석 사 학 위 논 문

Master’s Thesis

유한요소해석 기반 PT 텐던 시스템의

자기변형에 대한 수치적 연구

Numerical Studies of Magnetostriction in Post-tensioning Tendon System based on

Finite Element Analysis

2017

이 준 (李 晙, Lee, Jun)

한 국 과 학 기 술 원

Korea Advanced Institute of Science and Technology

석 사 학 위 논 문

유한요소해석 기반 PT 텐던 시스템의

자기변형에 대한 수치적 연구

2017

이 준

한 국 과 학 기 술 원

건설 및 환경공학과

유한요소해석 기반 PT 텐던 시스템의

자기변형에 대한 수치적 연구

이 준

위 논문은 한국과학기술원 석사학위논문으로

학위논문 심사위원회의 심사를 통과하였음

2016년 12월 28일

심사위원장 손 훈 (인 ) 심 사 위 원 김 진 근 (인 )

심 사 위 원 곽 효 경 (인 )

Numerical Studies of Magnetostriction in Post-tensioning Tendon System based on

Finite Element Analysis

Jun Lee

Advisor: Hoon Sohn

A dissertation/thesis submitted to the faculty of Korea Advanced Institute of Science and Technology in partial fulfillment of the requirements for the degree of

Master of Philosophy in Civil and Environmental Engineering

Daejeon, Korea December 28, 2016

Approved by

Hoon Sohn

Professor of Civil and Environmental Engineering

The study was conducted in accordance with Code of Research Ethics1).

1) Declaration of Ethical Conduct in Research: I, as a graduate student of Korea Advanced Institute of Science and Technology, hereby declare that I have not committed any act that may damage the credibility of my research. This includes, but is not limited to, falsification, thesis written by someone else, distortion of research findings, and plagiarism. I confirm that my dissertation contains honest conclusions based on my own careful research under the guidance of my advisor.

초 록

프리스트레스콘크리트 구조물에서 PT 텐던 긴장력은 구조물의 사용성 및 안전성과 연관 되어있는

가장 중요한 요소 중에 하나다. 와전류 탐상 기법은 텐던 정착부 구조물에 응력 변화가 발생 시

자기변형(Magnetostriction)효과를 통해 자기장 변화가 발생하게 되며, 이를 계측 하는 방법이

다. 본 연구에서는 유한요소 해석 기법을 활용하여 프리스트레스 텐던 구조물의 긴장력 저하 시

텐던 정착부에서 발생되는 자기장 변형에 대하여 3차원으로 묘사할 수 있는 시물레이션 기법을

개발하였다. PT 텐던 정착부의 자기적 성질을 재료 실험단계에서 구하였으며, 3차원 시물레이션

을 통하여 PT 텐던 긴장력 변화에 따른 응력 변화를 확인하였으며, 응력에 따른 자기장 발생을

확인하였다. 3.3 m 강연선을 통하여 긴장력에 따른 자기장 변화에 대해 실험값과 비교를 통하여

검증하였다.

핵 심 낱 말 자기변형, 유한 요소 해석, 와전류 탐상 기법, 프리스트레스콘크리트, 텐던 구조물

Abstract

PT tendon system is important elements in terms of service and safety in prestressed concrete structure. The

tensile force monitoring in PT tendon system by eddy current technique measures the changes of magnetic flux

on the wedge surface under different tensile force by magnetostriction. This thesis presents a numerical analysis

of PT tendon system to estimate the magnetic flux density under different stress condition. The magnetic material

properties are achieved by material properties test using MTS machine, and Three dimensional model of a pre-

stressing strand, a wedge, and a barrel are developed. Variation of stress distribution and magnetic flux density

are observed using commercial numerical simulation program, ABAQUS. Then, laboratory experiments are

carried out to evaluate tensile force loss of a 3.3 m long pre-stressing strand. The monotonic relationship between

tendon force loss and magnetic flux density is successfully observed.

Keywords Magnetostriction, Numerical analysis, Eddy current test method, Prestressed concrete, Posttensioning

tendon system

MCE 20153474

이준. 유한요소해석 기반 PT 텐던 시스템의 자기변형에 대

한 수치적 연구. 건설 및 환경공학과. 2017년. 40+v 쪽. 지도교수: 손훈. (영문 논문)

Jun Lee. Numerical Studies of Magnetostriction in Post-tensioning Tendon System based on Finite Element Analysis. Department of Civil and Environmental Engineering. 2017. 40+v pages. Advisor: Hoon Sohn. (Text in English)

i

Contents

Contents ................................................................................................................................................................. i

List of Figures and Table ................................................................................................................................... iii

Chapter 1. Introduction ....................................................................................................................................... 1

1.1Research background ...................................................................................................................................................... 1

1.2 Literature review of other PT tendon monitoring techniques .............................................................................. 3

1.3 Objective and uniqueness ............................................................................................................................................. 5

Chapter 2. Theoretical Background .................................................................................................................. 6

2.1 Magnetostriction ............................................................................................................................................................. 6

2.1.1 Basic concept of magnetostriction ................................................................................................. 6

2.1.2 Direct effect and inverse effect ...................................................................................................... 7

2.1.3 Piezo-magnetic material properties equation specification ............................................................ 8

2.1.4 Relationship between piezo-electricity and piezo-magnetism ....................................................... 8

2.2 Eddy current technique ............................................................................................................................................... 10

Chapter 3. Numerical Simulation of Magnetostriction on PT Tendon System ............................................. 11

3.1 Experimental Estimation of Magnetoelastic material properties ............................................................ 11

3.1.1 Magnetic field – Magnetic flux curve test by SQUID-VSM ....................................................... 11

3.1.2 Magnetic field – Magnetostriction curve under different stress condition by using MTS machine .............................................................................................................................................................. 15

3.2 Numerical simulation setup .................................................................................................................... 19

3.2.1 Geometry, loading and boundary condition in first step .............................................................. 19

3.2.2 Geometry, loading and boundary condition in second step ......................................................... 21

3.3 Numerical simulation result .................................................................................................................... 23

3.3.1 Energy history of numerical analysis ........................................................................................... 23

ii

3.3.2 Stress concentration in wedge surface ......................................................................................... 24

3.3.3 Magnetic flux difference at difference loading conditions........................................................... 26

Chapter 4. Lab Scale Experimental Validation Test ........................................................................................ 28

4.1 Experimental set up ..................................................................................................................................................... 28

4.2 Experimental result...................................................................................................................................................... 30

Chapter 5. Summary and Conclusion ............................................................................................................... 32

Bibliography ........................................................................................................................................................ 33

Acknowledgments in Korean ............................................................................................................................. 36

Curriculum Vitae ................................................................................................................................................ 38

iii

List of Figures and Table

Figure 1.1 Compensating a low tensile strength by prestressing with longitudinal compression ................ 2

Figure 1.2 A schematic overview of post-tensioning (PT) construction, which applies compression force to

the individual concrete segments using PT tendons ........................................................................................... 2

Figure 1.3 Previous researches for PT tendon tensile force estimation technique: (a) Electromagnetic

technique (Zhao et al., 2008), (b) Optical fiber with distributed Fiber Bragg Grating technique (Kim et al., 2012),

(c) Mechanical Strain Technique (Abdullah et al., 2015), (d) Ultrasonic technique (Bartoli et al., 2011) and (e)

