The receiving girder 4 and the hanging girder 5 constitute a passage through which vehicles and people can pass. In FIG. 6, three of the plurality of receiving girders 4 are shown, and two of the plurality of hanging girders 5 are shown. Hereinafter, the three receiving girders 4 will be referred to as receiving girders 4A to 4C in order from the left in FIG. 6, and the two hanging girders 5 will be referred to as hanging girders 5A and 5B in order from the left in FIG. To do.

The receiving girder 4A is supported on the piers 2A and 2B via bearings positioned between the receiving girder 4A and the piers 2A and 2B, respectively. Receiving digits. 4A, piers 2A, a portion 4A ₁ extending between 2B, piers 2A, a pair of portions 4A ₂ extending up to the end of 2B, portions _4A 1, 4A pair of portions 4A ₃ extending between the _two Including.

The receiving girder 4B is supported on the piers 2C and 2D via bearings positioned between the receiving girder 4B and the piers 2C and 2D, respectively. Receiving digits. 4B, piers 2C, a portion 4B ₁ extending between 2D, piers 2C, the pair of portions 4B ₂ extending up to the end from 2D, portions _4B 1, 4B pair of portions 4B ₃ extending between the _two Including.

The receiving girder 4C is supported on the piers 2E and 2F via bearings positioned between the receiving girder 4C and the piers 2E and 2F, respectively. Receiving digits. 4C, piers 2E, a portion 4C ₁ extending between 2F, piers 2E, a pair of portions 4C ₂ which extends up to the end of 2F, portions _4C 1, 4C pair of portions 4C ₃ extending between the _two Including.

  The suspension girders 5A are supported on the receiving girders 4A and 4B via supports (gel bar hinges) positioned between the hanging girders 5A and the receiving girders 4A and 4B, respectively. The hanging girder 5B is supported on the receiving beams 4B and 4C via supports (gel bar hinges) positioned between the hanging beam 5B and the receiving beams 4B and 4C, respectively.

  An acceleration sensor AM (hereinafter referred to as acceleration sensor AM1) is attached to the central portion of the receiving beam 4B. An acceleration sensor AM (hereinafter referred to as acceleration sensor AM2) is attached in the vicinity of a support (Gelbar hinge) positioned between the receiving beam 4B and the hanging beam 5B. An acceleration sensor AM (hereinafter referred to as acceleration sensor AM3) is attached to the central portion of the hanging beam 5B. The acceleration sensors AM1 to AM3 measure the acceleration generated at the attachment position when the vehicle V travels on the bridge 1 (on the upper structure 3). Although there is no restriction | limiting in particular in the attachment position of acceleration sensor AM1-AM3, when it considers the ease of installation, it is preferable if it fixes to the upper surface of the floor boards 4a and 5a.

  Examples of the vehicle V used in the vibration measurement experiment include a vehicle having a predetermined weight such as an automobile and a railway vehicle. The weight of the vehicle V may be appropriately set according to the size and structure of the bridge 1. The weight of the vehicle V is not particularly limited as long as the bridge 1 can generate vibrations that can be detected by the acceleration sensors AM (AM1 to AM3) when the vehicle V travels on the bridge 1. It is preferable that the weight of the vehicle V is larger because vibration is easily generated in the bridge 1 when the vehicle V travels on the bridge 1. However, since the bridge 1 may be damaged if the weight of the vehicle V becomes too large, it is preferable that the weight of the vehicle V does not exceed a size that the bridge 1 can support. The weight of the vehicle V may be, for example, about 10,000 kg to 40000 kg when the bridge 1 is a general road bridge. For example, when the bridge 1 is a general railway bridge, It may be about 30000 kg to 50000 kg.

  In the vibration measurement experiment, the vehicle V travels on the bridge 1 (on the upper structure 3) at a predetermined speed, and acceleration is detected by at least one acceleration sensor AM (any one or more of acceleration sensors AM1 to AM3). The vehicle V may travel all or some of the bridge 1 (upper structure 3). When the vehicle V travels a part of the bridge 1 (superstructure 3), when the bridge 1 is a simple girder bridge, at least one span (the distance between adjacent piers, also referred to as one span). The vehicle V only needs to travel, and when the bridge 1 is a continuous girder bridge, the vehicle V only needs to travel at least the girder to which the acceleration sensor AM is attached and the girder on both sides thereof.

  The traveling speed of the vehicle V may be substantially constant or may vary. When the traveling speed of the vehicle V is substantially constant, it becomes easier to set the speed of the vehicle model when the simulation is performed on the computer as compared with the case where the traveling speed is changed. To do). The traveling speed of the vehicle V is not particularly limited, but since the sampling frequency of the acceleration sensor AM (the number of acceleration data acquired by the acceleration sensor AM per unit time) has an upper limit, the number of samples is small when the traveling speed is too high. Therefore, the accuracy of detection of the acceleration generated in the bridge 1 tends to be low, and when the traveling speed is too slow, the bridge 1 hardly vibrates or vibrates at a very low frequency. Therefore, the acceleration sensor AM detects the vibration. It tends to be difficult. The traveling speed of the vehicle V may be, for example, about 20 km / h to 60 km / h when the vehicle V is an automobile, or about 80 km / h to 300 km / h when the vehicle V is a railway vehicle. It may be.

