Recycled aggregates are classified into four types of quality (JIS A 5023, JIS A 5022, JIS A 5021, JASS 5N). Although there were many reports and studies on recycled aggregate and concrete, few reports focused on adhered mortar which were contained in high-quality recycled aggregate.

　In the previous paper we reported that the adhered mortar ratio of the recycled aggregate and mortar particle amount had a high correlation with the water absorption and the absolute dry density, and most of the mortar content was the weight of mortar particles.

　We are studying the properties of concrete using high-quality recycled aggregate from demolished concrete.

　This paper presents the effect of mortar particle amount on the unit water content, compressive strength, and the Young’s Modulus of concrete.

(1) Regardless of the method for producing high-quality coarse aggregate, the larger the amount of mortar particle in recycled aggregate, the lower the unit water content of fresh concrete with the same slump. There is a negative correlation between the amount of mortar particles and the unit water content. Since the resistance between coarse aggregates was reduced by the rounded mortar particle and the shape of original gravel.

(2) Regardless of the method of producing high-quality coarse aggregate, the compressive strength of concrete using recycled aggregate has a negative correlation with the amount of mortar particles, since mortar particles cause fragile parts in hardened concrete, so the compressive strength of concrete in the seal curing trend to be lower than in standard curing.

(3) The compressive strength of concrete using high-quality recycled coarse aggregate concrete was about 20% lower than crushed stone. The ratio was increasing steadily in the case of the high strength concrete. It was due to the fragility of the mortar particles, as well as the different of surface structure on aggregate and shape of them between recycle aggregate and crushed stone.

(4) The Young’s Modulus of recycled aggregate concrete decreases depending on the amount of mortar particles in recycled coarse aggregate, as well as the compressive strength. The relationship between the compressive strength and the Young’s Modulus of high-quality recycled coarse aggregate concrete can be evaluated by the formula of Architectural Institute of Japan.

　The durability of Reinforced concrete (RC) buildings are basically determined by the presence or absence of steel corrosion. In order to prevent corrosion of steel in concrete, degradation factors such as carbon dioxide (CO₂) and chloride ions need to be prevented from penetrating into concrete. In many RC buildings, finishes are applied to concrete surfaces, which prevent the penetration of degradation factors. While many studies have been reported on the preventive effects of finishing materials based on accelerated testing, little is based on the long-term exposure to the outdoor conditions. In this study, a model building made from concrete block specimens whose surfaces were coated with various finishing materials was constructed for the outdoor exposure test. After 30 years of outdoor exposure, the deterioration of finishes and their carbonation and corrosion inhibition effects were investigated.

　The follow conclusions could be drawn based on this study:

(1) As for the degradation status of finish materials, all finish materials undergo some kind of degradation after a long period of outdoor exposure. In the case of dirt and chalking, rainwater and other influences may improve the judgment of the deterioration status.

(2) As for the carbonation depth, those with multiple layers of finishes, such as multi-layer type, those with low moisture permeability, such as waterproofing materials, and those with thick finishes are more effective in inhibiting carbonation. On the other hand, those with water repellency or air permeability will accelerate the carbonation.

(3) Carbonation rate was less than 0.5 for coating materials, such as No.1, 4, 5, 7, 10, 16, 17, 21 and No.22. Because of the large variation, it was not possible to confirm the change in the carbonation protection effect of the finishing materials over time.

(4) The corrosion depth was smaller in those with multiple layers of finishes, those with low moisture permeability, and those with thick finishes, as well as the carbonation depth.

(5) The corrosion progress of concrete, containing chlorides, is inhibited by the use of finishing materials that inhibit the penetration of oxygen and moisture.

(6) The relationship between the corrosion depth and the carbonation depth is that the carbonation progresses first and then the corrosion progresses.

　We have proposed a characterization method that can reproduce the velocity waveform using a few parameters (characteristic values) while retaining the characteristics of the pulse-like ground motions observed near the epicenter (velocity amplitude, pulse period, wavenumber, etc.). In this paper, based on Sugino et al., which can be characterized easily and objectively, we propose a characterization method that takes the phase into consideration. Then, it was applied to the observation record of 2018 Hualien earthquake in Taiwan, where pulse-like ground motions were observed at high density near the fault, and its effectiveness was shown.