Eddy current technique (Schoenekess et al., 2007) .......................................................................................... 3

Figure 2.1 The concept of magnetostriction, (a)Disordered regime(over the curie point), (b) Ferromagnetic

regime (demagnetized), (c) Ferromagnetic regime (magnetized to saturation) ............................................. 6

Figure 2.2 The effect of magnetostriction (a) Joule effect, (b) Villari effect ................................................. 7

Figure 2.3 Basic principles of eddy current technique ................................................................................. 10

Figure 3.1 A Josephson junction device, which consists of a superconductor with a poorly conducting weak link . 11

Figure 3.2 Magnetic field – Magnetic moment curve under different temperature ..................................... 12

Figure 3.3 Magnetic field – Magnetic flux density curve under different temperature ............................... 12

Figure 3.4 The result of magnetic flux density at saturation point under different temperature .................. 14

Figure 3.5 The specification of magnetic effect test specimen under compressive and tensile stress ......... 15

Figure 3.6 Experimental setup of magnetic effect test under different stress condition .............................. 15

Figure 3.7 The result of magnetic flux density under difference stress condition ....................................... 16

iv

Figure 3.8 Magnetic field – Magnetic flux density curve under different stress condition ......................... 18

Figure 3.9 The relationship between magnetic flux and stress under different stress condition .................. 18

Figure 3.10 Specification of numerical analysis model (a)Dimension of model, (b) Loading and boundary

condition ......................................................................................................................................................... 20

Figure 3.11 Loading and boundary condition of numerical analysis model ................................................ 20

Figure 3.12 Numerical analysis model at second step applied pre-stressing at the end of tendon 180 kN .. 22

Figure 3.13 The strain energy of numerical analysis ................................................................................... 23

Figure 3.14 Stress concentration result of PT tendon system (a) 180 kN, (b) 120 kN, (c) 60 kN ............... 24

Figure 3.15 The stress distribution at wedge surface under tensile force loss condition ............................. 25

Figure 3.16 Magnetic flux result of PT tendon system (a) 180 kN, (b) 120 kN, (c) 60 kN ......................... 26

Figure 3.17 The Magnetic flux distribution at wedge surface under tensile force loss condition ................ 27

Figure 4.1 Experimental setup to measure magnetic flux at PT tendon system using eddy current sensor . 29

Figure 4.2 Eddy current sensor specification;(a) diameter of eddy current sensor, (b) actual sensor .......... 29

Figure 4.3 The average eddy current measurement, at 60 kN, 120 kN, and 180 kN ................................... 30

Figure 4.4 The comparing magnetic flux result between experiment and numerical analysis..................... 31

Table 2.1 Comparison of piezomagnetic and piezoelectric governing equations .......................................... 8

Table 2.2 Comparison of piezomagnetic and piezoelectric quantities ........................................................... 8

v

Table 2.3 Comparison of interface conditions for piezomagnetism and piezoelectricity .............................. 9

Table 3.1 Magnetic permeability of each specimen ..................................................................................... 13

Table 3.2 Piezomagnetic coefficient of each specimen................................................................................ 17

Table 3.3 Mechanical properties of numerical analysis ............................................................................... 21

Table 3.4 The mesh properties of numerical analysis .................................................................................. 21

Table 4.1 Experimental setup parameters value ........................................................................................... 30

1

Chapter 1. Introduction

1.1 Research Background Lately, Prestressed Concrete (PsC) structures have been constructed widely as alternative structures for

reinforced concrete structures since 1980s, because the introduction of prestressing to a reinforced concrete girder

overcomes its disadvantages such as limited span length and cracking under service conditions. As shown in

Figure 1.1, PsC is structural concrete in which internal stresses have been introduced to reduce potential tensile

stress in the concrete resulting from loads. The fundamental and advantages of prestressing can be summarized as

the possibility to limit cracking, deformations in structural members with large span, and to increase the load

capacity for given span and dimension.[1] In the prestressed concrete system is separated two method, Pre-

Tensioning system and Post-Tensioning. The Post-Tensioning (PT) is stressed after hardening of the concrete. The

anchorages are fixed into the concrete, and without need for external anchorages the method is suitable for in-situ

construction. This allows larger tendons than in pre-stressing, since anchorage no longer depends on bond. Also,

PT construction has been broadly used to assemble pre-fabricated concrete bridge using a bundle of pre-stressing

strands which is also called as PT tendon. Figure 1.2, shows a schematic overview of PT construction which is

consist of pre-fabricated concrete bridge segments.

The structure will have an enhanced resistance to share and torsion due to compressive stresses. Also,

prestressed tensile force at tendon will reduce deflections under service loading conditions, due to both the reduced

external load and the increased stiffness effects caused by delayed or eliminated cracking. Therefore, prestressed

tensile force at tendons are critical structural parameters in PsC and also tendon system is critical structural

members. If the prestressing tensile force is reduced below predicted values, then cracking of the concrete and

excessive deflection or even collapse of a bridge can occur. The prestressed tensile force carried in these tendons

can be reduced by fracture due to corrosion, failure or slip at anchorages, excessive elastic losses, concrete creep,

and elastic shortening of the prestressed concrete section.[2]

So, PsC structure engineers have an interest in monitoring tendon force. However, currently there in no

practical solution yet. There are some difficulties. At first, Access to tendon system is very limited because tendon

system is embedded in the concrete. Secondly, to apply prestress at concrete structure, tendon is installed across

concrete structure.[1]

2

Figure 1.1 Compensating a low tensile strength by prestressing with longitudinal compression

Figure 1.2 A schematic overview of post-tensioning construction, which applies compression force to the individual concrete segments using PT tendons

3

1.2 Literature reviews of other PT tendon monitoring techniques

Figure 1.3 Previous researches for PT tendon tensile force estimation technique: (a) Electromagnetic technique (Zhao et al., 2008), (b) Optical fiber with distributed Fiber Bragg Grating technique (Kim et al., 2012), (c) Mechanical Strain Technique (Abdullah et al., 2015), (d) Ultrasonic technique (Bartoli et al., 2011) and (e) Eddy current technique (Schoenekess et al., 2007)

To monitor the tensile force loss of tendon, various approaches using electromagnetic sensor, optical

fiber, strain gauge, ultrasonic guided wave or eddy current have been studied.

Magnetoelastic sensor has been proved that EM measurement method is able to precise tensile force

quantification of actual stress in steel wire, Prestressed concrete bar and tendon precisely without destroying their

polyethylene covering sheath.[3-8] Ming L. Wang (2006) discussed the calibration and workability of magneto-

elastic stress monitoring sensors for large steel cables used in actual bridge at China.[4] However, for

magnetization of a tendon, the size of EM sensor should be larger than the size of anchor head. because of high

power consumption in the range of a few hundred Watts, its installation and maintenance can be challenging.

Particularly, its applicability to the PT tendon embedded inside concrete is limited because of its bulkiness.

Yoshiyuki Sakairi (2008) developed a Brillouin optical time-domain reflectometry prototype system,

4

for monitoring strain and temperature along fiber with a sub- meter spatial resolution.[9] Jae-Min Kim (2012)

investigated a novel method for prestressed tensile forces of prestressing tendon by using the fiber Bragg grating

sensor system. The straight king wire of the 7-wire prestressing tendon is replaced with an instrumented steel tube

in which the FBG sensor is embedded.[10] However, its installation can be extremely difficult and expensive

because a hole should be drilled inside the center steel wire of the tendon and the optical fiber needs to be inserted

into the hole.