  Subsequently, acceleration data detected by the acceleration sensor AM is transmitted to the computer main body 102. Thereby, the computer main body 102 acquires acceleration data (step S11). At this time, the computer main body 102 (I / O port 102d and external storage device 102e) operates as an acceleration data acquisition unit.

  Subsequently, the computer main body 102 acquires actual displacement data from the acceleration data (step S12). At this time, the computer main body 102 (CPU 102a) operates as a means for acquiring actual displacement data. As a method for acquiring the actual displacement data from the acceleration data, for example, a method of performing second-order integration while performing baseline correction on the acceleration data in the time domain, or performing second-order integration using a bandpass filter in the frequency domain. A method is mentioned. For example, the band pass filter may cut a frequency of 0.1 Hz or less where noise components are dominant.

  Subsequently, the computer main body 102 calculates a natural frequency as an experimental value from the acceleration data (step S13). At this time, the computer main body 102 (CPU 102a) operates as means for acquiring the natural frequency as an experimental value. An example of a method for calculating the natural frequency from the acceleration data is a modified periodogram method (Welch method). In the modified periodogram method, noise is removed from measurement data including noise that occurs at random using a predetermined taper (for example, Hamming taper, Hanning taper). As a result, the power spectrum obtained from the acceleration data is smoothed, and the dominant frequency in the acceleration data can be easily identified.

  Subsequently, a vibration simulation is performed on the computer main body 102 (step S14). Specifically, a bridge model 1M (see FIG. 1) simulating an actual bridge 1 is configured in the virtual space of the computer main body 102. The bridge model 1M may be a model in which the bridge 1 is modeled by a solid element, or may be a model in which the bridge 1 is modeled by a beam element. At this time, the bridge model 1M includes a secondary moment of section I obtained based on the cross section of the bridge 1 (see FIG. 7), an elastic modulus E of concrete obtained based on the design criteria at the time of construction of the bridge 1, The bending rigidity EI is set. Examples of design criteria include concrete standard specifications, road bridge specifications, and the like. Thereby, a bridge model 1M simulating the bridge 1 at the time of construction (that is, the bridge 1 in a healthy state) is obtained.

  Subsequently, a vehicle model (not shown) simulating the actual vehicle V is configured in the virtual space of the computer main body 102. The vehicle model may be, for example, one mass point, or may be modeled using a spring element and a damper (dashpot) in consideration of the position of the center of gravity. The spring constant of the vehicle model may be obtained by conducting a predetermined experiment on the actual vehicle V (for example, Akito Yabe, “Basics on Development of an Existing Bridge Deformation Detection System Using a Route Bus” Research ”, Yamaguchi University dissertation, Science and Technology Hiroko No.556).

  Subsequently, the virtual displacement data is calculated by causing the vehicle model to virtually run the bridge model 1M in the virtual space of the computer main body 102 (step S15). At this time, the computer main body 102 (CPU 102a) operates as virtual displacement data acquisition means. Specifically, in the virtual space of the computer main body 102, as a result of the vehicle model virtually running on the bridge model 1M, a virtual displacement generated at a predetermined location of the bridge model 1M is acquired as virtual displacement data. The predetermined location corresponds to the installation position of the acceleration sensor AM on the actual bridge 1.

固有ベクトル｛ｕ｝を用いると、式１の解は

とおくことができる。式２を式１に代入すると、任意時刻において

が成り立つ必要があるので、式３が有意な解を持つための条件は

となる。そうすると、式４を満足するω^２（固有値）を求めることにより、式５からｉ（ｉは自然数）次の固有振動数ｆ_ｉを算出することができる。

Subsequently, the computer main body 102 calculates a natural frequency as an analysis value (step S16). At this time, the computer main body 102 (CPU 102a) operates as means for acquiring a natural frequency as an analysis value. Specifically, the analysis value of the natural frequency is obtained as follows. First, from Lagrange's equation of motion, if [M] is a mass matrix, [K] is a stiffness matrix, and {x} is a displacement vector, the equation of motion of free vibration of a multi-degree-of-freedom system (Equation 1) without damping is can get.

Using the eigenvector {u}, the solution of Equation 1 is

It can be said. Substituting Equation 2 into Equation 1, at any time

Therefore, the condition for Equation 3 to have a significant solution is

It becomes. Then, by obtaining ω ² (eigenvalue) that satisfies Equation 4, the i-th natural frequency f _i (i is a natural number) can be calculated from Equation 5.

  Subsequently, the computer main body 102 determines whether or not the actual displacement data and the virtual displacement data are approximated (step S17). At this time, the computer main body 102 (CPU 102a) operates as a determination unit for determining whether or not the actual displacement data and the virtual displacement data are approximate. Specifically, as a result of comparing the actual displacement data and the virtual displacement data, the computer main body 102 determines that the two are approximated when the value obtained by dividing the virtual displacement data by the actual displacement data is within a predetermined range. (YES in step S17), the process proceeds to step S18. On the other hand, as a result of comparing the actual displacement data and the virtual displacement data, the computer main body 102 determines that the two are not approximated when the value obtained by dividing the virtual displacement data by the actual displacement data is not within the predetermined range (step (NO in S17), the process proceeds to step S19. In the determination, at least the extreme value of the virtual displacement data and the extreme value of the actual displacement data may be compared, but all the data may be compared. The predetermined range may be, for example, 95% to 105%, 97% to 103%, or 98% to 102%. Therefore, when comparing extreme values, the computer main body 102 may determine that both are approximate when the extreme value of the virtual displacement data is within ± 5% of the extreme value of the actual displacement data. . In this case, the virtual displacement data is accurately approximated to the actual displacement data at the extreme value where the absolute value of the displacement tends to be large. Therefore, the set value of the bending stiffness EI in the bridge model 1M can be made closer to the actual bending stiffness of the bridge 1 with high accuracy.