　The results obtained are listed below.

1) The advantage of the proposed method is that it can be applied to pulse-like ground motions with multiple pulse periods, and the velocity waveform and displacement waveform of the observed pulsed ground motion can be roughly reproduced including permanent displacement. (Here, the reproduced waveform is called a characterized pulse.)

2) By drawing the velocity orbit of the horizontal two components by the characterized pulse, it is possible to simplify and analyze the pulse-like ground motions characteristics when the horizontal two components are input.

3) By applying the characterization method to the observed ground motions near the fault of the Taiwan Hualien earthquake where the fault slip occurred, on the east side of the fault where the slip occurred, it was illustrated that (a) Multiple pulse periods tend to occur in the Fault-Parallel component, and (b) The pulse period tends to be larger and the wave number tends to be smaller than the west side.

　By using the characterization method for pulsed ground motions proposed in this paper, it is possible to easily quantify the characteristic values of many observed and predicted ground motions. In the future, if the relationship between the source characteristics, propagation characteristics, amplification characteristics and pulse characteristic values in seismic motion becomes clear quantitatively, the range of pulse characteristic values that should be assumed at the building location can be specified, and it is expected to lead to reduction of damage caused by pulse-like ground motions.

　The consideration of indoor safety during severe earthquake is necessary to minimize casualties in buildings. Most human casualties occur as injuries, such as bruises caused by overturning of furniture. However, injuries from the seismic response of humans, for example, being knocked down by earthquake shaking or the head collision to the wall, also could occur. It is important to consider the human response to the earthquake motions for the evaluation of indoor earthquake safety.

　Several studies have been conducted on human seismic response, and the seismic response analysis model of human bodies was developed based on the shaking table test which input earthquake motions to human subjects. However, the individual differences in human seismic response have not been studied so far. When investigations on human damages under severe earthquake shaking are conducted, person-to-person differences are considered critical.

　The purpose of this study was to identify the individual differences of human seismic response, and to develop the seismic response analysis model of human bodies which can consider the difference. For that purpose, first, a difference in human seismic response was investigated by comparing the shaking table test results of two human subjects. The comparison results showed that there is a difference between two subjects’ seismic response in the existence of residual displacement and the coping strategy to the shaking.

　Then, a novel feedback control method of the seismic response analysis model of human bodies was proposed where a time dependent reference value of the control system was adopted. Then the proposed model succeeded to reproduce the person-to-person difference between two subjects. In addition, response analyses of human bodies were conducted with the developed model inputting various strong motion records. The response analysis results showed that the characteristics of the differences in seismic response between subjects in the shaking table test were also shown in the results of the response analysis, and they vary depending on the frequency characteristics of the input earthquake motion.

　It has been reported that the vibration generated in a live hall located in an urban area during a live performance causes unpleasant vibration to surrounding buildings. This is because the vertical motion performed by a large number of audiences in accordance with a song and music. This vertical motion works as an excitation force to generate vibration and the vibration propagates to surrounding buildings through the ground. As a countermeasure against this vibration problem, we have developed a vibration-proof floor system using inertial mass devices. The feature of this vibration-proof floor system is that the forces of the spring supporting the floor and the inertial mass device arranged in parallel with the spring cancel each other out in opposite phases. Due to this feature, the force transmitted to the ground within a certain frequency range can be greatly reduced.

　In this paper, first, we show the response characteristics of the vibration-proof floor system in the frequency domain. From this response characteristics, it is confirmed that this vibration-proof floor system is effective countermeasure for the vibration problem caused by vertical motion of audiences in a live hall. Next, an excitation test of the principle model of the vibration-proof floor system is reported. Four disc spring units support weight of the model and an inertial mass device composed of a ball screw and a rotational weight is arranged in parallel. In this test, the model was excited with a sinusoidal wave using an excitation device. As a result of this excitation test, it is confirmed that the expected response characteristics for sinusoidal excitation can be realized. In addition, since the disc spring units and the inertial mass device have a slight hysteresis, it is necessary to evaluate this hysteresis damping in order to determine the performance of the vibration-proof floor system.