Additionally, strain gauge[11], peizoceramic[12], and Eddy Current Sensors (ECS)[13] with low costs

and low power consumption have been explored to evaluate tensile force in tendon system.

Abdullah et al. (2015) attached strain gauges at anchor heads, and related local strain measuring the

tendon force. This technique was tested only for the breakage of several tendons, therefore its sensitivity to initial

tendon force reduction was not reported.[12]

D.G. Aggelis (2010) investigated tensile stress measuring method from the anchor head by using

ultrasonic techniques. [11] Claudio Nucera (2011) used of ultrasonic nonlinearity as a means to determine the

level of load applied to the tendons. [14] Jinyoung Min (2012) monitored relaxation of tendon with a damage

index which defined the changes of impedance signal at specific frequency band.[15] However, ultrasonic waves

can’t reach the receiver which were applied opposite side of tendon.[16] Also, if ultrasonic generator is attached

on the anchor head for measuring the condition of wedge and anchor, single tendon’s status cannot be measured.

So using the ultrasonic wave is of difficulties to measure the tensile stress of tendon.

Ricken et al. (2006) directly installed an eddy current coil onto the surface of a steel tendon, producing

magnetic flux passing through the tendon.[13; 17; 18] Then, they investigated the relationship between the change

of magnetic flux density and the tendon force. In fact, however, the direct installation of the eddy current coil

inevitably brings about coil breakage and malfunction problems due to a potential collision with adjacent tendons

in a multi-tendon system.

5

1.3 Objective, uniqueness and thesis organization This study is motivated by the need to develop appropriate numerical model for accurately predicting

magneto-elastic effect in a PT tendon system. The objectives are summarized as follows;

l To develop schemes and methods for numerical simulations of Magnetostriction in PT tendon

system

l To figure magnetic flux distribution in PT tendon system using in numerical simulation

Chapter 2 deals with the theories which are basis for understanding the Magnetostriction concept in

subsequent chapters. The governing equation of motion in the magneto-elastic viewpoint is reviewed. In Chapter

3, to figure magnetic material properties at numerical analysis, the material test is occurred. Using previous result,

a detailed description on numerical analysis for Magnetostriction is introduced. Chapter 4 demonstrates

experimental validation of the Magnetostriction in laboratory scale test. Finally, Chapter 5 summarized the thesis

and draws conclusions. Suggestions are given regarding future works of finite element analysis for prediction of

magnetic field under different stress.

6

Chapter 2. Theoretical Background

2.1 Magnetostriction

2.1.1 Basic concept of magnetostriction

Magnetostriction is a coupling phenomenon involving a magnetization process and dimension/shape

change in ferromagnetic materials. [19]

λ = (2.1) When a ferromagnetic material is cooled until Curie point which had completely random alignment

above the Curie point because of the disordered magnetic moments, become ordered over volumes containing

large numbers of atoms. These volumes in which all moments lie parallel are called domains and can be observed

under a microscope. In this condition, the bulk magnetization is zero. As shown in Figure 2.1 (a), when the

material becomes ferromagnetic at the Curie points, spontaneous magnetization appears within the domains and

with it an associated spontaneous strain e or magnetostriction λ, along a particular direction, as shown in Figure 2.1 (b). [20]

Within each isotropic domain, the strain varies with angle θ from the direction of spontaneous

magnetization according to the following relation in equation 2.2. () = cos (2.2)

Figure 2.1 The concept of magnetostriction, (a)Disordered regime(over the curie point), (b) Ferromagnetic regime (demagnetized), (c) Ferromagnetic regime (magnetized to saturation)[20]

7

The average deformation throughout the solid due to the onset of spontaneous magnetostriction ca then

be obtained by integration, assuming that the domains are oriented at random, so that any particular direction is

equally likely.

= cos sin ⁄ = 3 (2.3)

2.1.2 Direct effect and inverse effect

(a)

(b)

Figure 2.2 The effect of magnetostriction (a) Direct effect, (b) Inverse effect As illustrated in Figure 2.2 (a), when a magnetic field is applied, the direction elongates into an ellipsoid

with the symmetry axis along the direction of the applied magnetic field. The size-changing effect is called the

“Joule effect”.[21] Joule effect is the direct effect of magnetostriction. There is a limit to this induced strain, which

is known as the saturation magnetostriction. The direct effect is very well documented in the literature. Much

research has been conducted to optimize the properties of magnetostriction material, such as Terfenol-D or

Galfenol, for magnetostrictive transducers and actuators. Most of this transducer research uses one-dimensional

piezomagnetic theory.[22-25]

On the other hand, the “Villari effect” which is called inverse effect of magnetostriction that is illustrated

in Figure 2.2 (b) refers to the reverse phenomenon such that if there is any change in the size of a ferromagnetic

8

material, the material induces a magnetic field.[26] There is significantly less published work on the inverse effect

than the direct effect. The Villari effect in ferrites has been proposed as a possible means of measuring stresses

[27]. Steel and nickel samples were externally stressed, yielding changes in the magnetic field measured with a

Permalloy magnetoresistor [28; 29].

2.1.3 Piezo-magnetic material properties equation specification

In general, magnetostriction is a nonlinear effect, but it can be treated as a linear effect over a certain

range of operation.[30] The linear magnetostrictive effect is called Piezomagnetism and is described by the

equations, = + = + (2.4)

2.1.4 Relationship between piezo-electricity and piezo-magnetism

As, shown in Table 2.1, the governing equations for piezomagnetism and piezoelectricity are similar

each other. The important quantities in piezomagnetism are magnetic flux density (B) and magnetic field (H).

These are corresponding to electric displacement (D) and electric field (E) in piezoelectricity, respectively.[31]

To complete the analogy, a comparison of all the pertinent quantities is shown in Table 2.2.[31; 32]

Table 2.1 Comparison of piezomagnetic and piezoelectric governing equations

Piezo magnetism Piezoelectricity = + = + = + = +

Table 2.2 Comparison of piezomagnetic and piezoelectric quantities

Piezo magnetism Piezoelectricity

Quantity Symbol Quantity Symbol

Stress Strain Magnetic flux density Magnetic field Elastic compliance (H=constant) Piezomagnetic constant Permeability (T=constant) Magnetic scalar potential

T S B H sH d μT ϕm

Stress Strain

Electric displacement Electric field

Elastic compliance (E=constant) Piezoelectric constant

Permittivity (T=constant) Electric potential

T S D E sE d εT ϕe

9

Along with the piezomagnetic governing equations, boundary conditions are necessary to solve a

problem. [33]

As same in piezoelectric governing equations, which hold for magneto-static problems when there are

no free currents.

These analogous sets of equations lead to important continuity conditions at the interface of two

materials.[33]

∇ ∙ B = 0 (2.5.a) ∇ × H = 0 (2.5.b) ∇ ∙ D = 0 (2.6.a) ∇ × E = 0 (2.6.b)

10

2.2 Eddy current technique Eddy current inspection is one of several non-destructive inspection testing methods that use the

principal of inverse effect of magnetostriction which is consist of electromagnetism. Other methods taking use of

this principle include remote field testing, flux leakage and barkhausen noise, etc.

