KR20090082613A - Hybrid damage monitoring system for prestressed concrete girder bridges - Google Patents

Abstract

A hybrid damage monitoring system for a prestressed concrete girder bridge is provided to apply damage monitoring techniques for the existing PSC girder bridge by classifying damages, to secure the usability and safety of a bridge by constructing a monitoring system using a developed system, and to heighten the efficiency of a safety diagnosis for a structure by automation of a monitoring system. A hybrid damage monitoring system for a prestressed concrete girder bridge uses acceleration and impedance signal characters and comprises the first step of monitoring the variation of the acceleration signal characters obtained from a PSC girder bridge and alarming damage to the whole area and the second step of distinguishing the type of damage from the variation of the impedance signal characters obtained from a tendon anchorage.

Description

Hybrid damage monitoring system for prestressed concrete girder bridges

The present invention relates to a damage monitoring system for the safety diagnosis of a PSC girder bridge, and more particularly, to alert the occurrence of a decrease in the bending stiffness, such as a decrease in tension of PC tendon, which is a typical damage type of the PSC girder bridge, and cracking of concrete. The present invention relates to a hybrid damage monitoring system that uses a vibration-based damage monitoring technique and an impedance-based damage monitoring technique to classify which of the damage types is due to.

In general, prestressed concrete is defined as concrete that has given the strength by artificially determining the distribution and size of the stress in advance so that the stress caused by the external force can be canceled to a predetermined limit. Thus, even if tensile stress acts on the lower surface of the concrete beam due to the initially introduced tension force, the tensile stress is controlled so as not to exceed the flexural tensile strength of the concrete to prevent the occurrence of cracking.

Among the typical types of damage that threaten the safety of PSC bridges, flexural cracking of concrete is formed when the stress at the ultimate tensile surface is greater than the failure coefficient of concrete, and is caused by internal expansion or contraction of concrete components, corrosion, dead weight, or the weight of the member. The microcracks generated by the small trailing stresses cause the main crack to occur due to the stress difference between the reinforcing bar and the concrete in the cracked section. Larger cracks can corrode reinforcing bars and PC tendons, which leads to a crack-promoting effect through secondary degradation of flexural stiffness, leading to flexural failure of the members. Reduction of tension in PC tendon also includes the behavior of elastic deformation, friction and anchorage stages during prestressing, and after the introduction of dry shrinkage and creep of concrete, relaxation of PC steels, corrosion of PC anchorages and stress corrosion and fatigue cracking of PC steels. It is caused by such a cause. This reduction in the amount of prestress results in a decrease in the flexural resistance of the concrete member, a decrease in the tensile tension cracking and the crack closing effect.

Since the 1990s, many researchers have been trying to apply the change of dynamic characteristics as a monitoring method for specific damage types of PSC bridges. In 1994, Saiidi et al. 1 attempted to monitor changes in tension from natural frequency changes, and in 2000 Miyamoto et al. Conducted a study on the dynamic behavior of prestressed composite girder bridges to provide basic information for monitoring. To provide. In addition, a study was conducted to identify the change in tension from the dynamic response measured in Prestress Bo, by Kim et al. [2003] and [2005] Law and Lu [2005]. In addition, in 1999, Abdel Wahab and De Roeck [5] used the mode curvature change to monitor the actual damage of PSC bridges. In 2003, Kim [6] proposed a damage detection method using modal strain energy. In 2005, Huth et al. Analyzed the sensitivity of a vibration-based damage detection method to monitor the progression of cracks in PSC highway bridges.

However, the techniques used in these studies are limited in that they can be used when the type of damage is known. This is because both the reduction in tension and the damage caused by lowering of flexural stiffness result in a change in the mode characteristics. Therefore, it is difficult to monitor both damage types directly using the above techniques, and a new approach is required.

Recently, with the development of smart materials, impedance-based monitoring using smart piezoelectric materials has been widely used as a new damage detection technique. The basic idea of this method is to give a high frequency structural stimulus to monitor the local area of the structure. In the high frequency area, it is sensitive to the initial changes that have little effect on the integrity of the structure. Compared with vibration-based techniques, impedance-based techniques are an effective method for locating damage that requires more precise measurements, such as smaller cracks. However, local monitoring using the impedance technique is difficult to monitor the overall characteristic change of the structure.

