The apparatus with electromagnetic waves of frequency 2.45GHz, such as microwave ovens, are mass-produced, therefore an oscillator for the electromagnetic wave of frequency 2.45GHz is inexpensive. We study the heating mortar block system using oscillator of frequency 2.45GHz for melting snow to convert an electromagnetic wave into heat. The heating mortar block system can melt snow in a shorter time than systems with electric heating wires.
 The heating mortar block system for melting snow needs to be composed of a leaky waveguide for providing electromagnetic waves to the heating mortar blocks. The leaky waveguide has many slots emitting electromagnetic waves on the top surface. In this study, the influences that the figure of the slots gives to quantity of emitted electromagnetic waves are confirmed.
 First, the characteristic of the single slot on a metal board about emitted electromagnetic waves was confirmed by the S parameter. The metal board that a slot was installed in is connected to a coaxial waveguide, and return loss is measured. When a slot grows longer in the direction inhibiting the electric current in the leaky wave guide, return loss greatly changed. The slot with 60mm long emitted most electromagnetic waves. This length is equal to a half wavelength of the electromagnetic wave of 2.45GHz in the free space.
 Second, the characteristic regarding emitted electromagnetic waves of the slots that were arranged continually on a leaky waveguide was confirmed by equipment measuring quantity of emitted electromagnetic waves on the leaky waveguide.
 The equipment measuring the quantity of emitted electromagnetic wave on a leaky waveguide is composed of electromagnetic waves oscillator, a leaky waveguide and a coaxial waveguide converter with the metal board that a slot was installed in. The slot that emitted the most electromagnetic waves on the metal board is 60mm long. Electromagnetic waves that emitted out from each slot on a leaky waveguide is received in a coaxial waveguide converter with the metal board that the slot was installed in.
 When the length of the slot on the leaky waveguide was 30mm, quantity of emitted electromagnetic waves is stable. In addition, emitted electromagnetic waves have the period that depend on the wavelength in the leaky waveguide when they are reflected in the terminal of the leaky waveguide. However, when they are absorbed in the terminal of the leaky waveguide, the period of emitted electromagnetic waves is lost. This is because there are no stationary waves since there are no reflection waves.
 The ability of the heating mortar block system for melting snow was confirmed by melting snow piled up in Tsuruoka city, Yamagata prefecture. The heating mortar block system is composed of electromagnetic waves oscillator, leaky waveguide and heating mortar blocks. The slots that emitted electromagnetic waves stably on leaky waveguide is 30mm long. In addition, the rest of electromagnetic waves are emitted at the terminal of the leaky waveguide by the slot of 60mm long.
 As a result, the heating mortar block system for melting snow has ability of melting snow. However, the heating mortar blocks did not have the equal temperature rise under the influence of the dielectric constant of themselves. It is necessary to make a space between a leaky waveguide and the heating mortar blocks for the equal temperature rise.

抄録全体を表示

 1. Introduction
 Realistic three-dimensional (3-D) shake table tests using E-Defense were conducted for full-scale 5-story steel building specimens with/without dampers. Four different types of dampers, steel, viscous, oil, and viscoelastic dampers were alternatively attached to the frame, and small- to full-scale of the ground motions recorded at JR Takatori station during the 1995 Great Hansin earthquake were applied to the test specimen. After finishing the test series, the frame without dampers were also tested using the same motion but scaled to 0.7 times, due to the concerns of safety. This paper discusses the detailed 3-D analytical modeling for the frame without dampers, such that it can be used for the subsequent correlative analyses of the damped cases.
 2. Outline of Analysis Model
 The model closely reflects the actual properties of the frame. It considers slab-girder interaction which appeared to be significant even under the negative bending moment. The detailed model includes the slab effective width measured from the test and varied beam-by-beam. The haunch and gusset plate are considered for beam and column modeling. Rational models for the panel zones and column bases were formulated, and rotational (torsional) inertia due to the size of the floor and scattered simulation weights are considered.
 3. Simulation by Analysis
 The model appears to accurately simulate the experimental results. The accuracies of the global responses such as peaks and histories of story displacements and torsional rotations, drifts, accelerations, and shear forces are verified. Those of the local responses such as bending moment, axial force, and shear force of the beams and columns, as well as panel zone moments and deformations are also verified. Comparison with experimental results enabled to understand and clarify the observed behavior of the frame.
 4. Effects of Modeling Schemes on Prediction
 The accurate model discussed above is also used as the benchmark for studying effects of various modeling techniques on accuracies. In total, eighteen different models of the 3-D frame produced and the sensitivity of the analytical responses due to them are discussed systematically. In addition to the JR Takatori motion, Japan's design basis artificial earthquake called BCJ-L2 motion is also used in order to draw more general observations based on much less irregular spectrum characteristics of the motion.
 5. Analytical Study on Typical Design Model
 The practical and reasonably accurate modeling scheme is chosen from the above studies and is used to analytically evaluate a much weaker frame satisfying the minimum code requirements. The results indicate the frame although satisfying the minimum code requirements will be damaged severely under the JR Takatori motion, and will still undergo large inelastic deformation even with the BCJ-L2 motion. The present frame performs much better, but shows larger drifts than desired to keep building's functional continuity, thereby warranting the use of the dampers.
 6. Conclusions
 This paper discusses correlative time-history analyses of the frame portion of the specimen that was tested after the whole test series using the dampers. The detailed model closely reflecting the actual properties of the specimen appears to accurately simulate the experimental results, and is utilized to analyze and evaluate the frame for the larger table motion not used due to the safety reasons.
 It is also used as the benchmark for studying effects of various modeling techniques on accuracies. The practical and reasonably accurate technique is used to simulate a much weaker frame satisfying the minimum code requirements.