Eddy currents are created by a process called electromagnetic induction. When alternating current is

applied to an eddy current sensor, which is winded by copper wire, a magnetic field forms in and around the eddy

current sensor. This magnetic field expands as the alternating current rises to maximum and collapses as the

current decreases to zero. If another electrical conductor is brought into this changing magnetic field, current will

be induced in the second conductor. Eddy currents are induced electrical currents that flow in a circular path. They

get their name from Eddies that are formed when a liquid or gas flows in a circular path around obstacles when

conditions are right.

Figure. 2.3 shows the basic working principles of eddy current technique. [34]

At first, the alternating current flowing through the eddy current sensor at an excitation frequency generates a

magnetic field around the coil. Secondly, when coil is placed close to target structure, which is consisted of an

electrically conductive material, such as steel, nickel, and cobalt, eddy current is induced in the target structure.

If a flaw in the target structure disturbs the eddy current circulation, the magnetic coupling with the sensing part

of eddy current sensor is changed and a defect signal can be read by measuring the coil impedance or signal

variation.

Figure 2.3 Basic principles of eddy current technique

11

Chapter 3. Numerical Simulation of Magnetostriction

on PT Tendon System

3.1 Experimental Estimation of Magnetoelastic material properties Three types of steel in PT tendon system, which are anchor head, wedge, and tendon, are ferromagnetic

materials. However, the actual magnetic properties such as magnetic hysteresis loop, magnetic permeability, and

magnetoelastic coefficient, etc. To solve magnetoelastic analysis, some properties have to achieve by the material

test.

3.1.1 Magnetic field – Magnetic flux curve test by SQUID-VSM

To measure delicate magnetic hysteresis loop, The Superconducting Quantum Interference Device –

Vibrating Sample Magnetometer(SQUID-VSM) is used in Korea basic science institute. SQUID-VSM provide

the ultimate in resolution for field measurements. The SQUID-VSM consists of a superconducting ring with a

small insulating layer known as the weak link, as shown in Figure 3.1. The weak link is also known as the

Josephson junction. The resolution of these devices is 10-14 T. The flux passing through the ring is quantized once

the ring has gone superconducting but the weak link enables the flux trapped in the ring to change by discrete

amounts.[20]

Specimens are three kind of steel from each part of PT tendon system. The specimen size is Cubic shape

(1 mm x 1 mm x 1 mm) because to saturate magnetization of a ferromagnet. Before measuring magnetic flux by

Figure 3.1 A Josephson junction device, which consists of a superconductor with a poorly conducting weak link

12

Figure 3.2 Magnetic field – Magnetic moment curve under different temperature

Figure 3.3 Magnetic field – Magnetic flux density curve under different temperature

13

magnetic field, specimen is magnetized at saturation point. Also, to find the tendency of magnetic field under

different temperature condition is measured. In PsC structures, the temperature difference is occurred -40 ℃ to

40 ℃. So temperature difference is consisted 5 steps from -40 ℃ to 40 ℃, increasing 20 ℃.

Figure 3.2 is the result of Magnetic hysteresis loop (Magnetic field strength, H – Magnetic moment, M)

under different temperature condition. All of specimen is very similar magnetic characteristic. The area of

hysteresis loop is very narrow so that a small amount of dissipated energy in repeatedly reversing the

magnetization. By using H and M, Magnetic flux density, B is calculated by equation 3.1. B = ( + ) (3.1) where, (= 4 × 10) is the magnetic permeability of free space, which is a universal constant.

Figure 3.3, is the result of Magnetic hysteresis loop (Magnetic field strength, H – Magnet flux density,

B) under different temperature condition. In this loop, magnetic permeability is calculated by equation 3.2.

= (3.2) Magnetic permeability of each specimen is in Table 3.1.

As shown in Figure. 3.4, The result of magnetic flux density at saturation point under different

temperature. The linear relationship is showed between temperature and saturation point of magnetic flux density.

However, the change of magnetic flux density under different temperature is too small comparing whole magnetic

flux density, so that the change caused by temperature is not considered at PT tendon system.

Table 3.1 Magnetic permeability of each specimen

Anchor head Wedge Tendon

Magnetic permeability [Hm-1] 6.6353*10-6 7.5140*10-6 6.1827*10-6

14

Figure 3.4 The result of magnetic flux density at saturation point under different temperature

15

3.1.2 Magnetic field – Magnetostriction curve under different stress condition by using

MTS machine

All of the specimens were tested in uniaxial compression and tension on an Instron 8801. The size of

specimen is shown in Figure 3.5 and Figure 3.6 shows experimental setup of magnetic effect test under different

stress condition. The step of stress at each specimen is set 7 steps, from 300 MPa tensile condition to 300 MPa

Figure 3.5 The specification of magnetic effect test specimen under compressive and tensile stress

Figure 3.6 Experimental setup of magnetic effect test under different stress condition

16

compressive condition, with 100 MPa interval. The power supply (Agilent N7951A) is used for generating

Magnetic field around specimen. The power supply can apply maximum 1000 W. the voltage is up to 20 V and

current is 50A limitation. The Coil is designed around the specimen. 1.5 mm diameter copper wire is 370 turns in

the coil. It can generate 0.2 MA/m magnetic field at 40 A condition. The magnetic response was measured with a

Hall probe and Tesla-meter (TM-801, Kanetec). The Teslar-meter measure the signal from the Hall probe and

output magnetic flux reading. The resolution of hall probe is 0.00001 T (=0.1 G) under the 1.5 T, and after 1.5 T,

the resolution is set 0.001 T. The sampling frequency of Teslarmeter is set 10 Hz. The limitation of the setup is

that the hall probe can measure only one direction of the magnetic flux at a time. The direction of stress is defined

3, and the measured direction of magnetic flux density is defined 1 and 2 as shown in Figure 3.6. Therefore, the

test is occurred two times in same stress condition.

The result of magnetic flux under different stress condition is shown in Figure 3.7. The location of hall

sensor is vertical direction of specimen, so that the stress direction and the measured magnetic flux density

direction is orthogonal. When increasing stress, the magnetic flux is gradually decreased. The ratio between stress

and magnetic flux density is piezomagnetic coefficient. The piezomagnetic coefficient can be simplified in the

isotropic material as shown in equation 3.3.[35] By using this equation, results of piezomagnetic coefficient each

specimen are shown in Table 3.2.

Figure 3.7 The result of magnetic flux density under difference stress condition

17

= =

⎣⎢⎢⎢⎢⎡ −

− − − − − ⎦⎥⎥⎥⎥⎤ ( ) (3.3)

Table 3.2 Piezomagnetic coefficient of each specimen

Anchor head Wedge Tendon

Piezomagnetic coefficient [m/A] ( = , = , , , = , , ) 1.35e-12 1.86e-12 2.54e-12 Piezomagnetic coefficient [m/A] ( ≠ , = , , , = , , ) -6.75e-13 -9.30e-13 -1.27e-12

The relationship between magnetic field and magnetic flux density under different stress condition is

shown in Figure 3.8. H-B curve is similar to result of SQUID-VSM test. The anchor head specimen’s ratio is

different comparing to the wedge and tendon specimen, because the hysteresis energy which is indicated by the

area of hysteresis loop is smaller than other specimen. As same as temperature condition, the change magnetic

flux density because of stress condition is not significant. However, when the stress increase, the magnetic flux

density is also increased in the same magnetic field, as shown in Figure 3.9. Using equation 3.2, the magnetic

permeability is also calculated. In this result, the magnetic permeability is decrease when the compressive stress

is applied.