In the study of applying this impedance-based technique to the soundness evaluation of civil engineering structures, a study was conducted to monitor local hazards affecting the usability of the entire structure. Impedance techniques were applied to the modeled bridge members to monitor the damage of the members and to verify the effects of temperature. In 2003, Bhalla and Soh [9] conducted a study to monitor the damage of model reinforced concrete frames. Also, in 2006, Kim et al. [10] conducted a study to monitor the change of point condition through model plate bridge test.

Saiidi, M., Douglas, B., and Feng, S., Prestress force effect on vibration frequency of concrete bridges, Journal of Structure Engineering, 1994, Vol. 120, ISBN 0733-9445, pp. 223-2241

Miyamoto, A., Tei, K., Nakamura, H., and Bull, J.W., Behavior of prestressed beam strengthened with external tendons, Journal of Structural Engineering, 2000, Vol. 126, ISBN 0733-9445, pages 1033-1044

[3] Kim, J.T., Yun, C.B., Ryu, Y.S., and Cho, H.M., Identification of prestress-loss in PSC beams using modal information, Structural Engineering and Mechanics, 2003, Vol. 17, ISBN 1225-4568, pp. 467-482

4, Law, S.S, and Lu, J.R., Time domain response of a prestressed beam and prestress identification, Journal of Sound and Vibration, 2005, Vol. 288, ISBN 0022-460X, pp. 1011-1025

[5] Abdel Wahab, M.M., De Roeck, G., Damage detection in bridges using modal curvatures: Application to a real damage scenario, Journal of Sound and Vibration, 1999, Vol. 226, ISBN 0022-460X, pp. 217-235

[Document 6] Kim, JT, Ryu, YS, Cho, MH, Stubbs, N., Damage identification in beam-type structures: frequency-based method vs mode-shape-based method, Engineering structures, 2003, Vol. 25, ISBN 0141-0296, pp. 57-67

[7] Huth, O., Feltrin, G., Masck, J., Kilic, N., Motavalli, M., Damage identification using modal data: Experiences on a prestressed concrete bridge, Journal of Structural Engineering, 2005, Vol . 131, ISBN 0733-9445, pp. 1898-1910

8, Park, G., Gudney, H., and Inman, D.J., Impedance-based health monitoring of civil structural components, Journal of Infrastructure Systems, ASCE, 2000, Vol. 6, ISBN 1076-0342, pp. 153-160

[9] Bhalla, S., and Soh, C.K., Structural impedance based damage diagnosis by piezo-transducers, Earthquake Engineering and Structural Dynamics, 2003, Vol. 32, ISBN 0098-8847, pp. 1897-1916

10, Kim, J.T., Na, W.B., Park, J.H., and Hong, D.S., Hybrid health monitoring of structural joints using modal parameters and EMI signatures, proceeding of SPIE-the international Society for Optical Engineering, 2005, Vol. 288, ISBN 0277-786X, no. 617420

Two types of damages of PSC girder bridges are damage due to cracking of concrete girder and reduction of tension of tendon tendon. Existing damage monitoring and nondestructive diagnostic techniques can be selectively applied to any of these types of damages, but it is difficult to detect which of the two types of damage has occurred in practice. Therefore, accurate structural diagnosis of PSC bridges requires a system that can alert the damages and classify the types of damages.

In the present invention, a system using a vibration-based monitoring method and an impedance-based monitoring method was developed in order to alert the damage generated in the PSC girder bridge and classify the damage type.

Since the vibration signal of the structure measured from the accelerometer changes for all damages occurring in the structure, it is easy to determine whether it is damaged by using a few sensors, and the electro-mechanical impedance signal measured from the piezoelectric sensor is very close to the sensor. Respond sensitively to damage in the local area.