抄録全体を表示

 In the 2016 Kumamoto Earthquake, ground motions of seismic intensity 7 occurred twice in a short period. When subject to these ground motions, wooden structures with stiffness decreasing type of load-deformation property are very likely to collapse. In order to provide for this type of severe earthquake, we have been developing a response control method is for low-rise structure with a variety of dampers. Among them, a negative stiffness control method found to be advantageous to reducing the absolute acceleration response. The negative stiffness is produced by making use of high speed moving fluid's inertia mass. However, most existing fluid dampers have not made the best use of inertia mass effect because their damping effect is dominated by viscous damping force.
 In this paper, we propose a method for performance based design of a Linked Fluid Inertia Mass Damper (LFIMD). In this study, we regard high-performance as exerting inertia force relatively greater than viscous damping force. The damper is composed of two cylinders, two piston rods and two link tubes that connect each oil chamber of a pair of cylinders. There is no orifice on the piston and the fluid runs in the tube at high velocity when piston rods move back and forth. The damper has three distinguished mechanical properties; they are damping effect by viscous fluid, inertia mass effect by moving fluid at high velocity and the displacement control by hydraulic link mechanism. The damper is advantageous to reducing response acceleration, because the load-displacement relationship of the damper exhibits negative stiffness by inertia force, which works to reduce the inertia force in the whole structure. The LFIMD can exert negative stiffness by means of simple passive mechanism.
 In the first part of the paper, we study how damper's performance depends on link tube specifications. If we use a softer plastic tubes, the damper exerts smaller link effect. But, in the case of using more rigid copper tubes, two piston rods link fairly well and load-displacement relation of the damper draw a clear oval with negative stiffness. The test shows it is necessary to use rigid tubes to exert stable linking performance and large damping force.
 In the Second part of the paper, we try to improve the performance of the damper. There are two methods; first one is to use high density fluid and second one is to reduce the viscosity of the fluid. The first method is to achieve high-performance by directly increasing inertia force. The second method is to relatively increase the inertia force by reducing viscosity of the fluid. We confirmed by experiments that two methods are effective for high performance.
 In the last part of the paper, we conducted a shaking table test to demonstrate how the negative stiffness works to improve seismic response control performance. In the test, we use two kinds of fluid with different density. The test shows that increasing inertia force with high density fluid is advantageous to reducing acceleration response. Finally, we composed analytical model for shaking table test, showing that the analytical results simulate the experimental results fairly well.

抄録全体を表示

 It is being revealed that occurrence of the Nankai-Trough earthquake might cause significant damage to low-damped super high rise steel buildings in some metropolitan areas in Japan. Addition of supplemental damping devices is effective means to mitigate the damage caused by the great earthquake. However, the seismic effectiveness of dampers connected to a structural frame tends to lose due to overall bending deformation in a slender steel building. On the other hand, a tuned mass damper (TMD) connected to the frame work well even if the overall bending deformation is quite large. The combination of the two kinds of dampers therefore is expected to be effective in slender buildings with a large aspect ratio. However, an optimal design parameter of the TMD mounted on highly damped buildings and its seismic effectiveness are not clarified. The objective of this paper is to develop a design procedure of the two different damping devices. Both the dampers are limited to a linear viscous damper in this study.
 Firstly, a lumped mass stick model with shear springs and rotational springs is employed to take account of the overall bending deformation. Two degree of freedom (2-DOF) system is derived from this model with the help of dynamic reduction technique. The 2-DOF system becomes non-proportional damping system. Subsequently, a linear viscous TMD are connected to the 2-DOF system and a 3-DOF system is obtained for the TMD-building interaction system. Optimal design parameters of the TMD are obtained by solving a Lyapunov equation of the 3-DOF system. The optimal stiffness of the TMD is summarized in terms of equivalent damping factor due to only dampers installed into the stories. The computed optimal stiffness is almost consistent with the result of SDOF system with the same equivalent damping factor. This fact admits to use regression formulae of the optimal stiffness proposed by several researchers. It is also confirmed that the optimal parameters obtained by the criterion of minimizing top displacement and story drift is almost the same.
 Secondary, equivalent damping factor of the TMD-building interaction system are investigated through stochastic dynamic analysis. As a result, the equivalent damping decreases according to increase of the amount of the dampers within stories. The convenient regression formula is proposed in terms of the equivalent damping factor of the building. The equivalent damping factor is directly evaluated with a slenderness ratio of the building and a damping factor by using the formula.
 Finally, a design procedure of the TMD is formulated. Required amount of the TMD mass is proposed based on the concept that the loss of damping factor is equal to the additional equivalent damping factor added by the TMD.