18

Figure 3.8 Magnetic field – Magnetic flux density curve under different stress condition

Figure 3.9 The relationship between magnetic flux and stress under different stress condition

19

3.2 Numerical simulation setup As mention in chapter 2, Villari effect is not modeled with any of finite element analysis(FEM) program

which are used commercially to solve problems in engineering. By using the strong analogy between

piezoelectricity and piezo-magnetism, Inverse effect can be calculated using piezoelectric analysis. In ABAQUS

2016, has piezoelectric elements and piezoelectric analyses.

Piezoelectric analysis can be used in linear analysis. So the mechanical behavior of the material can

include linear elasticity only. However, part of wedge and tendon in PT tendon system are stressed and deformed

over the yield point of material property. For applying actual condition of PT tendon system under the pre-

stressing, simulation has to process two step. In the first step, the stress and deformation are analyzed by dynamic,

implicit procedure. And then, the piezoelectric analysis is performed using the result of the result of the first step.

3.2.1 Geometry, loading and boundary condition in first step

For simulating an actual PT tendon system, a three dimensional model is structured by Auto-Cad 2017.

A wedge is modeled with three-pieces and a tendon modeled a seven steel bars. As shown in Figure 3.9, size of

tendon is set Φ 15 mm × 150 mm (each steel bar is Φ 5 mm × 150 mm) and an anchor head is set Φ 50 mm × 60

mm. The detailed mechanical properties of an anchor head, a wedge, and a tendon are decided in Table 3.3.

To minimize the friction force between anchor head and wedge, grease is spread at inside of anchor

head in actual PT tendon system, and inner surface of wedge has threads to strongly grab tendon. Contact condition

between wedge and anchor head is set frictionless condition, between wedge and tendon is set rough condition.

And the general contact condition is set at every model if the contact is occurred because of deformation. To avoid

surface penetration between surfaces in contact surface, the contact pressure-overclosure relationship is defined.

In this condition, the surfaces separate if the contact pressure reduces to zero. Separated surfaces come into contact

when the clearance between them reduces to zero.

Figure 3.10 is the loading and boundary condition of PT tendon system. The maximum tensile load at

end of tendon model is applied 180 kN which is same as maximum load in actual PT tendon system. By using

load control method, the velocity of model is very important so that the loading curve is made smoothly using

smooth condition, as shown in Figure 3.11. Compressive stress is applied wedge outer surface so that the tendon

is grabbed by wedge. The outer surface of anchor head is fixed all direction.

20

(a)

(b) Figure 3.10 Specification of numerical analysis model (a)Dimension of model, (b) Loading and boundary condition

Figure 3.11 Loading and boundary condition of numerical analysis model

21

Table 3.3 Mechanical properties of numerical analysis

Anchor head Wedge Tendon

Density [ton/mm3] 7.85E-09

young's modulus [MPa] 210000 200000 185000

Poisson’s ratio 0.3 0.4 0.23

yield stress [MPa] 476 392 1860

tensile stress [MPa] 657 628 2140

plastic elongation 0.2 0.17 0.055

The Element type of PT tendon system is defined 3D stress mesh. C3D8R mesh is for 3D stress element

type which is consists of linear brick, and have 8-node in one element. In case of full integration, all the stiffness

coefficients of an un-distorted elements can be exactly integrated, but the computing cost is too high comparing

the reduced integration method. The problem with reduced integration elements is hour-glass mode. For

reasonable numerical analysis, the hourglass control is set 0.5 (the default value at C3D8R). Size of element is set

according to characteristic of parts, as shown in Table 3.4. To figure the stress concentration and magnetic flux

density at surface of wedge, the size of wedge and tendon element is smaller than anchor head. The anchor head

is stiffest structure compare to others, so that the anchor head is set largest value of element size.

3.2.2 Geometry, loading and boundary condition in second step

After the first step of numerical analysis using dynamic analysis, piezo-magnetic effect analysis is progressed.

Analysis model in second step is applied prestressed model which is the result of first step, as shown in Figure

3.11. The deformation occurred at first step is in this model at each tensile force. The stress result is applied by

predefined fields of each tensile force. The boundary condition is fixed boundary at the outer surface of Anchor

head as same as previous step model. In the piezoelectric case, the electrical potential would be set to zero along

Table 3.4 The mesh properties of numerical analysis

Anchor head Wedge Tendon

Element type – first step C3D8R

Element type – second step C3D20RE

Element size [mm] 1 0.5 0.5

Number of element 9,720 13,800 34,500

22

the top and bottom of the wedge and anchor head. The analogous piezomagnetic boundary condition is to set the

magnetic potential equal to zero on top and bottom edges in wedge and anchor head.

The element type of model is set C3D20RE. In this element type can applied to three-dimensional model,

and consisted of 20-nodes. Increasing the number of node is more effective method to increase accuracy and to

decrease computing cost of numerical analysis result than decreasing mesh size. After the number of nodes, R is

defined reduced integration, and the E is defined piezoelectric material calculation element.

Material properties of piezoelectric analysis have to several assumptions is demand to complete the

model. At first, the assumption of isotropic elasticity is assumed because of the piezoelectric element can solve

the elastic material. The mechanical properties except yield stress, tensile stress, and plastic elongation is applied

as same value of Table 3.2. used in previous simulation. For the piezomagnetic coefficient, the stress coefficients

are used. The value of piezomagnetic coefficient is applied about the result from chapter 3.1.2. The value of

magnetic permeability is used from chapter 3.1.1.

Figure 3.12 Numerical analysis model at second step applied pre-stressing at the end of tendon 180 kN

23

3.3 Numerical simulation result

3.3.1 Energy history of numerical analysis

Figure 3.13 is shown the energy history of numerical analysis. The total energy indicates the

reasonability of numerical analysis result by equation 3.4.

where, ETOTOAL is total energy in numerical analysis, ALLIE is total internal energy, ALLKE is kinematic

energy, ALLVD is viscous dissipation energy, ALLFD is frictional dissipation energy, and ALLWK is external

work. The kinetic energy is increased before contact between wedge and anchor head. The strain energy and

internal energy are gradually increasing up to maximum tendon loading, and then slowly decrease until tensile

force is reached zero.

The total internal energy is calculated by equation 3.5. = + + + , (3.5) where, ALLSE is recoverable strain energy, ALLPD is plastic dissipation energy, ALLCD is energy dissipated

by creep, viscoelasticity and swelling, and ALLAE is artificial strain energy. In the artificial energy is increased

after contact between wedge and anchor head.

= + + + – = (3.4)

Figure 3.13 the strain energy of numerical analysis

24

3.3.2 Stress concentration in wedge surface

(a)

(b)

(c)

Figure 3.14 stress concentration result of PT tendon system (a) 180 kN, (b) 120 kN, (c) 60 kN

25

Figure 3.14 is shown the stress concentration of numerical analysis result of PT tendon system. The

stress concentration is occurred at contact point between wedge and tendon. After maximum load, some element

of wedge connected tendon or anchor head are excessed the yield stress, so that the plastic behavior is shown in

that elements under tensile force loss condition at tendon. When the tensile force is decreased, the stress

concentration of wedge surface is also reduced.