Using the characteristics of these two techniques, the following monitoring system can be configured to alert the occurrence and types of damage to the two representative types of damage of the PSC girder bridge. 1) Install the accelerometer at any two positions to measure the vibration signal of the PSC girder bridge, and install the piezoelectric sensor at the anchorage of the tendon for impedance measurement. 2) Select the frequency range where the frequency response ratio of the acceleration signal measured at two positions before damage and the change of the impedance signal can be easily measured. 3) Periodically measure the change in the frequency response ratio of the acceleration signal to alert the occurrence of damage. 4) When an alarm of damage occurs, monitor the change in the impedance signal within the selected frequency range. 5) According to the change of the impedance signal, it is possible to discriminate whether it is due to the decrease of tension force of tendon or the decrease of bending stiffness such as cracking.

Through the monitoring system configured as above, first, damage classification is possible, so that the existing damage monitoring techniques for PSC girder bridges can be applied to each damage, and second, a continuous monitoring system using the developed system is established. If so, the usability and safety of the bridge can be secured. Third, the efficiency of structural safety diagnosis can be increased through automation of the monitoring system.

The present invention is to develop a hybrid damage monitoring system that can be damaged alarm and classification for the degradation of flexural stiffness such as the reduction of tension of PC tendon and concrete cracks that is a typical type of damage threatening the safety of PSC girder bridge as shown in FIG. In this study, two phases were constructed using the vibration-based technique, which is easy to monitor the abnormal state in the global area of the target structure, and the impedance technique, which is easy to monitor the initial damage only in the local area.

In the first step (Step 1 of Figure 1), the change in acceleration signal characteristics obtained from the PSC girder bridge is monitored to alert the occurrence of global damage, and the second step (Step 2 in Figure 1) is obtained from the tendon anchorage The type of damage is determined from the change in the impedance signal characteristics.

Looking at the algorithm used in the first and second stages,

First, in the first step, a frequency ratio assurance criterion (FRRAC) was newly proposed to inform the occurrence of global damage by using the frequency response function of the acceleration signal. In order to obtain the frequency response function of the conventional acceleration signal, information over time of the load is required. However, for most civil structures, it is difficult to measure the time history information of the load. Therefore, the frequency response function ratio is used to extract the information irrelevant to the load. This is described below.

와 구조물응답

사이에는 다음과 같은 관계가 성립된다.The frequency response ratio guarantee index can be derived from the frequency response function obtained by converting the vibration response in the time domain to the frequency domain. The frequency response function can be calculated from the relationship between the spectra and the spectrum obtained from the measurement records of the input and response. Input load to the structure

And structure response

The following relationship is established between them.

(1)

(One)

,

,

는 각각 구조물의 질량, 감쇠, 강성계수를 나타내며, 위의 식을 푸리에 변환하여 다음과 같이 주파수 영역에서의 입력하중과 구조물응답 사이의 관계로 나타낼 수 있다.here,

,

,

Represents the mass, damping, and stiffness coefficients of the structure, respectively, and can be expressed as the relationship between the input load and the response of the structure in the frequency domain by Fourier transform.

(2)

(2)

)는 다음과 같이 구할 수 있다.At this time, the frequency response function of the structure (

) Can be obtained as

(3)

(3)

와

는 각각 주파수영역에서의 하중과 변위 응답을 나타낸다.here,

Wow

Are the load and displacement responses in the frequency domain, respectively.

위치에 가진주파수

의 동적 하중이 가해진다면, 구조물과의 공진에 의하여

번째 절점과

번재 절점에서의 응답은

번째 모드형상에 따라 진동한다. 따라서,

번째 절점과

번째 절점의 주파수응답함수비는 다음과 같으며,Node of Structure with Proportional Attenuation

Frequency with position

If a dynamic load of is applied, the resonance with the structure

With the first node

The response at the burnt node is

Vibrate according to the first mode shape. therefore,

First node

The frequency response function ratio of the first node is

(4)

(4)

The frequency response ratio (FRR) function is defined as follows.

(5)

(5)

,

는 각각 상호 스펙트럼 밀도 함수(cross-spectral density function), 자기 스펙트럼 밀도 함수(auto-spectral density function)를 나타내고,

는 평균을 의미한다. here,

,

Denote cross-spectral density functions and auto-spectral density functions, respectively.

Means mean.

       By comparing the frequency response ratio in the intact state with the frequency response ratio in the intact state, the following frequency response ratio guarantee index is defined.