抄録全体を表示

 We propose the evaluation and design method of a viscos damping mechanism in the passive-type base isolation system aiming at the acceleration reduction. Through the appropriate setting of the stiffness and the damping between the ground and the building, the longer-period and the additional high damping are attained and the approximate input insulation can be achieved. To get the input insulation effect, the analytical model, the base isolation period of 9 seconds, is analyzed. To restrain the displacement of the base isolation story, the high damping factors of 40% is considered. We found that the input insulation effect can be kept favorably by the Maxwell mechanism, named ‘Maxwell model’ here, where a spring and a dashpot are arranged in series.
 In this paper, we have analyzed the characteristics of ‘Maxwell model’ by theoretical analyses of the transmissibility, Fourier spectrum analyses, and seismic response analyses. The results of this study are summarized as follows:
 (1)We made it clear from the displacement-acceleration earthquake spectrum that the specifications satisfying the target performance are the natural periods of 6-8 seconds and damping factors of 30-40%.
 (2)From the analyses of the transmissibility of the single-degree-of-freedom model, the ‘Maxwell model’ can expect the response reduction effect by the smaller transmissibility compared to the dashpot model, named ‘Usual model’ here, above the specific frequency ratio. Its frequency ratio is about 5 times the natural frequency of the ‘Usual model’.
 (3)From the analyses of the Fourier spectrum ratio of the fifth-degree-of-freedom model, ‘Maxwell model’ can expect the better response reduction effect for the shorter natural period of the building and the longer isolation period, since the natural frequencies of the insulated system tend to increase.
 (4)’Maxwell model’ indicates the drastic response reduction effect of the 1st floor's maximum acceleration response compared to the ‘Usual model’ for the longer period and the larger damping factors of the base isolation, and the shorter natural period of the building. ‘Maxwell model’ can reduce the maximum acceleration response to 60 % of the response of the ‘Usual model’ on average for the analytical cases of various earthquakes.
 (5)The maximum isolation displacement response of ‘Maxwell model’ is almost equal to the response of the ‘Usual model’.
 For the proposed system, the acceleration response can be less than 80cm/s2 and the isolation displacement can be less than 50 cm for extremely severe earthquakes, in the setting condition of long natural period of 7-9 seconds and the high damping ratio of 20-40%.

抄録全体を表示

 It is recognized that a base-isolated building exhibits a large response to a long-duration, long-period ground motion and an interconnected building system is not necessarily effective for a pulse-type ground motion. To compensate for each deficiency, a new hybrid passive control system is investigated in which a base-isolated building is connected to another building (free wall) with oil dampers. It is shown that the mechanism of response reduction in the proposed hybrid passive control system is closely related to the property of natural frequency and damping ratio with respect to increased capacity of connected dampers in a translation mode and a meeting-parting mode.
 First of all, the model H (hybrid) and the model N (normal) are defined in Section 2 and the natural frequencies and damping ratios of these models are formulated in Section 3 through the complex eigenvalue analysis. There are three cases (A, B, C). Case A is the case where all four eigenvalues are complex values, Case B is the case where two are complex and the other two are real and Case C is the case where all four eigenvalues are real. The total damping coefficient of the connecting dampers is increased in order to investigate the properties of the models H and N. It is shown that the three cases (Case A-C) appear corresponding to the range of the total damping coefficient of the connecting dampers.
 In Section 4, it is demonstrated that there are two eigenmodes, called the translational mode and the meeting and parting mode, for the infinite total damping coefficient of the connecting dampers. The eigenmode for a finite total damping coefficient of the connecting dampers corresponding to the translational mode is called ‘a mode toward the translational mode’ and that for a finite total damping coefficient of the connecting dampers corresponding to the meeting and parting mode is called ‘a mode toward the meeting and parting mode’. The eigenvalues and eigenmodes of the mode toward the translational mode and the mode toward the meeting and parting mode are derived in Section 4.2 and 4.3.
 In Section 5, the response characteristics of both models (model H and model N) to the pulse-type earthquake ground motions and the long-period, long-duration ground motions are investigated. It is shown that the relation between the fundamental natural period of the main building and the predominant period of the long-period, long-duration ground motions plays a key role in the understanding of the response characteristics of both models (model H and model N). Furthermore it is also made clear that the relation between the fundamental natural period of the main building and the predominant period of the pulse-type ground motions plays a key role in the understanding of the response characteristics of both models (model H and model N).
 Finally it is concluded that the mechanism of response reduction in the proposed hybrid passive control system is closely related to the property of natural frequency and damping ratio with respect to increased capacity of connected dampers and the relation between the fundamental natural period of the main building and the predominant period of the pulse-type or long-period, long-duration ground motions plays a key role.