　The vibration-proof floor system is installed in an actual live hall with standing seats of 1,600. The specifications of the hall, the device configuration and the design performance of the vibration-proof floor system are shown. An excitation test was conducted in which the vibration-proof floor was excited by vertical motion of 1025 people. In this test, the verticals motions to the tempo of 2.0, 2.5 and 3.0 Hz and free verticals motions to the actual song were performed. From the measurement value obtained by the test, the vibration-proof floor behavior at the time of vertical motion was almost as designed. The reaction force response magnification of the vibration-proof floor system applied to the actual hall during the excitation test is obtained through a simulation analysis of the test. In this analysis, the device characteristics are evaluated using bilinear model that expresses friction of devices. The excitation wave used in this analysis is created as a combination of a sinusoidal wave of excitation frequency and twice and three times frequency of that. As a result of this analysis, it was confirmed that the reaction force reduction effect of proposed vibration-proof system is higher than that of the vibration-proof floor without inertial mass device.

　Through this paper, it is proved that the proposed vibration-proof floor system is very effective for the vibration problem caused by vertical motion of audiences during live performance.

　In recent years, time history response analysis has been carried out using a more elaborate three-dimensional model. The dynamic explicit method does not require solving simultaneous equations and is suitable for parallel computing. However, if the damping model is Stiffness-proportional damping and Rayleigh damping, the stability conditions become stricter than the mass-proportional damping, and it is necessary to make the analysis time step finer.

　There are many unclear points about the damping, but a model in which the attenuation ratio is almost constant is desired in terms of engineering.

　Co-authored Nakamura proposed two causal damping models called Causal Hysteretic damping and Extended Rayleigh damping. Compared with the conventional Rayleigh damping, two causal damping models can be analyzed with a damping ratio that is constant over a wide range. However, the formulation and verification of these damping models was based on the dynamic implicit method.

　Therefore, the purpose of this report is to examine the applicability of two causal damping models to the dynamic explicit method. We also confirm that using two causal damping models may improve computational efficiency compared to using a conventional damping model. First, the outline of two causal damping models and the outline of the dynamic explicit method that approximates the velocity term by backward difference are shown. Furthermore, numerical analysis was performed using a simple multi-degree-of-freedom model and a model of shell elements with about 3000 nodes, and the applicability was examined. Since this is the initial stage of research, we examined its applicability in elastic linear problems.

　As a result, followings were obtained.

1) Using two causal damping models, it was confirmed that the dynamic explicit method exhibits the same performance as the dynamic implicit method by a simple multi-degree-of-freedom model.

2) Two causal damping models could be applied to the dynamic explicit method, which approximates the velocity term by backward difference, by the model of shell elements with about 3000 nodes. In addition, it was confirmed by comparing the two causal damping models and the mode damping that the damping ratio can be evaluated constant and accurately over a wide frequency range.

3) Two causal damping models and conventional Rayleigh damping require the velocity term to be solved by dynamic explicit method that approximates the backward difference. It was revealed that two causal damping models may have a larger analysis time step than the conventional Rayleigh damping model.

　Simulation analysis is an effective method for assessing building safety after an earthquake. This type of analysis is effective for evaluating foundations or piles, of which the safety cannot be confirmed by a visual check.

　During such analyses, the input motion should be set to reproduce the observed waves at given points. In these cases, the modal iterative error correction method can be used to set the input motion, even if the analysis model exhibits nonlinear behavior. However, the conventional method requires considerable calculation time, especially when the duration of the input motion is long.

　This study proposes a new execution method for faster inversion. The outline and details of the method are described, and the results of the experiments are presented to confirm the accuracy of the method.

　The main findings of this study are as follows:

(1) This study proposed a new input motion inversion method to determine the input motion gradually from the front by performing partial inversions repeatedly instead of the entire duration inversion at once. The proposed method can reduce the computational cost of executing the perturbation analyses for the perturbation impulse response matrix and generating the generalized inverse matrix of the perturbation impulse response matrix.