Figure 3.15 is the stress distribution at wedge surface. Two paths are set in the wedge surface. Start of

paths is inner side of wedge elements which are closed to tendon element. The first path (p1) is middle of wedge

surface, which is directly connected with tendon elements. The second path (p2) is set edge of wedge surface. in

this path, the connection between tendon and wedge is occurred at the maximum tensile stress condition at tendon.

Comparing two paths, the elements in p1 are stressed high in the maximum tensile force. But in the p2, the stress

is decreased in proportion to the displacement under the maximum tensile force. In the maximum tensile force,

the whole elements in p1 and inner side of p2 is exceed the yield stress, these elements applied plastic behavior.

When tensile force loss is occurred, the stress distribution is changed significantly at outside of p1 and inside of

p2. In these elements, the contact is week because of the plastic behavior.

Figure 3.15 the stress distribution at wedge surface under tensile force loss condition

26

3.3.3 Magnetic flux difference at difference loading conditions

(a)

(b)

(c)

Figure 3.16 Magnetic flux result of PT tendon system (a) 180 kN, (b) 120 kN, (c) 60 kN

27

Figure 3.16 is shown the magnetic flux density of numerical analysis result of PT tendon system. The

magnetic flux density is inverse proportional to the stress. Therefore, the high stress concentration area at the

interface between wedge and tendon is lowest in the surface of PT tendon system. Similar tendency between

magnetic flux density and stress distribution, after maximum load at end of tendon, magnetic flux density is

gradually increased when the tensile force loosed.

Figure 3.17 is the magnetic flux density at the two paths in the wedge surface. Start of paths is inner

side of wedge elements which are closed to tendon element. The first path (p1) is middle of wedge surface, which

is directly connected with tendon elements. The second path (p2) is set edge of wedge surface. in this path, the

connection between tendon and wedge is occurred at the maximum tensile stress condition at tendon. The

magnetic flux density is gradually decreased paths in tensile force loss condition. The tendency of magnetic flux

density is similar to each paths. But the variation of magnetic flux density at p1 is higher than p2, because the

stress variation of p1 is larger p2.

Figure 3.17 the Magnetic flux distribution at wedge surface under tensile force loss condition

28

Chapter 4. Lab Scale Experimental validation

4.1 Experimental setup The experimental setup to measuring magnetic flux under tensile force loss condition in PT tendon

system is described in Figure 4.1. A mono-tendon was inserted into a steel frame, and the tension is applied to the

tendon using a custom-designed universal testing machine (UTM). The UTM consists of a 2400 × 220 × 220 mm3

steel frame, a hydraulic actuator and a load cell. The maximum load capacity of the hydraulic actuator is 250 kN,

and the rated measurement range of the load cell is from 0 kN to 300 kN with 0.1 kN load resolution. A user can

easily set a desired force level in the control unit and the hydraulic actuator accordingly adjusts the tension force

of the tendon to the prescribed level. The tension force measured by the load cell was considered the ground-truth

in this experiment.

A 3.3 m-long, 15.2 mm-diameter and normal-relaxation tendon was used as a test specimen. A wedge

and an anchor head (KTA-MA-Type produced by Korea Total Anchorage Inc.) were installed at both ends of the

tendon, and its allowable strength was 1,860 MPa. When the wedge and the anchor head were fully combined, its

overall size was Φ 45 mm × 60 mm. Eddy current sensor is designed to measure the magnetic flux density at

wedge surface of PT tendon system. Eddy current sensor is composed of two separate coils, 20 turn coil, excitation

coil (100 turn coil) and sensing coil (20 turn coil). Each coil is manufactured using a copper wire of 0.1 mm-

diameter. Diameter of coil shown in Figure 4.2 (a). The sensing coil is located inside of the driving coil, and the

overall size of the eddy current sensor is less than 3 mm and 1.3 mm-thick. The electric resistance and inductance

of the driving and sensing coils are measured to be 3.5 Ω, 1.0 Ω, 44.01 μH and 1.77 μH, respectively. And arbitrary

waveform generator (AWG) and a digitizer is used exciting and measuring eddy current sensor. Slot-type

commercial products, NI PXI-5421, NI PXI-5122 and NI PXI-8119, from National Instruments Inc. are used for

the AWG, the digitizer, and the control unit. The AWG has 16-bit resolution and 100 MHz sampling frequency

for digital-to-analog conversion and its maximum output voltage is ±6 V. The digitizer can sample eddy current

data with 14-bit resolution and 100 MHz sampling frequency. The control unit contains a 2.3 GHz processor and

a 4 GB RAM for post-processing of the measured data.

The specific parameters value of experimental setup is shown in Table 4.1, the tension force of the

tendon was increased up to 180 kN initially, and gradually reduced to 30 kN with 30 kN interval. At each force

29

Figure 4.1 Experimental setup to measure magnetic flux at PT tendon system using eddy current sensor

(a) (b)

Figure 4.2 Eddy current sensor specification;(a) diameter of eddy current sensor, (b) actual sensor

30

Table 4.1 Experimental setup parameters value

Tensile force steps 180 kN ~ 30 kN (30kN decrease, 6steps)

Strand Type Φ15.2 mm ⅹ 3200 mm

Resistance of Sensor (Drive – Sensing) 0.60 Ω – 3.15 Ω Inductance of Coil (Drive-Sensing) 1.77 μH – 44.01 μH

Frequency 100 ~ 2,000 kHz

Input Volt. -2 V ~ 2 V

Sampling frequency 20 MHz

DAQ Time 0.005 sec

Num. of measurements 10

level, magnetic flux is measured. The AWG applied a modulated linear chirp signal varying from 100 kHz to

2,000 kHz to the eddy current sensor in 1 msec, and its amplitude was set to ±2V. Data Sampling frequency set

20 MHz in DAQ.

4.2 Experimental result Figure 4.3 shows the average eddy current measurement of time response. When the tensile force is

decreased, the amplitude of average eddy current measurement is gradually increased.

Figure 4.3 the average eddy current measurement, at 60 kN, 120 kN, and 180 kN

31

From the output voltage result of eddy current, magnetic flux is calculated by equation 4.1.

= = ∗ √2 ∗ (4.1) where denotes the length of coil, N denotes the turns of coil, I denotes the current, and is the magnetic permeability of target structure. The direct voltage amplitude is same as alternating voltage divided by square root

2. By using the magnetic flux density of sensing eddy current and generating magnetic field at excitation coil, the

magnetic flux of changed by stress is calculated by equation 4.2. = + (4.2) Figure 4.4 is the result of magnetic flux at experimental test and numerical analysis. When the tensile

force is reduced, the magnetic flux is gradually increased both result. In the experimental result, the ratio of

magnetic flux change is reduced in low stress. However, in the numerical analysis, the relationship between

magnetic flux and load is monotonic. In the numerical simulation cannot be applied the frequency of input

magnetic flux.

Figure 4.4 the comparing magnetic flux result between experiment and numerical analysis

32

Chapter 5. Summary and Conclusion

In this thesis, a numerical analysis of PT tendon system is presented to predict the magnetic flux density

under the different stress condition. Since the inverse effect of magnetostriction is not modeled with any of

commercial numerical analysis program, Therefore, to simulate the inverse effect of magnetostriction, the

piezoelectric analysis solver is used.

Before numerical analysis, the material test is occurred by MTS machine. When the stress is increased,

the magnetic flux density of same direction is also increased under zero magnetic field condition. By using these

material properties, three dimensional model of a pre-stressing tendon, wedges, and a barrel are developed.