(6)

(6)

,

는 각각 비손상과 손상 상태를 표시한다. 식 (6)은 비손상 상태의 주파수응답비

과 손상 상태의 주파수응답비

의 선형 관계식이다. 만약,

값이 1에 가까운 값이라면 비손상상태로, 1보다 작은 값을 가진다면 손상발생 상태로 판별하게 된다.Where subscript

,

Indicates an intact and damaged state, respectively. Equation (6) is the frequency response ratio in the intact state.

Frequency Response Ratio in Over and Damaged Conditions

Is a linear relation of. if,

If the value is close to 1, it is intact, and if it is less than 1, it is determined as damaged.

Next, in the second step, the change in impedance was monitored to determine the cause of the damage alarm. This technique is intended to measure impedance after attaching a piezoelectric zirconate-titanate (PZT) patch to the tendon's fusing unit, and directly monitor the tension reduction from the characteristic change of the measured impedance.

The impedance of a structure-sensor system is defined as the inverse of the admittance:

(7)여기서,

,

,

는 각각 PZT 패치의 폭, 길이, 두께를 나타내며,

,

,

는 각각 PZT 패치의 전기적 성질을 나타내는 것으로, 유전율, 유전체 손상 탄젠트 및 압전결합상수이다.

는 PZT 패치의 역학적 성질을 나타내는 것으로, 복소탄성계수이다. 또한,

,

는 각각 PZT 패치와 구조물의 역학적 임피던스를 나타낸다.

(7) where

,

,

Represents the width, length and thickness of each PZT patch,

,

,

Are the electrical properties of the PZT patch, respectively, dielectric constant, dielectric damage tangent and piezoelectric coupling constant.

Represents the mechanical properties of the PZT patch and is a complex modulus of elasticity. Also,

,

Represents the mechanical impedance of the PZT patch and the structure, respectively.

가 변화하게 되고, 이는 곧 전체 임피던스의 변화를 의미한다. Thus, if damage occurs around the sensor, the mechanical impedance of the structure

Will change, which means a change in the overall impedance.

As such, by measuring the electrical impedance generated due to the interaction between the structure and the PZT patch, and monitoring the change in the impedance, it is possible to determine the abnormal state of the local region of the structure. As an index indicating a quantitative change in impedance, a root-mean-square deviation (RMSD) such as Equation (8) is used, or a correlation coefficient such as (9) is used.

(8)

(8)

(9)

(9)

와

는 각각 손상 전후의 임피던스 응답을 나타내며,

와

는 각 임피던스 응답의 평균과 표준편차를 나타낸다. 또한,

는 괄호 내에 주어진 값의 평균을 나타낸다. RMSD나 상관계수값이 각각 100%나 1 보다 작아지면 손상 발생을 의미한다.here,

Wow

Represents the impedance response before and after each damage,

Wow

Represents the mean and standard deviation of each impedance response. Also,

Represents the mean of the values given in parentheses. If the RMSD or the correlation coefficient is less than 100% or 1, respectively, it means damage occurs.

In order to verify the applicability of the hybrid monitoring system to the actual PSC girder bridge, damage monitoring was performed by model PSC girder bridge.

Figure 2 shows the shape and reinforcement of the model PSC girder fabricated for verification. As shown in FIG. 2, a model PSC girder having a total length of 6.4m having a T-shaped cross section was manufactured, and the reinforcing bars were reinforced using a deformed bar having a nominal diameter of 9.53 mm D10. In addition, 7 stranded wire (Grade 250) having a diameter of 15.2 mm was used as a tendon, and was arranged in a straight line at the center of the lower flange. The tendon was installed in a 25 mm diameter duct and was not grouted. The 28-day compressive strength of concrete is 23.6 MPa and the unit weight is 2400 kgf / m ³ . Initial tension was introduced using a hydraulic jack (see Table 1). In addition, the wedge for fixing the tendon when the tension force is introduced on the other side of the hydraulic jack is not installed.

<Table 1: Introduction of tension force using hydraulic jack>

3 and 4 show the position of the acceleration sensor and the PZT sensor installed in the model PSC girder. As for the acceleration signal, a total of two piezoelectric accelerometers (PCB 393B04) (two at the top of the girder) were used, and the positioning of the two acceleration sensors at the top of the girder was found in the analysis of the mode shape in the dynamic characteristics analysis on a simple supported beam. Several high mode sensitive locations were selected and attached. In addition, the excitation force was applied at the position of 0.95m from the right end of the girder using an impact hammer. In case of PZT sensor for impedance signal measurement, MFC patch (25.4x12.7x0.254mm) was used to make up for the weakness of the existing PZT patch and to easily attach the surface of the wedge having curved surface.