(2) The effectiveness of the proposed method is confirmed using examples. It was confirmed that by the proposed method, the calculation cost required for inversion could be reduced to approximately 40% that of the conventional method. Furthermore, it was confirmed that the precision of the reproduced time history of the ground surface acceleration and the precision of the identified input seismic motion could be kept.

(3) It was confirmed that further reduction in execution time is possible by shortening the overlapping time, resulting in an execution time as low as approximately 1/4 that of the conventional method. However, because sufficient reproduction accuracy is not obtained when the overlapping time is eliminated, it is necessary to set an appropriate overlapping time considering the delay time and nonlinearity of the analytical model.

Recently, from view－point of Global Environment, timber, i.e., one of nature-cycle materials, is being tried to be utilized as structural members of large timber buildings in Europe and North America.

A representative timber of the members is Cross-laminated timber (CLT), however, CLT structural system very often restricts planning of building because of CLT being plate member. High-stiffness-strength-timber slender beam and column are significantly desired.

This study focuses a hybrid glulam timber beam with steel deformed bar (rebar) and Epoxy resin adhesive. The aim of this study is to clarify mechanism of suppressing bending creep of the hybrid beam by rebars and to develop method of estimating its relative creep coefficient in design. This paper reported creep behavior of hybrid glulam timber beams with a small size under 4-year term loading and additional coated beams under 1.5-year term loading. The coated beam specimens were prepared in order to investigate adverse effects of moisture on shear stiffness of wood around the rebars. The result and discussion of the two experiments are summarized as follows:

(ⅰ) Creep of curvature of the hybrid beam rapidly boosted against prediction based on assumption of Navier hypothesis, during same term in every year when humidity increases in Japan, i.e., July-September, as Fig.11.

(ⅱ) The boost results from position of rebar-embedded lamina in the beam section because moisture in air around beam percolates surface of its timber and decreases shear stiffness of wood part around rebars. One of basic ways to protect the decrease is to arrange the rebar-embedded lamina as the second layer from outer layer because adhesive layers between the outer layer and the second, as seen in Fig. 20, prevent the moisture from percolating the second layer , i.e., rebar lamina.

(ⅲ) This problem may occur in the small-sized beam, however, the real full-sized hybrid beam can readily prevent it in some ways as shown in Fig.20(b),(c), by using waterproof coating, burning marginal later, or refractory coating.

(ⅳ) The protected hybrid beam is estimated to significantly suppress its bending creep by rebars, as Relative creep coefficient C_φ of curvature in 50-year for design listed in Table 8.

(ⅴ) Deflection-stiffness of timber beam and hybrid beam rarely deteriorated during creeping, as Fig.25.

(ⅵ) A prediction method for relative creep fluctuating intensely owing to seasonal influence was proposed by using two - tangent lines on logarithmic chart, as Fig.21, Fig.22, and Fig.23.

　Geopolymer concrete has the potential to be an eco-friendly alternative to cement concrete because CO₂ emissions in its production process can be reduced and industrial by-products such as fly-ash can be used effectively.

　The present study investigates flexural behavior of fly-ash based geopolymer concrete beams before yielding and its applicability as a flexural member. Eight beams consisting of four fly-ash based geopolymer concrete beams and four ordinary Portland cement concrete beams were cast and tested under four-point flexural loading. The experimental variables are combinations of reinforcing bar diameter and number of reinforcing bars, compressive strength and concrete types. In addition, non-linear three-dimensional finite element analyses were carried out.

　Bending moment-deflection behavior, crack width, crack space and number of cracks were examined. The experimental results were compared with the proposed equations provided by the AIJ design standard for reinforced concrete. The deflection and maximum crack width of the fly-ash based geopolymer concrete beams subjected to bending moment based on long/short-term allowable stress were similar to those of ordinary Portland cement concrete beams, and, therefore, they have an applicability as flexural members. The flexural behavior of fly-ash based geopolymer concrete beams prior to flexural yielding, except for the yielding deflection, could be predicted using the AIJ design standard. Making modifications to the coefficients related to the Young’s modulus ratio improved the accuracy of the prediction of yielding deflection.