Variation of stress distribution and magnetic flux density are observed using commercial numerical simulation

program, ABAQUS.

To validate the result of numerical analysis, the experimental test is carried out to evaluate tensile force

loss of a 3.3 m long pre-stressing tendon. As similar result between numerical and experimental test, the

monotonic relationship between tendon force loss and magnetic flux density is successfully observed.

Future works will be focused on the magnetic material properties updating to predict more accurate

magnetic properties changes in the PT tendon system. The model of numerical analysis has to update the actual

tendon shape. We expect that the helically wounded shape is affect the tendency of magnetic flux density. Also,

in the model, the magnetic permeability changed by stress have to considered.

33

Bibliography

[1] (2005). fib Bulletin 31: Post-tensioning in Buildings. [2] Washer, G.A., Green, R.E., and Pond, R.B. (2002). "Velocity constants for

ultrasonic stress measurement in prestressing tendons." Research in Nondestructive Evaluation, 14, pp. 81-94. [3] Wang, G. et al. (2005). "Application of EM stress sensors in large steel

cables." Smart Structures and Materials: pp. 395-406.

[4] Wang, G. et al. (2006). "Application of magnetoelastic stress sensors in large

steel cables." Smart Structures and Systems, 2, pp. 155-169. [5] Sumitoro, S. et al. (2001). "Long span bridge health monitoring system in

Japan." 6th Annual International Symposium on NDE for Health Monitoring and

Diagnostics: pp. 517-524.

[6] Wang, M.L. et al. (2000). "Magnetoelastic permeability measurement for

stress monitoring in steel tendons and cables." SPIE's 5th Annual International

Symposium on Nondestructive Evaluation and Health Monitoring of Aging

Infrastructure: pp. 492-500.

[7] Chen, Z. et al. (2000). "A New Magnetoelastic Stress/Corrosion Sensor for

Cables in Cable-Stayed Bridges using Measurement of And Curves." Proc Of the

2nd Workshop on ATUEDM, Kyoto, July 11.

[8] Kurokawa, S. et al. (2001). "Stress measurement of various cables by EM

sensor." Proc of JPCEA 10th Annual Conference: pp. 101-106.

[9] Sakairi, Y. et al. (2008). "Prototype double-pulse BOTDR for measuring

distributed strain with 20-cm spatial resolution." SICE Annual Conference, 2008:

pp. 1106-1109.

[10] Kim, J.-M. et al. (2012). "FBG sensors encapsulated into 7-wire steel strand

for tension monitoring of a prestressing tendon." Advances in Structural Engineering, 15, pp. 907-917. [11] Kleitsa, D. et al. (2010). "Assessment of metal strand wire pre-stress in

anchor head by ultrasonics." NDT & E International, 43, pp. 547-554. [12] Abdullah, A., Rice, J.A., and Hamilton, H. (2014). "Wire breakage detection

using relative strain variation in unbonded posttensioning anchors." Journal of Bridge Engineering, 20, pp. 04014056. [13] Schoenekess, H.-C., Ricken, W., and Becker, W.-J. (2007). "Method to

determine tensile stress alterations in prestressing steel strands by means of an

eddy-current technique." IEEE Sensors Journal, 7, pp. 1200-1205. [14] Nucera, C., and di Scalea, F.L. (2011). "Monitoring load levels in multi-wire

strands by nonlinear ultrasonic waves." Structural Health Monitoring, 10, pp. 617-629.

[15] Min, J., Shim, H., and Yun, C.-B. (2012). "Electromechanical impedance-

34

based health diagnosis for tendon and anchorage zone in a nuclear containment

structure." SPIE Smart Structures and Materials+ Nondestructive Evaluation and

Health Monitoring: pp. 83454O-83454O-83459.

[16] Bartoli, I. et al. (2011). "Use of interwire ultrasonic leakage to quantify loss

of prestress in multiwire tendons." Journal of Engineering Mechanics, 137, pp. 324-333.

[17] Liu, J.-G., and Becker, W.-J. (2001). "Force and stress measurements with

eddy-current sensors." Proceedings of Second Conference of SENSOR: pp. 23-

28.

[18] Ricken, W., Schoenekess, H., and Becker, W.-J. (2006). "Improved multi-

sensor for force measurement of pre-stressed steel cables by means of the eddy

current technique." Sensors and Actuators A: Physical, 129, pp. 80-85. [19] Jiles, D. (1995). "Theory of the magnetomechanical effect." Journal of physics D: applied physics, 28, pp. 1537. [20] Jiles, D. (2015). Introduction to magnetism and magnetic materials. CRC press.

[21] Joule, J.P. (1847). "XVII. On the effects of magnetism upon the dimensions

of iron and steel bars." The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science, 30, pp. 76-87. [22] Dapino, M.J. et al. (1996). "Measured Terfenol-D material properties under

varied applied magnetic field levels." 1996 Symposium on Smart Structures and

Materials: pp. 697-708.

[23] Moffett, M.B. et al. (1991). "Characterization of Terfenol-D for

magnetostrictive transducers." The Journal of the Acoustical Society of America, 89, pp. 1448-1455.

[24] Clark, A.E. et al. (2002). "Smart Materials-Fundamentals and Applications.

Magnetostrictive Properties of Galfenol Alloys Under Compressive Stress."

Materials Transactions, 43, pp. 881-886. [25] Downey, P., and Flatau, A. (2005). "Magnetoelastic bending of Galfenol for

sensor applications." Journal of Applied Physics, 97, pp. 10R505. [26] Villari, E. (1865). "Ueber die aenderungen des magnetischen moments,

welche der zug und das hindurchleiten eines galvanischen stroms in einem stabe

von stahl oder eisen hervorbringen." Annalen der Physik, 202, pp. 87-122. [27] Bieńkowski, A. (1990). "The possibility of utilizing the magnetomechanical

Villari effect in ferrite as stress transducer." Journal of Magnetism and Magnetic Materials, 83, pp. 411-412. [28] Kaleta, J., Tumański, S., and Żebracki, J. (1996). "Magnetoresistors as a

tool for investigating the mechanical properties of ferromagnetic materials."

Journal of Magnetism and Magnetic Materials, 160, pp. 199-200. [29] Kaleta, J., and ŽEbracki, J. (1996). "Application of the Villari effect in a

fatigue examination of nickel." Fatigue & Fracture of Engineering Materials &

35

Structures, 19, pp. 1435-1443. [30] Lacheisserie, E.d.T.d. (1993). Magnetostriction Theory and Applications of Magnetoelasticity. CRCPress, Inc., Boca Raton, FL. [31] Institute, A.N.S. (1973). IEEE Standard on Magnetostrictive Materials: Piezomagnetic Nomenclature. [32] Institute, A.N.S. (1987). "IEEE Standard on Piezoelectricity."

[33] Griffiths, D.J., and College, R. (1999). Introduction to electrodynamics, Vol 3. prentice Hall Upper Saddle River, NJ.

[34] Buckley, J.M. (2003). "An introduction to Eddy Current Testing theory and

technology." technical paper eddyc pdf available from the internet at http://joe buckley net/papers, downloaded on Sep, 8, pp. 7. [35] Ribichini, R. et al. (2010). "Quantitative modeling of the transduction of

electromagnetic acoustic transducers operating on ferromagnetic media." IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, 57, pp. 2808-2817.