Subsequently, the computer main body 102 determines whether or not the experimental value and the analysis value of the natural frequency are approximated (step S18). At this time, the computer main body 102 (CPU 102a) operates as a determination means for determining whether or not the experimental value and the analysis value of the natural frequency are approximate. Specifically, the computer main body 102 compares the experimental value of the natural frequency with the analytical value, and as a result, the value obtained by dividing the analytical value of the natural frequency by the experimental value of the natural frequency is within a predetermined range. It is determined that both are approximated (YES in step S18), and the rigidity estimation process is terminated. Thereby, it is estimated that the actual bridge 1 has the bending rigidity EI set to the bridge model 1M. On the other hand, as a result of comparing the experimental value of the natural frequency with the analytical value, the computer main body 102 approximates both when the value obtained by dividing the analytical value of the natural frequency by the experimental value of the natural frequency is not within the predetermined range. It determines with not having carried out (it is NO at step S18), and progresses to the process of step S19. The even-order value of the natural frequency f _i as the analysis value is the natural frequency in the inverse target mode and cannot occur in the actual bridge 1. and odd values of f _i, is compared with the natural frequency of the experimental values. The predetermined range may be, for example, 95% to 105%, 97% to 103%, 97.5% to 102.5%, or 98% -102% may be sufficient. Therefore, when comparing the experimental value of the natural frequency with the analytical value, the computer main body 102 approximates both when the analytical value of the natural frequency is within ± 5% of the experimental value of the natural frequency. It may be determined that In this case, in addition to the approximation of the virtual displacement data and the actual displacement data, it is also determined whether or not the natural frequencies are approximated in the actual bridge 1 and the bridge model 1M. Therefore, even if the damage state of the bridge 1 is complicated, the damage state of the bridge 1 can be estimated more accurately by comparing the actual bridge 1 and the bridge model 1M from a plurality of viewpoints.

  When it is determined NO in step S17 or step S18, the process proceeds to step S19, and the computer main body 102 changes the value of the bending stiffness EI set in the bridge model 1M (step S19). At this time, the computer main body 102 (CPU 102a) operates as a means for changing the set value of the bending rigidity EI in the bridge model 1M. In step S19, the setting value of the bending rigidity EI in a part of the bridge model 1M may be changed, or the setting value of the bending rigidity EI in the whole bridge model 1M may be changed.

  By the way, when it is determined NO in step S17 or step S18, the bending stiffness EI of the bridge model 1M is different from the bending stiffness EI of the actual bridge 1, and the bridge model 1M reproduces the actual bridge 1. It will not be. Since the value of the bending rigidity EI initially set in the bridge model 1M is a value when the actual bridge 1 is healthy, the fact that the bending rigidity EI of both is different means that the actual bridge 1 is damaged. It is estimated that the bending rigidity EI is reduced. Accordingly, in step S19, the value of the bending stiffness EI set in the bridge model 1M is changed so that the value of the bending stiffness EI of the bridge model 1M is smaller than the initial value.

  When the set value of the bending stiffness EI is changed in step S19, the process returns to step S14. Thereafter, steps S14 to S19 are repeated until the actual displacement data and the virtual displacement data are approximated and the experimental value and the analysis value of the natural frequency are approximated. That is, in the rigidity estimation process, the process of changing the set value of the bending rigidity EI of the bridge model 1M is repeatedly performed until the bridge model 1M reproduces the actual bridge 1.

  Next, the damage degree estimation process will be described with reference to FIG. First, theoretical values of the bending stiffness EI and the skeleton curve (curve indicating the magnitude of the bending moment M with respect to the curvature φ) in the bridge 1 (superstructure 3) are calculated (step S20). At this time, a theoretical value when it is assumed that the bridge 1 (upper structure 3) is healthy, or a theoretical value when it is assumed that predetermined damage is generated in the bridge 1 (upper structure 3) is calculated. As this assumption condition, when the upper structure 3 is made of reinforced concrete, for example, the degree of cracking of the concrete, the number of ruptures of the reinforcing bars (main bars), and the like can be mentioned. In step S20, theoretical values are calculated for the bridge 1 (superstructure 3) having various degrees of damage by changing the assumed amount of cracks in the concrete and the assumed number of fractures of the reinforcing bars (main bars) as parameters. May be. On the other hand, in step S20, the theoretical value may be calculated only for the bridge 1 (superstructure 3) having a predetermined degree of damage. In this case, the assumption may be that the bridge 1 is damaged to the extent that the passage of the vehicle V should be restricted (for example, one main bar is broken). In addition, you may obtain | require the theoretical value when it is assumed that the bridge 1 (superstructure 3) is healthy based on the design standard at the time of construction of the bridge 1. FIG. Examples of design criteria include concrete standard specifications, road bridge specifications, and the like.