　Although Young’s modulus of fly-ash based geopolymer concrete was lower than that of ordinary Portland cement concrete, the post-cracking stiffness of fly-ash based geopolymer concrete beams were not largely different from that of ordinary Portland cement concrete beams. The results suggested that the effect of tensile strength on the flexural behavior of the beams was greater than that of Young’s modulus of concrete.

　In addition, the crack width of the fly-ash based geopolymer concrete beams was similar to that of the ordinary Portland cement concrete beams, and could be predicted by the AIJ design standard for reinforced concrete.

　The steel plate shear wall is one of the seismic resistance devices which is installed in the flame of building to improve the horizontal stiffness and strength. Shear force is loaded on the steel plate shear wall when the horizontal force acts the building. In order to prevent the overall buckling by shear force, some stiffeners are welded on the shear wall in some types of arrangements. Therefore, the design of stiffeners is very important to prevent the strength deterioration and to maintain the restoring force.

　In this paper, focus is on the stiffeners of one-sided stiffening type. Authors have proposed the new rational design method of the stiffeners arranged in one-side of the wall, based on that of the stiffeners arranged in front and back of the wall; cross stiffening type in our previous study. The purpose of this study is to directly estimate the plastic buckling strength and the plastic deformation capacity by using the width-thickness ratio β normalized by yield strength and elastic buckling coefficient. The relationship between β and plastic deformation capacity is formulated based on the results of FEM analysis, and required values of β are defined for three target performances; Grade I, II and III. In addition, the effects of the cyclic loading and the surrounding frame members are revealed by experimental tests.

　Results are summarized as follows:

　1) The plastic buckling strength and the plastic deformation capacity can be evaluated based on the estimation formulas, which are same as that of cross stiffening type.

　2) The specifications of the stiffeners of one-sided stiffening type are compared with those of cross stiffening type. The more rational stiffening type is different from each other depending on the aspect ratios of the wall and the target performances.

　3) In the experimental tests, the maximum value of the deformation angle on the skeleton curve has strong correlation with the β of the specimen, which is not depending on the stiffening type. It indicates that the plastic deformation capacity of one-sided stiffening type can be evaluated based on β, which are same as that of cross stiffening type.

　4) In the experimental tests of specimens which are evaluated as Grade I and II, the effect of the cyclic loading on the plastic deformation performance is small. Although the proposed estimation formula is based on the results of FEM analysis by the monotonic loading, it can evaluate the plastic deformation capacity conservatively in the case of Grade I and II.

　5) The effect of the axial loading on the plastic deformation performance is small in the experimental tests, when the strength of the surrounding frame members is larger than the shear yielding of the wall. It is because the surrounding frame members have the rigidity against the axial loading.

　It is difficult to construct a rigid frame structure without seismic walls and bracing because of the difficulty in ensuring the strength and rigidity of the joints. In the previous paper, dog bone timber-steel hybrid connections, which consist of a dog bone part and a timber-steel hybrid connection that transfers the bending moment of the laminated timber beam to the steel, is proposed. The proposed connection has enabled the realization of a structure with sufficient strength, stiffness and toughness, but the deformation capacity of the connection has not been studied for another cross-sectional dimensions. In this study, based on the previous evaluation equations, an analytical method to derive the history of the M-q relationship by increasing the angle q of the core of the laminated wood to the H-shaped steel of the wood-steel joint, taking the shear force and a single neutral axis calculation were performed to obtain the rotational stiffness Kw, the yielding bending moment My, and the ultimate bending moment Mu of the timber-steel hybrid connection. The validity of the shear force evaluation method and the effect of shear force on the hysteresis properties of the timber-steel hybrid connection are discussed. Four different test specimens (MD600, MD360, CD360 and MS360) are fabricated for the proposed connection with both flanges fixed by lag screws to increase the stiffness and strength of the wood-steel joints. The MD600 is 600 mm height of glued wood and has two bodies with monotonic loading, and the MD360 is 360 mm height of glued wood and has two bodies with monotonic loading. The CD360 is 360 mm height of glued wood and has three bodies with cyclic loading. The MS360 is strength tests of timber-steel hybrid connections, which is 360 mm height in glued wood and has three bodies with monotonic loading. From the joint bending tests, the hysteresis properties and plastic deformation capacity of the proposed connections with dog bone are verified, while the stiffness and holding strength of the timber-steel hybrid connection are verified. As results, the following conclusions are obtained:

1) In the joint bending tests, it was confirmed that the dog-bone part plasticized while wood part is in elastic range for all the test specimens with dog-bones (MD600, MD360, CD360), and that the dog-bone portion was stable and had an excellent plastic deformation capacity exceeding the plasticity ratio of 5. In the MS360 test case, the strength of the wood joint was superior to that of the dog-bone in both stiffness and flexural strength.

2) A new evaluation method that takes into account the shear forces on the wooden joints is proposed and it is confirmed that the proposed method can roughly reproduce the hysteresis properties of the joints. It was confirmed that the proposed method was able to simulate the historical properties of the joints in general, although the effect of shear forces was not so great and the flexural strength of the joints could be reduced by shear forces.

　In steel roof gymnasiums with RC substructures, the out-of-plane response of Cantilevered RC walls is predominant during seismic responses, which triggers sequential damages of structural or non-structural components. Particularly, in 2016 Kumamoto earthquake, structural damages related to space frames (i.e., member buckling, member local buckling, post-buckling ductile fracture and connection bolt fracture) were reported, which suggest structural engineer may be required to perform a non-linear numerical simulation of the spatial roof to ensure the seismic performance of similar gymnasium used as shelter in the future. In this paper, a member global buckling strength estimation of spatial truss roof with semi-rigid joints is proposed according to both the results of the connection bending tests of spatial roof truss joints. The proposed estimation method is introduced to a parameter formulation for the phenomenological constitutive model simulating yield strength increment by cold-forming, post-buckling behavior and post-buckling ductile fracture. The accuracy is validated against static unit-truss element test and dynamic shaking-table test, including truss member buckling and member fracture.

　In summary, the following results were obtained:

1) A simple equation is proposed to evaluate the rotational stiffness of a semi-rigid joint based on coupler and bolt specifications. The proposed method applies to arbitrary coupler and bolt specifications.

2) The global buckling strength estimation equation for system truss member considering two types of joint rotational stiffness depending on the rotational constraints of the spherical nodes are proposed. The proposed evaluation method was predicted with an error of -5.7 to +2.3 kN in quasi-static cyclic loading tests of the system truss unit and an error of -3.4 to +5.7 kN in shaking-table test results of a reduced partial model of a gymnasium.

3) The proposed global buckling strength evaluation method and the modified Shibata-Wakabayashi phenomenological buckling hysteresis model considering member fracture can capture the relationship between the global buckling strength and load deformation of system truss members for both static and dynamic inputs, and member rupture can be conservatively evaluated.

　Mid-story pin column system has been proposed to realize beam yielding mechanism of a steel moment-resisting frame. In this system, a pin connection system is applied inbetween the upper steel column and the lower RC column extended from the base beam to control moment demands on both ends of the first story column. An exposed-type column base has been adopted as a pin connection connected by a single anchor bolt. Static cyclic horizontal loading tests of a partial frame consisting of the first story column and the second floor beams showed a frame with the proposed column base system exhibited stable hysteresis performance up to a story drift angle of 0.05 rad, remaining column in elastic. In addition, the RC column with the connection of a single anchor bolt exhibited slippage at the connection while this problem was solved when an anchor bolt was stiffened by cross-shaped four shear plates. A cover plate was also welded on the top of the shear plates to cover the top surface of the RC column from bearing pressure of the upper steel column. An anchor bolt with the welded plates are called an anchor bolt set, or ABS, hereinafter. Ensuring a reliable seismic performance of an upper steel frame, damages at the proposed connection have to be minimized. So far, compressive stress transfer mechanism was clarified, and the ultimate compressive strength of the connection was evaluated in the previous study.