National Instruments' PXI 4472 DAQ and PXI-8186 controllers were installed to measure the acceleration, and LabVIEW measured the acceleration signal at a sampling frequency of 1 kHz. For impedance measurement, HIOKI 3532, an impedance analyzer, was installed and the signal was measured by applying a voltage of 1V using LabVIEW. Table 3 shows the acceleration and impedance signals measured using LabVIEW.

Table 3: Acceleration and Impedance Signals Measured Using LabVIEW

Based on the experimental configuration described above, the results of monitoring the tension reduction and the decrease in the flexural stiffness of concrete, which are typical damage types of the PSC girder bridge of the present invention, were examined.

PC Tendon Tension  For reduction hybrid  damaged monitoring

In order to perform the experiment on the PC Tendon's tension reduction, the tension force was introduced using a hydraulic jack, and the magnitude of the tension force was measured using a load cell. In order to facilitate the introduction of the tension force, the hydraulic jack is attached to the end of the tendon with a threaded steel bar, and then fixed with a nut. The opposite side is fixed with a wedge. In order to minimize the dynamic change of the structure caused by the load of the hydraulic jack, the hydraulic jack was separated into the structure and the signal was acquired after the tension force was introduced. The experiment was performed with an initial tension decreasing from 17.6 kN (PS 1) to at least 39.2 kN (PS 5) by 19.6 kN for each dynamic test step (see Table 3). In addition, eight experiments were performed at each tension reduction step.

<Table 3: Damage Scenarios Selected for Tension Reduction Experiments of PC Tendon>

Prestress Level PS 1 PS 2 PS 3 PS 4 PS 5 Prestress Force (kN) 117.6 98.0 78.4 58.8 39.2

Step 1 of the hybrid monitoring system of FIG. 1 is a damage warning step to signal the occurrence of global damage of the PSC bridge. The frequency response ratio guarantee index was used to alert the occurrence of damage by using the frequency response function of the acceleration signals measured from accelerometers (Sensor 1 and Sensor 2) at two different positions. FIG. 5 shows the frequency response ratios calculated by the frequency response functions of the sensor 1 and the sensor 2 and the frequency response functions calculated at the two positions for each tension reduction step. FIG. 6 shows a result of calculating the frequency response ratio guarantee index by comparing the frequency response ratios of the remaining stress reduction stages (PS 2-PS 5) based on the frequency response ratio in the state where PS 1 is introduced. In addition, in each of the eight experiments, all experimental data were compared based on the first experiment of PS 1. As can be seen in Figure 6, in the case of PS 1 it can be seen that the change in the frequency response ratio guarantee index hardly occurs, the frequency response guarantee index can be seen at the stage of PS 2 occurs, until PS 3 occurs Stays constant. The same is true for PS 4 and PS 5. From these results, it can be seen that the damage alarm for all stress reduction steps is achieved by using the frequency response ratio guarantee index.

Next, Step 2 of the hybrid monitoring system of FIG. 1 identifies whether the cause of the damage alarm is caused by a decrease in the tension of the PC tendon or a decrease in the bending stiffness. 7 shows the impedance signal measured from the MFC patch attached to the wedge of PC tendon. A change in the impedance signal indicates that a structure damage or other physical change of the structure has occurred in the vicinity of the MFC patch, and it can be seen that the impedance signal moves to the right. From this, it can be classified that the cause of the damage alarm is due to the decrease in tension.