The theoretical value of the bending moment M can be obtained by increasing or decreasing the curvature φ in Equation 6 where y is the distance from the neutral axis of the bridge girder cross section to an arbitrary point and σ is the degree of stress at the arbitrary point. Further, the theoretical value of the bending stiffness EI can be obtained from Equation 7.

  Subsequently, the computer main body 102 evaluates the actual degree of damage to the bridge 1 according to the bending rigidity EI of the bridge model 1M obtained in the rigidity estimation process (step S21). At this time, the computer main body 102 (CPU 102a) operates as an evaluation means for evaluating the actual degree of damage to the bridge 1. Specifically, by comparing the bending rigidity EI of the bridge model 1M obtained in the rigidity estimation process with the theoretical value of the bending rigidity EI calculated in step S20, the degree assumed in the theoretical value of the bending rigidity EI. It is estimated that damage has occurred in the actual bridge 1. Further, the surplus proof stress is estimated from the difference between the maximum proof stress in the theoretical value of the skeleton curve at that time and the maximum proof stress of the bridge 1 at the time of soundness. Thus, the damage degree estimation process ends.

[Action]
In the present embodiment as described above, the acceleration generated at a predetermined location of the bridge 1 is measured by the acceleration sensor AM, and the actual displacement data is acquired based on the acceleration data obtained from the acceleration sensor AM. When acquiring actual displacement data based on acceleration data measured by the acceleration sensor AM, the vibration state of the bridge 1 is detected with higher accuracy than directly acquiring actual displacement data using a displacement sensor. That is, the displacement (deflection) during traveling of the vehicle V at a predetermined location of the bridge 1 can be obtained with high accuracy. Therefore, when it is determined that the virtual displacement data is approximate to the actual displacement data, it can be determined that the bridge model 1M accurately reproduces the actual bridge 1. Therefore, when the bending stiffness set in the bridge model 1M is estimated, the bending stiffness of the bridge 1 that cannot be directly measured can be grasped from the bending stiffness of the bridge model 1M. As described above, according to the present embodiment, the bending rigidity of the bridge 1 can be grasped only by the vehicle V traveling on the bridge 1 without accumulating data over time. 1 damage state can be estimated.

  In the present embodiment, the degree of damage to the bridge 1 is estimated by comparing the bending stiffness EI of the bridge model 1M obtained in the stiffness estimation process with the theoretical value of the bending stiffness EI calculated in step S20. . In this case, it is possible to estimate the degree of damage to the bridge 1 even when damage that is difficult to confirm by visual inspection of the appearance has occurred inside the bridge 1.

[Other Embodiments]
As mentioned above, although embodiment of this invention was described in detail, you may add a various deformation | transformation to said embodiment within the range of the summary of this invention. For example, in the above embodiment, the damage state of the bridge 1 is estimated as an example of the structure. However, if the vehicle V can travel, the damage state of other structures other than the bridge 1 can be estimated. It is particularly preferable that the structure for which the damage state is estimated is made of reinforced concrete because the damage state of the structure can be estimated with extremely high accuracy by the estimation method of the above embodiment.

  The number of acceleration sensors AM attached to the bridge 1 (structure) may be one or plural. That is, the measurement location of the acceleration of the bridge 1 (structure) by the acceleration sensor AM may be one or a plurality of locations. When only one acceleration sensor AM is used, the acceleration of the bridge 1 (structure) at a plurality of locations is measured by attaching the acceleration sensor AM to a different location each time a vibration measurement experiment using the vehicle V is performed. May be.

In step S19, the set value of flexural rigidity EI in the upper structure portion _4A 2 of the pier 2 close out of _{3, 4A 3, 4B 2,} 4B 3, 4C 2, 4C 3 bridge model 1M, superstructure of the bridge model 1M portions _4A 1 _of, 4B 1, may be set smaller than the set value of flexural rigidity EI at 4C _1. Of the superstructure 3 of the actual bridge 1, the portions 4A ₂ , 4A ₃ , 4B ₂ , 4B ₃ , 4C ₂ , 4C ₃ in the vicinity of the pier 2 are supported by the pier 2 and receive the load of the superstructure 3. . In addition to this, when the bridge 1 is a continuous girder bridge, a particularly large force is applied to the portions 4A ₂ , 4A ₃ , 4B ₂ , 4B ₃ , 4C ₂ , 4C ₃ when the vehicle V passes through the bridge 1. Cheap. Therefore, a long period of time has passed after the building 4A ₂ , 4A ₃ , 4B ₂ , 4B ₃ , 4C ₂ , 4C ₃ of the actual bridge 1 or the vehicle V repeatedly passes through the bridge 1. Easy to concentrate, damage. Therefore, the setting value of the bending stiffness EI in the portions 4A ₂ , 4A ₃ , 4B ₂ , 4B ₃ , 4C ₂ , 4C ₃ of the bridge model 1M is used as the bending stiffness EI in the portions 4A ₁ , 4B ₁ , 4C ₁ of the bridge model 1M. The setting value of the bending stiffness EI in the bridge model 1M can be made closer to the actual bending stiffness EI of the bridge 1 by setting it to be smaller than the setting value. As a result, the actual damage state of the bridge 1 can be estimated with higher accuracy.