　This paper discusses shear stress transfer mechanism and the ultimate shear strength of the RC column with proposed column base connection with the ABS. An anchorage rebar was welded on each shear plates to anchor the ABS to the RC column until the connection reaches the ultimate strength. Series of static cycle horizontal loading tests were conducted. Test parameters are a diameter of anchorage rebars and stirrup arrangement. Test results showed that anchorage rebars increased the shear strength twice as large as the connection by ABS without anchorage rebars, mitigating bearing crack propagation spread from the ABS (Chapter 2).

　Chapter 3 shows that Navier hypothesis is applicable to strain distribution of a group of anchorage rebars, and that their neutral axis can be estimated as the gravity center of the semicircle of the cover plate under uniform pressure. A formulation of the yielding moment of group of anchorage rebars is proposed based on these findings.

　Chapter 4 discusses the bearing crack initiation strength and the ultimate shear strength of the RC section where the ABS is embedded. With a combination of lever action by the embedded ABS and reaction moment retained by a group of anchorage rebars, shear force sustained by the RC section with the embedded ABS, or Q_RC, is amplified or reduced from the horizontal force P acting on the connection. The shear crack initiation strength Q_RC’,_c and the ultimate shear strength Q_RC’,_umean sufficiently agrees with test results, calculated by AIJ RC design standard with the effective width of the RC section considering the width of the ABS.

　Chapter 5 proposes the stress transmission mechanism of the connection based on that for the embedded-type column base system. The shear strength estimation formulas for the RC connection with the ABS is proposed by a superposed strength method considering the effect of reaction moment provided by a group of anchorage rebars.

1. Introduction

The objective of this study is to examine the properties of the stability index SI, and the relations among the effective length factors obtained from different buckling conditions. Overview of the relations among the stability indexes (SI and _GSI) and the effective length factors (γ_exact, _SIγ, _SIγ *, _Gγ, _Gγ˜ and _Gγ˜* ) is shown in Fig. 1. The targets are rectangular 2-span plane frames, and total 96 frames are examined, taking the condition of the column base, number of story, stiffness ratio, stiffness ratio reduction factor and distribution of the story shear as the parameters.

2. Stability index and effective length factors

Stability indexes SI and _GSI are calculated by Eqs. (1) and (12). For the overall structural system, the effective length factors _SIγ and _Gγ˜ are calculated by Eqs. (5) and (17), respectively. For the buckling of the story, the effective length factors _SIγ* and _Gγ˜* are calculated by Eqs.(9) and (19), respectively. The exact effective length factors γ_exact are calculated by the buckling slope-deflection method or the finite element method.

3. Results

Effects of story shear distribution and column base are shown in Figs. 6, 7 and Fig. 8, respectively. Comparison between the stability index SI obtained by the first-order lateral load elastic analysis and approximate _GSI obtained by G factor is shown in Figs. 9, 10, 11, and 12. Comparison of effective length factor between _SIγ and γ_exact is shown in Fig. 13, and comparison of effective length factor between _Gγ˜ and γ_exact is shown in Fig. 14.

4. Conclusions

The conclusions derived from this study are as follows:

1) The effects of the story shear distribution on the stability index are hardly observed (see Figs. 6 and 7). For frames with four stories or more, the stability index above the 4th story is almost same among the fixed base and pin base frame (see Fig. 8).

2) In the intermediate-stories, the approximate stability index _GSI obtained by G factor can be applicable as the stability index (see Figs. 9 10, 11, and 12).

3) The buckling-associated-story obtained from stability index _GSI agrees well with that obtained from stability index SI (see Table 3).

4) Effective length factors obtained by using the stability index agree fairly well with the exact factors about within 5% error (see Fig. 13).

5) The effective length factors _Gγ˜ estimate the exact values γ_exact up to 1.12 times (fixed base) and 1.2 times (pin base) (see Fig. 14).

6) In the intermediate-stories, the effective length factors _Gγ˜* can estimate the factors _SIγ*.