抄録全体を表示

 This paper proposes a new beam-column FEM which consists of several fiber elements considering shear deformation. The distribution of normal strain obeys Bernouli-Euler hypothesis. The distribution of shear strain is suggested here to be Equation (17). The proposed distribution is expressed by the product of distribution of shear strain and shear deformation angle at the elastic state. The formulation is accomplished based on the incremental perturbation method. Normal stress-strain relation and shear stress-strain relation in the fiber element are supposed here to obey each different consistent law independently.
 The new generalized displacement vectors with a shear deformation angle and the corresponding force vectors are defined here as shown in Eqs. (2) and (3). In order to analyze combined non-linear problem, the moving coordinate system in the previous beam-column FEM known as FERT is adopted here. The matrix of coordinates transformation is conducted here. Also, the formularization described in chapter 2 is developed by the incremental perturbation theory.
 The verification of the proposed numerical method named FERTs-P is carried out for the problem of elastic beam and elastic-plastic beam. Also the simulation of the steel beam experiment and the frame experiment with a shear panel are done by the FERTs-P. The section quantities consisted by fiber elements are set up to be equivalent to area, moment of inertia and plastic section modules. The tri-linear models are adopted here as the relation of shear stress-strain and that of normal stress-strain.
 The main observations are as follows:
 (1) The ratios of shear deformation to bending deformation in elastic beams are almost coincident with theoretical values due to Timoshenko beam.
 (2) The numerical results by FERTs-P are relatively good predictions for the experimental load-deflection curves of beams and frame yielding at shear.
 (3) The FERTs-P is effective to predict elastic-plastic bending behavior.
 In spite of bold modeling, this numerical method may be expected considerably. Moreover the consistent law in shear stress-strain relation will be considered.

抄録全体を表示

 The authors investigated the damage of pile foundations during the 2016 Kumamoto earthquake and carried out simulation analyses to estimate a cause of the damage. We show below conclusions provided by these studies.

 It occurred that a building was inclined and became unusable continuously for the damage of pile foundations, though there was little damage of the superstructure by the earthquake. The damaged building had the following characteristics, those were that the kind of piles was pre-stressed concrete piles, and the building was built on the alluvial plain, and there were various damage patterns of piles. These events were common with the 2011 off the Pacific coast of Tohoku earthquake.

 To analyze process reached to severe damage, a static analysis model consisted of both piles and ground springs considering nonlinearity was employed, and the lateral load calculated based on observed earthquake motion at pile head and ground displacement were given to this model. The results of our analyses showed that this analytical model was useful for evaluation of process reached to damage and estimation of seismic performance of pile foundations during severe earthquakes.

 It was confirmed that the damage did not occur when only load caused by earthquake motion was given to piles, in contrast the results obtained by our analyses almost accords with the actual damage situation when ground displacement by the cause except the earthquake motion was added. Influence of movement of the fault or embankment was thought about as for the prime cause of added ground displacement, but it is necessary to investigate quantitatively in the future.
 To secure the performance of piles and continuous use after severe earthquake of the building on particular ground condition like nearby the faults, the consideration for the ground displacement caused except the earthquake motion may be necessary even if it is relatively small building and is not the liquefaction ground.

 Results obtained our analyses accorded mostly with tendencies of real damage, and this study showed that axial load and the ground displacement distribution have a strong influence on the damage of each pile. On the other hand slightly difference of each pile in damage situation appeared. In order to explain the difference, analysis conditions must be set individually even if existing in the same pile cap. Therefore for realization of the quantitative method, accumulation of data about actual damage, structural experiments or analyses are necessary.

 Judging from obtained results, when seismic designed method based on the condition that design horizontal seismic intensity was decided to 0.6 without ground displacement was employed, it seems assumption on the safe side against severe earthquakes. However, the design using this method may not guarantee safety when ground displacement by the cause except the earthquake motion occur, in particular a close investigation for securing of shear capacity is required.

抄録全体を表示

 Tensile bolt joint is one of the joint methods of timber frame structure. In the failure mode of this joint method, shear fracture from the tensile bolt fixed portion to the beam or column “butt end” (= the surface of the end of beam or column) is very brittle. Therefore, it is important to predict precisely the fracture load of this failure mode.
 This study aims to propose the fracture load formula of tensile bolt fixed portion of glued laminated timber frame with tensile bolt joint from the relationship between “shear-length”, maximum load and failure mode. First of all, the destruction tests of the tensile bolt fixed portion of the same grade glued laminated timber were carried out. Moreover, the relationship between shear-length and fracture load or the area of tensile bolt fixed plate and fracture load are analyzed. The knowledges from this study are as shown below:

 1) About the load-deformation relationship of the specimen that the shear length is over 240mm, after the maximum load experience, deformation proceeds without lowering load. On the other hand, the specimens that the shear length is less than 180mm are destroyed in less load than the specimens that the shear length is over 240mm.