In step S19, the bending stiffness EI in the parts 4A ₂ , 4A ₃ , 4B ₂ , 4B ₃ , 4C ₂ , 4C ₃ of the bridge model 1M is used as the bending stiffness in the parts 4A ₁ , 4B ₁ , 4C ₁ of the bridge model 1M. The bending stiffness EI in the portions 4A ₃ , 4B ₃ and 4C ₃ of the bridge model 1M is set to be smaller than the set value of EI and the bending stiffness EI in the portions 4A ₂ , 4B ₂ and 4C ₂ of the bridge model 1M is set. It may be set smaller than the set value. The portions 4A ₂ , 4B ₂ , and 4C ₂ of the superstructure 3 of the actual bridge 1 receive the load of the suspension girder 5 because they support the suspension girder 5 via the support (Gelbar hinge). In addition, portions 4A ₂ , 4B ₂ , and 4C ₂ of the superstructure 3 of the actual bridge 1 tend to collect rain and the like from the gaps with the suspension girder 5, so that a long period of time has passed after construction and wind and rain It is easy to be eroded by being exposed to. Therefore, damage is particularly likely to concentrate on the actual portions 4A ₂ , 4B ₂ , 4C ₂ of the bridge 1. Accordingly, the setting value of the bending stiffness EI in the portions 4A ₂ , 4B ₂ , 4C ₂ of the bridge model 1M is set to be smaller than the setting value of the bending stiffness EI in the portions 4A ₃ , 4B ₃ , 4C ₃ of the bridge model 1M. Thus, the set value of the bending stiffness EI in the bridge model 1M can be made closer to the actual bending stiffness EI of the bridge 1. As a result, the actual damage state of the bridge 1 can be estimated with higher accuracy.

  Hereinafter, the present invention will be described in more detail with reference to examples. However, the present invention is not limited to the following examples.

In the examples, damage state estimation processing was performed on the former Sakae Bridge (removed as of the filing date of the present application) connecting Waki-machi, Gamagori, Yamaguchi Prefecture, and Otake City, Hiroshima Prefecture. First, the vibration measurement experiment of Sakae Bridge was conducted. Sakae Bridge is a Gerber structure bridge built near the mouth of the Kose River on National Route 2. Details of Sakae Bridge are shown below. Since the Sakae bridge and the bridge 1 in the embodiment have a Gerber structure, the symbols in the bridge 1 of the embodiment may be used in the following description of the Sakae bridge.
Start / end point: 2-chome Minamiei, Otake-shi, Hiroshima to 5-chome Waki, Waki-machi, Yamaguchi Prefecture
Year of completion: March 1953 Bridge length: 168.29m
Width: 11m
1 span: 22.2m
Girder length: 31.8m
Length of bridge girder from pier to end: 4.8m on both ends
Hanging girder length: 12.6m

Acceleration sensors AM1 to AM3 were installed on the upper surface of the upper structure of Sakae Bridge, at the same positions as shown in FIG. That is, the acceleration sensor AM1 is attached to the central portion of the receiving beam (receiving beam 4B) between the fourth diameters, and the receiving beam (receiving beam 4B) between the fourth diameters and the hanging beam (hanging beam 5B) between the fifth diameters. The acceleration sensor AM2 was attached in the vicinity of the support located between the two, and the acceleration sensor AM3 was attached to the central portion of the hanging girder (hanging girder 5B) between the fifth diameters. As the acceleration sensors AM1 to AM3, a servo acceleration sensor SF3000L (noise density at 10 Hz to 1000 Hz: 300 ng _rms / √Hz) manufactured by COLIBRYS was used.

A general sightseeing bus was used as a vehicle for vibration measurement experiments. Details of the bus are shown below.
Total weight: about 15000kgf (about 150kN)
Total length: 6.5m
Center of gravity: 2.17m from rear end
Travel speed during the experiment: approximately 30 km / h

  Under the above conditions, a bus was run from the Hiroshima side of Sakae Bridge to the Yamaguchi side. FIG. 8 shows acceleration data obtained in the acceleration sensor AM1 at this time. A symbol A in FIG. 8 corresponds to the symbol A in FIG. 6, and when the bus passes the position of the symbol A in FIG. 6 (the boundary position between the receiving beam 4 </ b> B and the hanging beam 5 </ b> A), The acceleration generated at the installation position (the central part of the receiving beam 4B) is shown. The same applies to symbols B to E in FIG. That is, FIG. 8 shows the acceleration measured according to the passage of time when the bus gradually moves away from the acceleration sensor AM1.

Next, the actual displacement data was obtained from the acceleration data by second-order integrating the acceleration data obtained by the acceleration sensor AM1 using a bandpass filter in the frequency domain. The acquired actual displacement data is shown in FIG. Reference signs A to E in FIG. 9 also correspond to reference signs A to E in FIG. According to FIG. 9, the following facts are understood.
-As the bus approached the receiving beam 4B from the receiving beam 4A, the displacement of the central portion of the receiving beam 4B gradually increased from 0 mm vertically upward.
-After the bus passed the pier 2C, the displacement of the central part of the receiving beam 4B gradually decreased vertically downward.
-When the bus reached the central part of the receiving beam 4B, the displacement of the central part of the receiving beam 4B became the largest vertically downward.
-After the bus passed through the central part of the receiving beam 4B, the displacement of the central part of the receiving beam 4B gradually increased vertically downward.
-After the bus passed the boundary position between the receiving beam 4B and the hanging beam 5B, the displacement of the central part of the receiving beam 4B gradually decreased vertically downward and approached 0 mm.