For reducing concrete flexural stiffness hybrid  damaged monitoring

와 같다. 여기서,

은 부가질량,

은 부가질량이 재하된 위치에서 거더단면의 질량,

는 거더의 휨 강성,

는 부가질량에 대응하는 휨 강성의 변화를 나타낸다. 실험은 초기 긴장력 117.6kN (PS 1)이 작용한 상태에서 부가질량을 재하하였으며, 콘크리트 블록 1개(Loading 1)와 2개(Loading 2)를 재하한 조건을 고려하였다. 각각의 재하조건에 대해 8회의 실험을 수행하였다.Added mass experiments were performed to test the degradation of the flexural stiffness of concrete without simulating artificial cracks in the PSC girders. The additive mass test is a method of indirectly simulating the flexural stiffness degradation of concrete against flexural cracking through the additive mass. For this purpose, a concrete block of 1.2 kN (0.5x0.5x0.25m) was used (see Table 4). The relationship for simulating the reduction in flexural stiffness through the additive mass test is

Same as here,

Silver addition mass,

Is the mass of the girder section at the location where the additional mass is loaded,

Is the bending stiffness of the girder,

Denotes a change in bending stiffness corresponding to the added mass. In the experiment, the additional mass was loaded in the state of the initial tension of 117.6 kN (PS 1), and the conditions of loading one concrete block (Loading 1) and two (Loading 2) were considered. Eight experiments were performed for each loading condition.

Table 4: Damage scenarios selected for concrete flexural stiffness degradation test

First, in order to perform the damage warning of Step 1 of FIG. 1, the frequency response function and the frequency response ratio were calculated in Sensor 1 and Sensor 2 as shown in FIG. From this, the frequency response guarantee index was calculated as shown in FIG. Also, the frequency response ratio guarantee index was obtained by comparing the frequency response ratios of the remaining bending stiffness reduction stages based on the frequency response ratio without loading the concrete block, and all experiments based on the first experiment without loading. The data was compared. As a result, there was almost no change in the frequency response ratio guarantee index in the unloaded state, and a decrease in the frequency response ratio guarantee index was observed at the loading 1 stage. This is the same trend in Loading 2. From this, it can be seen that by using the frequency response ratio guarantee index, the damage alarm for all the bending stiffness reduction step is made.

Next, the impedance signal was measured from the MFC patch attached to the wedge by Step 2 of FIG. As can be seen in Figure 10, it can be seen that the impedance signal is almost unchanged, it can be classified that the cause of the damage alarm is due to the decrease in the flexural rigidity of the concrete.

Through the above experimental verification, it can be seen that the proposed hybrid monitoring system is useful for monitoring the reduction of tension of PC tendon and the decrease of flexural stiffness of concrete, which are typical damage types of PSC bridges, and the real-time monitoring of actual PSC bridges. It can be seen that it is very practical to construct the system.

1 is a schematic diagram of a hybrid monitoring system for performing damage generation and damage classification by monitoring the reduction in tension and bending rigidity of the PSC girder bridge to be developed through the present invention.

Figure 2 is a shape and reinforcement diagram of the model PSC girder produced to verify the hybrid monitoring system to be developed through the present invention.

Figure 3 shows the position of the sensors to be attached to the model PSC girder fabricated to verify the hybrid monitoring system and the excitation position for the acquisition of the acceleration signal.

4 shows an accelerometer and MFC sensor attached to the top and wedge of a PSC girder to acquire acceleration and impedance signals.

Fig. 5 shows the frequency response function and the frequency response ratio calculated using the acceleration signals obtained from the accelerometers at different positions to alert the occurrence of the PC tendon tension reduction.

FIG. 6 shows the results of obtaining the frequency response ratio guarantee index using the frequency response ratio calculated to alert the occurrence of a decrease in the hepatic force of PC tendon.

FIG. 7 shows the change in the measured impedance signal to identify that the cause of the damage alerted in Step 1 is due to a lowering of tension.

FIG. 8 shows the frequency response function and the frequency response ratio calculated using the acceleration signals obtained from the accelerometers at different positions to alert the occurrence of degradation of concrete flexural stiffness.

9 shows the results of obtaining the frequency response ratio guarantee index by using the calculated frequency response ratio to alert the occurrence of degradation of concrete bending stiffness.

FIG. 10 shows the change in the measured impedance signal to determine that the cause of the damage alerted in Step 1 is due to a decrease in flexural stiffness.

Claims (3)

Hybrid Damage Alarm and Damage Classification System Using Acceleration and Impedance Signal Characteristics Damage warning technique using frequency response guarantee index of claim 1 Wedge with MFC sensor as in Figure 4

Images