 2) When the shear length is up to 300mm, shear failure is the major failure mode, and the fracture load is increased in proportion to the shear length. When the shear length is over 300mm, compression failure is the major failure mode, and the fracture load is about the same regardless of the length.

 3) The fracture load formula of tensile bolt fixed portion of glued laminated timber frame with tensile bolt joint was proposed. About the range of shear length from 120mm to 420mm, it is confirmed that the proposed formula can predict the fracture load to a safety side.

抄録全体を表示

 Wooden grid wall which is composed by joining half-lap wood in a grid generally has been used as bearing wall. Structural feature of the wooden grid wall is resistance to external force by embedment of half-lap joints at grid intersections, and it is known to have generally excellent deformation performance and high toughness. In addition, wooden grid wall constitutes assembling the half-lap wood in a grid. It was reported that the rigidity and strength vary by the number of half-lap joints. Due to these features, wooden grid wall performance can be altered by changing the number of grids. In this way it can be used as an earthquake resistant element. Because the embedment of wood affects the loading rate, this paper will consider how the structural performance of half-lap joints is affected by the loading rate.
 Based on the above background, the purpose of this paper is to accumulate basic data on the dynamic behavior of wooden grid wall. First, dynamic loading tests were done to understand the influence of the loading rate and the number of half-lap joints. In the experiment, as the target specimen had multiple joints, dynamic loading tests were performed to determine hysteresis curve representing the relationship between the bending moment in the joints and the rotation angle. Next, the envelope from the resulting history curve was calculated by considering the number of joints, and it was verified through the summation rule. Finally, the energy absorption performance of half-lap joint was evaluated by calculating the equivalent damping factor and hysteretic energy.
 The results of the experiment indicate that in the loading rate range from 0.01 to 50kine, about the strength, the impact of the equivalent damping factor and hysteretic energy on the loading rate could not be confirmed. Regarding rigidity, the large trend with increasing loading rate was confirmed. Through verification using the summation rule, methods to substantially increase rigidity and strength of the test specimen in proportion to the number of joints has been understood. In addition, that this is not affected by grid spacing was also understood. Therefore, the rigidity and strength of the wooden grid walls and the rigidity and strength of the joints constituting the walls could be inferred by multiplying the number of joints. The hysteretic energy become larger with the increase in the number of joints, and the nature of the hysteric energy was not affected by the grid spacing. Irrespective of the loading rate, when the rotation angle was greater than about 0.1 rad, the number of joints, grid spacing and the equivalent damping factor constantstended to converge to a range from 0.10 to 0.15.

抄録全体を表示

 A ductility-type seismic retrofitting technique is proposed for reinforced concrete (RC) columns. This involves arranging high-strength steel prestressing bars (PC bars) on the four faces of an RC column as external hoops. Hence, active confinement due to the tensile forces in the PC bars as well as passive confinement and transverse reinforcement can be expected. With regard to the active and passive confinements, previous authors have shown that the compressive strength and ductility of confined concrete are greatly improved and have proposed a stress–strain relation for confined concrete. The aim of the present study is to experimentally investigate the shear strength and shear resistance mechanism of RC columns retrofitted by this technique.
 The retrofitting details of the test specimens are shown in Table 1. The elevation and cross-section are shown in Fig. 1. To investigate the shear resistance mechanism (truss or arch mechanism), two types of specimen are considered in this study. One is a retrofitted RC column with no bond force between the concrete and the embedded longitudinal reinforcement, thereby generating the arch mechanism. The other is a retrofitted RC column with bonded rebars to evaluate the truss mechanism, which relies heavily on the bond resistance between the rebars and concrete. The column specimens used in this test had a square cross-sectional dimension of 250×250 mm², a height of either 500 mm or 750 mm, and a shear span-to-depth ratio (M/VD) of either 1.0 or 1.5. The longitudinal reinforcement ratio (p_g) of the retrofitted specimens with bonded rebars was either 3.67% (8-D19) or 5.51% (12-D19) and that of the specimens with unbonded rebars was 1.36% (12-D10). The shear reinforcement ratio (p_w) of the column specimens was 0.08%. The test parameters of the column specimens are the initial tension strain of the PC bars, their spacing, the axial force ratio, shear span-to-depth ratio, and the bonded/unbonded nature of the rebars.
 In Chapter 3, the following main points are discussed. (1) The initial tension force of the PC bars enhances the shear strength of the truss mechanism. (2) Applying a lateral confinement pressure to an unbounded RC column, the shear strength of the arch mechanism can be increased. (3) For the bonded specimens, the measured gradient of the compressive diagonal force of the truss mechanism, which is one of the internal forces of an analogous truss comprising PC bars that act as tension members, the bond force of rebars, and the diagonal concrete force was nearly 45°. (4) The shear strength of the arch mechanism increases with increasing axial force. (5) In the unbonded specimens, the slope of the line of thrust and the depth of the compression zone of the arch mechanism correlate with the lateral confinement pressure and the vertical axial load. (6) In the bonded specimens, the depth of the compression zone of the arch mechanism had no correlation with the lateral confinement pressure and the vertical axial load.
 Chapter 4 describes the comparison between the experimental shear strength and calculated results based on the modified Arakawa mean and Minami equations shown in Figs. 17 and 18. These figures show that the test results are in agreement with the modified Arakawa mean equation, and that the modified Minami equation can reasonably be applied to assess the shear strength considering the truss and arch mechanisms.