  Next, acceleration data obtained by the acceleration sensor AM1 was processed by a modified periodogram method to calculate a power spectrum (energy spectrum density). The calculated power spectrum is shown in FIG. As shown in FIG. 10, the response was excellent at 3.6 Hz, 4.6 Hz, 9.44 Hz, and 13.76 Hz. However, since the frequency at the axle position of the bus was measured and was excellent at about 12 Hz to 13 Hz, the response at around 13 Hz in FIG. 10 is the vibration introduced by the bus traveling on Sakae Bridge. Therefore, the natural frequencies in the vertical direction of the bridge girder of Sakae Bridge were 3.6 Hz, 4.6 Hz, and 9.44 Hz.

  Next, vibration simulation was performed on the computer main body 102. First, for the bridge model 1M, the elastic modulus E of concrete obtained based on the design criteria at the time of construction of Sakae Bridge and the bending stiffness EI are set as the values of Sakae Bridge at the time of healthy construction (Condition 1). The natural frequency as the virtual displacement data and the analysis value was calculated. The calculated virtual displacement data is shown in FIG. 11, and the calculated analysis value of the natural frequency is shown in FIG. In FIG. 11, the actual displacement data is indicated by a solid line, and the virtual displacement data is indicated by a broken line.

  Next, the computer main body 102 determines whether or not the actual displacement data and the virtual displacement data are approximate. Specifically, the extreme value of the virtual displacement data and the extreme value of the actual displacement data in the vicinity of 7.5 seconds in FIG. 11 are compared, and the value obtained by dividing the virtual displacement data by the actual displacement data is 95% to 105%. Judged whether it was within the range. In this example, the value obtained by dividing the virtual displacement data by the actual displacement data was 103%, so it was determined that both were approximate.

  Next, the computer main body 102 determined whether or not the experimental value and the analysis value of the natural frequency are approximate. Specifically, it was determined whether or not the value obtained by dividing the analysis value of the natural frequency by the experimental value of the natural frequency is in the range of 95% to 105%. In this example, since the value obtained by dividing the analysis value of the natural frequency by the experimental value of the natural frequency exceeded 105%, it was determined that both were not approximated. Therefore, the bridge model 1M in which the bending stiffness EI at the time of the construction of Sakae Bridge is set is the same as the actual Sakae Bridge, but the natural frequency is not approximated. It turned out that it was not reproduced.

  Therefore, the overall bending rigidity EI of the bridge model 1M was set to a value 30% lower than the bending rigidity EI of the Sakae Bridge when sound (condition 2), and the vibration simulation was performed again on the computer main body 102. As a result, the calculated virtual displacement data is shown in FIG. 12, and the calculated analysis value of the natural frequency is shown in FIG.

  Next, the computer main body 102 determines whether or not the actual displacement data and the virtual displacement data are approximate. In this example, the value obtained by dividing the virtual displacement data by the actual displacement data was 146%, so it was determined that the two were not approximate. Note that it was also determined by the computer main body 102 whether or not the experimental value and the analysis value of the natural frequency were approximate. In this example, the value obtained by dividing the analysis value of the natural frequency by the experimental value of the natural frequency was in the range of 91% to 103%. Therefore, the bridge model 1M when the overall rigidity is reduced by 30% is closer to the real Sakae bridge compared to the bridge model 1M when healthy, but the displacement behavior does not reproduce the real Sakae bridge. It turned out that the actual bending stiffness EI of Sakae Bridge was not reproduced. When the location and degree of damage are complex, the experimental value and the analysis value of the natural frequency are not approximated even if the actual displacement data and the virtual displacement data are approximated as in this example. Or vice versa.

Therefore, the set value of the bending stiffness EI in the portion near the pier (portions 4A ₂ , 4A ₃ , 4B ₂ , 4B ₃ , 4C ₂ , 4C ₃ etc.) in the bridge model 1M is used as the other portion (portion) of the bridge model 1M. 4A ₁ , 4B ₁ , 4C ₁ etc.) were set to be smaller than the set value of the bending rigidity EI, and the vibration simulation was performed again on the computer main body 102. Specifically, in the superstructure 3 of the bridge model 1M, the bending rigidity EI of the center quarter region of the span is set to the value at the time of soundness of the Sakae Bridge, and the bending rigidity EI of the other region is set at the time of the soundness. The value was set 30% lower than the value (Condition 3). As a result, the calculated virtual displacement data is shown in FIG. 13, and the calculated analysis value of the natural frequency is shown in FIG. At this time, since the value obtained by dividing the virtual displacement data by the actual displacement data was 127%, it was determined that the two were not approximate. Moreover, since the value obtained by dividing the analysis value of the natural frequency by the experimental value of the natural frequency was in the range of 93% to 101%, it was determined that the two were not approximated.

  Similarly, in the superstructure 3 of the bridge model 1M, the bending stiffness EI in the center half area of the span is set to the value at the time of soundness of the Sakae Bridge, and the bending stiffness EI of the other region is determined from the value at the time of soundness. Was also set to a value 30% lower (Condition 4). As a result, the calculated virtual displacement data is shown in FIG. 14, and the calculated analysis value of the natural frequency is shown in FIG. At this time, since the value obtained by dividing the virtual displacement data by the actual displacement data was 103%, it was determined that both were approximate. Moreover, since the value obtained by dividing the analysis value of the natural frequency by the experimental value of the natural frequency was in the range of 97% to 104%, it was determined that both were approximate. From the results in conditions 3 and 4, the bridge model 1M reproduces the actual Sakae bridge better than the condition 2 in both the conditions 3 and 4, but the reproducibility of the actual Sakae bridge by the bridge model 1M is particularly good in the condition 4. It was confirmed to be high.