36

Acknowledgments in Korean

한권의 논문을 통해 부족하지만 석사과정을 마무리 하게 되었습니다. KAIST 에서의

석사과정 2년동안 주변 분들의 관심과 도움이 있었기에 이 논문이 나올 수 있었습니다.

석사과정동안 많은 도움을 주신 분들께 본지를 빌어 감사를 표하고자 합니다.

먼저, 부족한 저에게 항상 관심과 조언을 아끼지 않고 지도해주신 손 훈 교수님께

감사드립니다. 현실에 안주하지 않고 끊임 없이 새로운 것을 추구하는 연구자에 대한 모습을

통해 많은 것을 배울 수 있었습니다. 또한 논문의 완성을 위하여 따뜻한 질타와 격려의 말씀을

아끼지 않으신 김진근 교수님, 곽효경 교수님께도 깊은 감사드립니다.

대학원 생활 동안 가장 곁에서 많은 조언과 생활의 기쁨이 있도록 도와준 항상

유머러스하신 같은 동네 주민 기영이형, 교수님이 되신 민구형, 저의 운동 우상 진열이형, 뒤에서

묵묵히 도와주시는 형진이형, 모든 궁금증을 해결해주는 병진이, 사수 이자 연구도 잘하고 투자도

잘하는 지민이, 항상 유쾌한 랩장 수영이, 대덕고 동문이자 저의 롤모델 준우형, 연구 말고

다른걸 더 잘하는 승환이형, 대학교부터 함께 지내와서 의지 되는 순규, 남몰래 웃음을 주는

재묵이형, 다른 공부 시작한 소울메이트 병주, ICT 급할 때 항상 함께해주는 성흠이, 독일 가고

싶어하는 용탁이, 미국에서 잘 지내는 현채, 1년에 딱 한번 웃기는 건희, 잘 따라주고 죽이 잘

맞는 익근이, 어떤 말이라도 잘 들어주는 준연이, 등은 살빠진 추성훈이라는 지호, 이제 시작하는

진호, 가영이, 한국사람보다 말 잘하는 Peipei, 함께하지 못해 아쉬운 Timo, 시물레이션함께 하는

Nazira, 저에게 잘생겼다고 말해주신 영화누나, 항상 저와 티격태격 해주시는 현미누나,

안된다고만 하시는 은혜누나, 믿음이 넘치시는 재신이누나까지 깊은 감사의 말씀 드리고

싶습니다. 그리고, 항상 바쁘단 핑계로 연락을 못하지만 언제나 그 자리에 있는 혜진이, 승우,

홍범이 정말 감사합니다.

37

또한 저의 곁에서 항상 잘되라고 지켜봐 주시고 아낌없이 지원해주시고, 묵묵히 곁에

있어준 가족이 있었기에 가능하였습니다. 주신 사랑보다 더욱 크게 성장할 수 있도록 열심히

노력하겠습니다. 그리고, 저를 믿고 항상 큰 버팀목이 되어주신 이경호 박사님께도 감사드립니다.

앞서 거론한 많은 분들의 도움과 관심이 없었더라면 이 글을 쓸 수 없었을 것입니다.

여전히 많이 부족하단 것을 알기에 초심을 잃지 않고 나아갈 수 있도록 하겠습니다. 다시 한번

모든 분들께 감사드립니다.

2017년 1월 카이스트에서 이준 올림

38

Curriculum Vitae

Personal Information

Name: Jun Lee

Place and Date of Birth: Daejeon, South Korea on July 5th, 1989

E-mail: leejun@kaist.ac.kr

Education

2017 February M.S., Dept. of Civil and Environmental Engineering, KAIST, Korea

2015 February M.S., Dept. of Architectural Engineering, ChungNam National University, Korea

2013 February B.S., Dept. of Architectural Engineering, ChungNam National University, Korea

Journal Publication

* The corresponding authors are underlined.

1. Ji-Min Kim, Jun Lee and Hoon Sohn, "Development of a Warning Technique for Tension Force Loss of

Pre-stressing Strand using Eddy Current Measurement", in preparation for NDT&E International.

Patent & Copyright

1. Hoon Sohn, Jimin Kim, Jun Lee, 구조물 진단 시스템 및 구조물 진단 시스템의 동작 방법, Korea

Patent (10-2015-0112894)

2. Hoon Sohn, Jimin Kim, Jun Lee, 텐던 긴장력 모니터링용 센서와 이를 이용한 텐던의 긴장력

진단 시스템, Korea Patent (10-2016-0161523)

39

Research Projects

2015-Present Bridge Life-Span Extension Using ICT, Partial Replacement and Low-Carbon Materials

(Primary Investigator and Director): Ministry of Land, Infrastructure and Transportation

(Funded: 28,320,000,000 KRW (28,320,000 USD) for 06/01/13 to 05/31/18)

2015-2016 Noncontact, Real-Time, and Autonomous Diagnosis of Fatigue Cracks in Industrial and

Aerospace Rotor Systems (Primary Investigator with Co-Investigator, Mohammad A.

Alshudeifat at Khalifa University in UAE): 2015 Seed Money Project, KAIST, (Funded:

60,000,000 KRW (60,000 USD) for 06/01/15 to 03/31/16

2014-2015 A Smart Scanning System for Green Energy Infrastructure (Primary Investigator): The

National Research Laboratory Program (NRL) at National Research Foundation of Korea

(Equivalent to National Science Foundation in US) (Funded: 1, 548,000,000 KRW (1,

548,000 USD) for 05/01/10 to 04/30/15)

Conference Proceedings

1. 정승환, 한민석, 손훈. 대형 구조물 안전 진단 센서용 RF 기반무선 전력 전송 시스템.

한국통신학회 종합 학술 발표회 논문집 (하계) 2014, 2014, 33-34.

2. Jimin Kim, Jun Lee, and Hoon Sohn, Detection of tensile force relaxation through eddy current

measurement of a pre-stressing strand, ASEM 2015, Incheon, Korea, 25~29 August 2015

3. Seung Hwan Jung, Kyung Hak Lee, Min Seok Han, HoonSohn and Jun Lee, “RF based 915MHz Tx, Rx

design to utilize the fusion wireless power transmission system for structural health monitoring system,”

KIBSE, Uiwang, Republic of Korea, November 13, 2015.

4. Ji-Min Kim, Jun Lee and HoonSohn, “Tensile force loss warning of a pre-stressing strand using eddy

current technique,” KIBSE, Uiwang, Republic of Korea, November 13, 2015.

40

5. 정승환, 손훈, 이경학, 한민석, 이준, 구조물 안전 진단 센서용 융합 무선 전력 전송을 위한

RF 기반 915 MHz 대역 전력 증폭기 및 변압기 설계, 201 년도 한국전자파학회 종합

학술발표회, Vol. 25, No. 1, 2015 년 11 월 26 일, 코엑스(COEX)

6. Jun Lee, Jimin Kim, Hoon Sohn, Detection of tensile force loss in a pre-stressing strand using coil

impedance measurement, International Conference on Smart infrastructure and Construction, 27-29 June

2016

7. Hoon Sohn, Hyung Jin Lim, Ji-Min Kim, Suyoung Yang, Jun Lee, Yongtak Kim, “ICT innovations for

reducing infrastructure lifecycle cost,” ASCE-Civil Engineering Confer-ence in the Asia Region (CECAR),

Hawaii, August 30-September 2, 2016.