抄録全体を表示

 Fiber reinforced cementitious composite is a composite consisting of concrete/mortar as matrix and short fibers as reinforcement. The fiber resists external forces by bridging effect across cracks. Polypropylene short fiber is one of the improved polymer fibers and has already been used to prevent spalling of cover concrete, to reduce drying shrinkage strain, and to prevent explosive spalling during a fire. Because of the wide applicability, the polypropylene short fiber is standardized in Japan Industrial Standard in 2015. The standardization is the first case as a reinforcing material for concrete and significantly benefitted manufacturers and consumers. Polypropylene is chemically hydrophobic and the bond between concrete is inferior comparing to hydrophilic polymers like vinylon fibers. The surface treatment, such like emboss processing, is therefore common for the polypropylene fiber to enhance the bond strength by mechanical method.
 The utilization of polypropylene short fibers for reinforced concrete structures have been studied from the viewpoints of prevention of spalling of cover concrete during a fire and durability under impulsive forces. Flexural characteristics of polypropylene reinforced concrete without steel reinforcing bars have also been investigated as well as bridging effect and fracture energy across cracks. On the other hand, only a few investigations have been attempted for flexural and shear characteristics of polypropylene fiber reinforced concrete with steel reinforcing bars unlike a number of researches on the structural behaviors of reinforced concrete members containing steel fibers or vinylon fibers. If the strength and ductility of reinforced concrete structures can be enhanced by mixing the polypropylene short fibers, then it may be applied to parts where steel reinforcing bars cannot easily be arranged due to complicated multi-axial stress condition, such like beam-column joints or strengthening around duct holes in walls and beams. This paper presents experimental investigations on basic mechanical properties of the polypropylene short fiber reinforced concrete, bond between steel reinforcing bars, and shear and flexural characteristics of beams.
 Chapter 2 describes plan and results of basic mechanical tests of the fiber reinforced concrete. Five kinds of concrete are prepared: concrete without fiber, concrete mixed with polypropylene fiber of 0.4% volume ratio, that of 1.2 %, concrete mixed with vinylon fiber of 0.3% volume ratio, and that of 0.9 %. Compressive tests of cylinders, three-point bending tests, and pull-out tests of steel bars. The test results show that the compressive strength and ductility, flexural strength and ductility, and bond strength increase as the volume ratio of the fibers increases. An excellent property is observed especially in concrete mixed with the polypropylene shot fibers of 1.2% volume ratio.
 Chapter 3 presents four-point bending tests of beams of 1,400 mm length and 100 mm × 200 mm cross section. Ten beams are prepared with five kinds of fiber-reinforced concrete mentioned above and two kinds of longitudinal bar ratios. Significant increase of post-yield ductility is observed in the beams with the larger amount of longitudinal bars while the ductility enhancement is not always significant in the beams with the smaller amount of longitudinal bars. In addition to these ten beams, four short span beams are prepared to examine the size effect. It is observed that the mixture of the polypropylene short fibers of 1.2% volume ration reduce size effect on the post-peak ductility after shear cracking.
 Chapter 4 resumes the above test results and concludes that the bridge effect of the polypropylene short fiber across the cracks prevent the localization of concrete fracture and improve the ductility of beam members.

抄録全体を表示

 In recent years, applications of buckling restrained braces (BRBs) in reinforced concrete (RC) frames have attracted much attention. It is necessary to reduce the damage in the BRB connections in RC frames to make the BRBs more effective in dissipating earthquake energy. In this paper, it is proposed to use an unconstrained gusset plate and a hinge relocation technique to control the damage to RC beam-column joints. The structural behavior of RC beams with BRB connections and relocated plastic hinges is investigated by both cyclic loading tests and finite element analysis. In the tests on RC beams with BRB connections, the BRB axial force was simulated by a force-controlled actuator to simplify the boundary condition of the test setup. Two different details of BRB gusset connections were examined. In one method, the gusset plate was fastened to the end of the RC beam by post-tensioned steel rods, whereas it was embedded in the RC beam and anchored there by a stud group in the other method. The test results confirmed that the proposed hinge relocation scheme is effective in controlling the locations of plastic hinges in RC beams so as to prevent the BRBs from being dislocated from the entire structural system as a result of the beam end failure at extreme earthquake scenarios. Both gusset connection details performed satisfactorily by providing sufficient strength and large stiffness to minimize the deformation loss in the BRB connections.
 Detailed finite element analysis was carried out to reproduce and extend the understanding of the test results. In the analysis, the posttensioned steel rods, the embedded studs and the bond-slip behavior between the embedded gusset plate and the concrete were explicitly modelled. The analysis results agree well with the test results. A parametric study was performed to determine the flexural strength demand for the beam sections that are subjected to the BRB-induced tensile force but are not expected to yield. The results show that the strength ratio, β_h, should be no less than 1.14 for the BRB connection to remain essentially elastic.