  Here, whether or not the bridge model 1M accurately reproduces the entire Sakae Bridge was confirmed by the following procedure. First, in the vicinity of the gel bar hinge between the receiving beam 4B and the hanging beam 5B, the computer main body 102 determines whether or not the actual displacement data of the acceleration sensor AM2 and the virtual displacement data under the condition 4 are approximate. . FIG. 16A shows the setting state of the bending stiffness EI and the measurement position of the displacement in the bridge model 1M under the condition 4, and FIG. 16B shows the actual displacement data of the acceleration sensor AM2 and the virtual displacement under the condition 4. Data.

  When the extreme value of the virtual displacement data and the extreme value of the actual displacement data in the vicinity of 7.5 seconds in FIG. 16B are compared, the value obtained by dividing the virtual displacement data by the actual displacement data is 103%. Therefore, it was determined that the two were not approximate. As a result, in condition 4, it was confirmed that the reproducibility of the actual Sakae bridge by the bridge model 1M was not sufficient, particularly near the Gerber hinge.

  Further, the computer main body 102 determines whether or not the actual displacement data of the acceleration sensor AM3 and the virtual displacement data in the condition 4 are approximate in the central portion of the hanging beam 5B. FIG. 17A shows the setting state of the bending rigidity EI and the measurement position of the displacement in the bridge model 1M under the condition 4, and FIG. 17B shows the actual displacement data of the acceleration sensor AM3 and the virtual displacement under the condition 4. Data.

  Comparing the extreme value of the virtual displacement data and the extreme value of the actual displacement data in the vicinity of 10 seconds in FIG. 17B, the value obtained by dividing the virtual displacement data by the actual displacement data was 103%. Therefore, it was determined that both were approximated at the central portion of the hanging beam 5B.

Therefore, the set value of the bending stiffness EI in the portion near the pier (portions 4A ₂ , 4A ₃ , 4B ₂ , 4B ₃ , 4C ₂ , 4C ₃ etc.) in the bridge model 1M is used as the other portion (portion) of the bridge model 1M. 4A ₁ , 4B ₁ , 4C ₁ etc.) and set to be smaller than the set value of bending rigidity EI. In addition, the set value of the bending rigidity EI in the part from the pier to the end of the bridge model 1M (parts 4A ₂ , 4B ₂ , 4C ₂ etc.) is set between the center part and the end part of the bridge model 1M. It was set smaller than the set value of the bending stiffness EI in (parts 4A ₃ , 4B ₃ , 4C ₃ etc.). Specifically, the bending stiffness EI of the center half area of the span of the superstructure 3 of the bridge model 1M is set to the value at the time of the soundness of the Sakae bridge, and from the bridge pier to the center half area. The bending rigidity EI was set to a value 30% lower than the value at the time of soundness, and the bending rigidity EI from the bridge pier to the end of the receiving girder was set to a value 50% lower than the value at the time of soundness (conditions) 5). Then, the vibration simulation was performed again on the computer main body 102.

  In this vibration simulation, first, the computer main body 102 determines whether or not the actual displacement data of the acceleration sensor AM1 and the virtual displacement data under the condition 5 are approximated at the center of the receiving beam 4B. 18A shows the setting state of the bending stiffness EI and the measurement position of the displacement in the bridge model 1M under the condition 5, and FIG. 18B shows the actual displacement data of the acceleration sensor AM1 and the virtual displacement under the condition 5. Data.

  When the extreme value of the virtual displacement data and the extreme value of the actual displacement data in the vicinity of 7.5 seconds in FIG. 18B were compared, the value obtained by dividing the virtual displacement data by the actual displacement data was 103%. For this reason, it is determined that the two are approximate in the central portion of the receiving beam 4B.

  Further, in the vicinity of the gel bar hinge between the receiving beam 4B and the hanging beam 5B, the computer main body 102 determines whether or not the actual displacement data of the acceleration sensor AM2 and the virtual displacement data under the condition 5 are approximate. . FIG. 19A shows the setting state of the bending stiffness EI and the measurement position of the displacement in the bridge model 1M under the condition 5, and FIG. 19B shows the actual displacement data of the acceleration sensor AM2 and the virtual displacement under the condition 5. Data.

  When the extreme value of the virtual displacement data in the vicinity of 7.5 seconds in FIG. 19B is compared with the extreme value of the actual displacement data, the value obtained by dividing the virtual displacement data by the actual displacement data is 104%. Therefore, it was determined that the two were approximated in the vicinity of the gel bar hinge between the receiving beam 4B and the hanging beam 5B.

  Further, the computer main body 102 determines whether or not the actual displacement data of the acceleration sensor AM3 approximates the virtual displacement data under the condition 5 in the central portion of the hanging beam 5B. 20A shows the setting state of the bending stiffness EI and the measurement position of the displacement in the bridge model 1M under the condition 5, and FIG. 20B shows the actual displacement data of the acceleration sensor AM3 and the virtual displacement under the condition 5. Data.