抄録全体を表示

 Removing formworks early is in great demand for reduction of construction period. It has been studied from both material and structural fields and is accepted in the AIJ Guidelines. However, the detailed effects of early removing such as bond stress are not well known. In the previous experiment conducted in summer, compressive strength of the upper layer of the reinforced concrete (RC) slab was found to be a half of that of the lower layer, which resulted in smaller bond stress between the upper reinforcement and the concrete and larger flexural crack width at the upper surface than the lower surface. In this research, eight specimens of RC slab are tested by applying sustained bending moment outside in winter. The specimens are sprinkled with water in the morning after the concrete casting and demolded after four or twenty-six days. The upper surfaces of the specimens are exposed to sunlight and rain for five or eight months while loading. The test parameter is the magnitude of the bending moment: zero, -0.5M_c', -1.0M_c', -1.5M_c and +1.5M_c', where M_c is the moment at flexural crack and “-” or “+” indicate the upper surface is subjected to tension or to compression. Figure 2 shows the reinforcement. Figures 3 and 4 show the loading setup. After the exposure of five months, two of the specimens were cut as shown in Figure 5 to investigate the strength of the concrete near the top and the bottom of the specimen. In the following discussion, the computed values are obtained using AIJ Guidelines and assuming that the concrete of each specimen is homogeneous and that the humidity at the upper and lower surfaces of each specimen is same as that of the open air.

 The following findings are obtained from the results of the experiment.
 (1) At the age of nine months, the compressive strength of the concrete of the upper layer of the specimen was 0.8 times of that of the lower layer (Figures 7b and 7c).
 (2) The observed crack widths and the bond stresses of the specimens with early demolding and +1.5 M_c moment were similar to those with -1.5 Mc moment (Figures 19, 20, and 21). The observed crack widths agreed with the predictions by the AIJ Guidelines (Figure 19).
 (3) Assuming curvature change to continue for 30 years, the curvatures after 30 years were predicted. The predicted curvatures of the specimens with early demolding fell within 6 to 12 times the elastic curvatures as indicated by the AIJ Guidelines (Figure 17).
 (4) In the specimens with early demolding, the increase of crack widths and curvatures and the decrease of bond stresses were less than those in the specimens with normal demolding (Figures 16, 17, 18, 19, and 21a).

抄録全体を表示

 Rectangular hollow section column-to-wide flange beam connections are commonly used in moment resisting frames in Japan. In such a connection, the moment in the beam web is transmitted to the column through the column skin plate. When the thickness of the tube wall is thin, out of plane deformation of the skin plate occurs easily, so the joint efficiency in beam web is low. The connection coefficient a is defined as the ratio of the maximum strength of beam connection to the full plastic strength of the beam, where the former is calculated considering the joint efficiency in beam web. The plastic deformation capacity of the connection is low when the connection coefficient a is small. For beam-to-column connections with low joint efficiency, the strength of the panel is relatively small due to small thickness. As a result, the panel might yield earlier than the beam. In this case, the plastic deformation capacity of the connection might increase. Hence, different failure modes have to be taken into consideration when evaluate the plastic deformation capacity of such connections. Furthermore, for moment resisting frames consisted of WF beam and RHS column with low joint efficiency in beam web, it is not clear how the yield of the panel would affect the seismic performance of the whole frame. In this study, to evaluate the plastic deformation capacity of the beam-to-column connection with low joint efficiency in the beam web considering that the panel yields, cyclic loading tests of beam-to-column connections were conducted. Moreover, nonlinear time-history response analyses of multi-story steel moment frames were conducted to assess whether the connections reach their ultimate states during earthquakes.
 Four beam-to-column connection specimens with same beam section and different panel section were tested. The test results showed that as the connection coefficient a decreased, the panel-to-beam strength ratio R_p also decreased. As a result, the plastic deformation capacity of the connection increased due to yielding of the panel. In this case, the ratio of energy dissipation of the panel and that of the whole connection increased. When R_p was about 1.1, the ratio of energy dissipation of the panel and that of the connection was about 0.5.
 In the nonlinear response analyses, moment resisting frames with 3 stories, 6 stories, 9 stories, and 12 stories were employed. The panel-to-beam strength ratio R_p in these frame verified from 0.8, 1.0, 1.2, to 1.5. Five seismic records with different amplification factors were input during the analyses. From the analytical results of 16 frames, it is concluded that (1) under Level 2 earthquakes (PGV=50cm/s), when the connection coefficient a is smaller than 1.2, “Class C” (defined by the Japanese design code) beam members are not likely to reach their ultimate states. However, “Class A” beam members might reach failure. (2) under Level 2 earthquakes (PGV=50cm/s), when the connection coefficient a is over 1.2, “Class A” beam members are also considered to be safe. (3) under Level 3 earthquakes (PGV=75cm/s), even when the connection coefficient a is larger than 1.2, both “Class C” and “Class A” are likely to reach failure. Nevertheless, if the panel-to-beam strength ratio R_p is small enough that the panel is yielded to a certain level, both classes of beam members would not fail during the earthquake.

抄録全体を表示

 A developmental study of Concrete Encase Steel (CES) composite structure system has been continuously conducted toward its practical use. CES structure is composed of steel and fiber reinforced concrete. In previous studies, it was found that the CES structure has good seismic performance from the experimental study of columns, shear walls and a two story two span frame. As for CES beam-column joints, it was confirmed that failure behavior, strengths and restoring force characteristics, CES beam-column joints also possess good seismic performance. It was shown that the flexural strength of the beam could be evaluated by superposition strength theory and the shear strength of the joint panel could be evaluated by a formula in AIJ SRC standard conservatively. The shear strength evaluation method in the SRC standard calculates shear strength as a product of the effective area of concrete and the shear stress of the concrete at its ultimate state. On the other hand, three-dimensional FEM analysis was conducted to understand the stress transfer mechanism of the joint panel. As a result, it was found that the stress state was different in the area which was surrounded by steel flange and the area which was not. However, it is thought that the structural performance of the joint panel is affected by many structural factors and it is still the case that experiments of CES beam-column joints with shear failure of the joint panel are low in number.
 Therefore, in the present study, static loading tests of CES beam-column joints with different cross sectional configurations of joint panels and length of columns were conducted to grasp the structural performance and stress state of the joint panel and examined by simulation with FEM analysis. Furthermore, parametric analysis was conducted in order to understand the effect of structural valuables on the stress state of concrete in the joint panel, and the validity of an effective area of concrete in the ultimate shear strength evaluation was examined.
 Firstly, in comparison between specimens with different cross sectional configuration of joint panel, the damage condition of concrete was different and it could be thought that the stress state of concrete was different depending on the peripheral configuration to the joint panel. FEM analysis of test specimens was conducted, where the analytical relationship of shear force and drift angle was in approximate agreement with that of test results and the validity of the analytical modeling method could be confirmed.
 Secondly, parametric analysis was conducted with a verified modeling method and the effect of axial force ratio, concrete strength, and width of steel flange in joint panel on shear strength of the joint panel was examined. From these analyses, the difference in shear strength of the joint panel was within 10% depending on axial force, the effect of the width of the steel flange was small although the width of the joint panel steel flange on shear strength affects to the shear stress state depending on the area inside or outside of the steel flange. In addition, the effect of the compressive strength of concrete on shear strength was dominant and shear strength increases with increasing the compressive strength of concrete.
 Finally, a calculation formula of the shear strength of the joint panel in a CES beam-column joint was proposed based on knowledge obtained from analysis and experimental results. The shear strength of the concrete panel was expressed with the effective area of concrete and the average shear stress of concrete at the ultimate state in the joint panel. Shear strength, calculated with the proposed formula, was in good agreement with the experimental data.

抄録全体を表示

 Recently, a number of designers have focused on free-form surface shell to realize a free architectural form that is different from the analytical curved surface such as cylindrical or spherical surfaces. However, in order to create rational architectural forms, constructability and cost are also essential factors to be considered.
 Developable surface is a special form of ruled surface generated by continuous movement of the straight line. It can be obtained by adding the condition that the normal vector of the surface does not change along the generating line (generatrix). Because the generatrix is a straight line without torsion, the formwork of continuum shell is easily created. Since the twisting process is not required, it has a high workability characteristics.
 In this study, several developable surfaces are combined to form a curved roof structure. The (n,1) Bézier surface is used for modeling the surface. Optimization problem is formulated for minimizing the maximum principal stress under several static loading conditions including vertical and horizontal loads. The coordinates of control points of the Bézier surface are chosen as design variables. The developability condition is numerically assigned so that the tangent vectors at the same parameter value of the two Bézier curves along the boundary exist in the same plane as the directing line. The G⁰ and G¹ continuity conditions are assigned for connecting the Bézier surfaces. Optimal solutions are found using nonlinear programming approach, where the sensitivity coefficients are computed by the finite difference approximation.
 As the result of optimization, a variety of developable surfaces are obtained by connecting Bézier surfaces. Since the control points of the curves are chosen as design variables, the calculation efficiency is high. The stress distribution also greatly improved by using the maximum stress as the objective function.

抄録全体を表示