  When the extreme value of the virtual displacement data and the extreme value of the actual displacement data in the vicinity of 10 seconds in FIG. 20B were compared, the value obtained by dividing the virtual displacement data by the actual displacement data was 104%. Therefore, it was determined that both were approximated at the central portion of the hanging beam 5B.

  From the above, it was confirmed that the real Sakae bridge could be reproducible with extremely high accuracy by the bridge model 1M under the condition 5.

1 ... bridges, 3 ... superstructure, 4 ... received digits, 5 ... hanging _{girder, 4A 1 ~4A 3, 4B 1} ~4B 3, 4C 1 ~4C 3 ... portion, 100 ... estimation system, 102 ... computer, 102a ... CPU, 102e ... external storage device, AM (AM1-AM3) ... acceleration sensor, V ... vehicle.

Claims (9)

A first step of measuring, by the acceleration sensor, acceleration generated at the predetermined location of the structure when the vehicle travels through the structure where the acceleration sensor is disposed at the predetermined location;
A second step of acquiring actual displacement data based on the acceleration data obtained from the acceleration sensor in the first step;
In the virtual space, the vehicle model which simulates the vehicle, by imitating to and rigid the structure causes virtually running on set structure model to a predetermined value, among the structure models the A third step of acquiring, as virtual displacement data, a virtual displacement occurring at a virtual position corresponding to the predetermined location;
Damage to the structure including the fourth step of determining whether the actual displacement data obtained in the second step is approximate to the virtual displacement data obtained in the third step A method of estimating the state. As a result of the determination in the fourth step, if not approximate, further includes a fifth step of changing a stiffness setting value in a part or all of the structure model,
The method according to claim 1, wherein the third and fourth steps are executed again using the stiffness set value after the change in the fifth step. The structure is a bridge comprising a plurality of piers and an upper structure supported on the plurality of piers,
In the fifth step, if the result of determination in the fourth step is not an approximation, the stiffness setting value in the first portion near the pier in the upper structure of the structure model is set. The method according to claim 2, wherein the stiffness is set to be smaller than a set value of rigidity in a second portion different from the first portion in the superstructure of the structure model. The structure is
First to fourth piers, and first receiving girders supported on the first and second piers,
A second receiving girder supported on the third and fourth piers;
A gel bar-structured bridge comprising a hanging girder positioned between the first and second girder and supported on the first and second girder via a gel bar hinge;
The first receiving digit is
A first portion extending from the first pier near the suspension girder to an end of the suspension girder,
A second portion extending between the first and second piers;
A third portion extending from the first pier to the second portion,
The second receiving digit is
A fourth portion extending from the third bridge pier close to the suspension girder to the end on the suspension girder side;
A fifth portion extending between the third and fourth piers;
A sixth portion extending from the third pier to the fifth portion,
In the fifth step, if the result of determination in the fourth step is not approximate,
Setting stiffness set values in the third and sixth portions of the structure model to be smaller than stiffness set values in the second and fifth portions of the structure model;
3. The method according to claim 2, wherein a stiffness setting value in the first and fourth portions of the structure model is set smaller than a stiffness setting value in the third and sixth portions of the structure model. .   In the fourth step, approximation is performed when the extreme value of the virtual displacement data obtained in the third step is within ± 5% of the extreme value of the actual displacement data obtained in the second step. The method as described in any one of Claims 1-4 which determines with having carried out. A sixth step of obtaining a natural frequency of the structure based on the acceleration data obtained in the first step;
In the third step, the natural frequency of the structure model is also obtained,
The method according to any one of claims 1 to 5, wherein in the fourth step, it is also determined whether or not the natural frequencies obtained in the third and sixth steps are approximate to each other.   If the result of determination in the fourth step is an approximation, the setting value of rigidity in the structure model at that time is the value obtained when it is assumed that predetermined damage has occurred in the structure. The method according to claim 1, wherein the degree of damage of the structure is estimated by comparison with a theoretical value of rigidity of the structure. On the computer,
Based on the acceleration data obtained by the acceleration sensor arranged at the predetermined location measuring the acceleration generated at the predetermined location of the structure when the vehicle travels on the structure, the actual displacement data at the predetermined location is obtained. A first process to acquire;
In the virtual space, the vehicle model which simulates the vehicle, by imitating to and rigid the structure causes virtually running on set structure model to a predetermined value, among the structure models the A second process of acquiring a virtual displacement generated at a virtual position corresponding to the predetermined location as virtual displacement data;
A third process for executing a third process for determining whether or not the actual displacement data obtained in the first process is approximate to the virtual displacement data obtained in the second process; A program that estimates damage status. An acceleration sensor that is arranged at a predetermined location of the structure and measures acceleration generated at the predetermined location of the structure when a vehicle travels through the structure;
First acquisition means for acquiring actual displacement data based on acceleration data obtained from the acceleration sensor;
In the virtual space, the vehicle model which simulates the vehicle, by imitating to and rigid the structure causes virtually running on set structure model to a predetermined value, among the structure models the Second acquisition means for acquiring virtual displacement occurring at a virtual position corresponding to a predetermined location as virtual displacement data;
A damage state of the structure, comprising: a determination unit that determines whether or not the actual displacement data acquired by the first acquisition unit is approximate to the virtual displacement data acquired by the second acquisition unit. Estimating system.

Images