　This study is devoted to clarifying the yielding behaviors of twist-formed or pretwisted bars having a narrow rectangular cross section subjected to axial compressive loading. In the first half of this paper, theoretical approaches provide explicit formulae for calculating the first yield and fully yield loads on the condition that the bars are free from buckling. The theoretical solutions are based on elastic and plastic distributions of axial and shearing stresses in the cross section and also on the Mises yield criterion. The theory suggests the ultimate behavior is governed by the plastic collapse in the mode of simultaneous development of twisting and contraction flows. In the latter half of this paper, a series of experimental tests of twist-formed steel bars clarifies the validity of the theoretical solutions and also revealed the predominant failure modes as follows: bending failure is prevailing for small pretwist, which is attributed to flexural instability, while twisting-contraction combined failure is prevailing for moderate to large pretwist as predicted by the theory of plasticity and this combined failure mode can be accompanied by overall flexural deformation due to initial crookedness.

　The distribution of stresses at the onset of yielding may be worthy of remark in relation to the issue of buckling. The compressive axial force induces a torque in the pretwisted bar, and consequently shearing stresses as well as axial stresses act on the cross section. The shearing stresses are uniformly distributed in the width and largest on the surface, but the compressive stresses are parabolically distributed with a maximum at the mid-width. Thus, the yielding starts at the mid-width on the surface and the elastic limit load is much smaller than that of an ordinary untwisted bar in which uniform axial stresses and zero shearing stress act on the cross section. Unfortunately most of or possibly all of the researchers involved in the buckling of pretwisted bars have been unaware of this matter.

　One of the most important and fundamental function of every building is to be a “shelter” for people inside, protecting them from various external disturbances. However once accidental fall of architectural finish suspended at a height, such as ceiling boards, happens, the building cannot keep people in the room safe anymore and the function of a “shelter” will be completely spoiled.

　There are various causes of the failure of ceilings, such as dew condensation, water leak, earthquakes, wind, various vibration, rusting, deterioration, thermal deformation, and invasion of living things into ceiling space. Earthquake is just one of them. It is obvious that the main force that drags down non-structural components is the gravity and the acceleration of the gravity is almost unique everywhere on the earth. Since the phenomena is reproducible, it is possible to define the safety of each ceiling board according to its material and suspended height.

　In this paper, the authors propose the quantitative evaluation procedure of the safety of ceiling boards, studying the knowledge of the human tolerance developed in the field of injury biomechanics.

　In the first and second part, the authors show their review of studies about human tolerance index with special attention to how to assess the damage to a head under the impact load and explain applicability of the threshold proposed by Nahum et al. to the ceiling problem.

　In the third part, they report the set-up and results of experiments in which they dropped many typical ceiling boards to a dummy head and recorded impact load to draw a diagram of the relationships between the maximum impact load and drop-height. The results show that the higher the drop-height is, the higher the impact speed and load become; the impact load varies very much according to the material, the lighter the material is, the faster it converges; and for brittle material, failure mode gradually becomes punching type as the drop-height goes up.

　Providing these results, in the fourth part, the authors propose a new criterion to assess the safety of ceiling boards, by comparing the impact load of the board with the threshold of linear fracture of a skull (equivalent to Abbreviated Injury Scale (AIS) 2) proposed by Nahum et al. They also give some cautions for the application of the criterion. Moreover, they point out the possible influence of the natural frequency of the set up system to the results of experiments. This may not affect the general stream of the study and detailed investigation and discussion will be given in the following paper, part 2.

　Full-scale fire resistance experiments for an axially loaded compressive steel column including high strength bolted connection were conducted under steady state condition ( the constant temperature and increasing load condition), to examine the load-bearing capacity and the possibility of shear fracture of high strength bolts after the column specimen exhibited the overall buckling. The specimen was fabricated from the JIS SS400 (wide flange shape and splice plate) and JIS F10T (high strength bolt). The column connection in the specimen was designed based on "AIJ Recommendation for Design of Connection in Steel Structures". The number of flange bolts arranged in the column connection was used as the experimental parameter. The three specimens without fire proofing materials and the specimen protected by that except for the connection part were prepared, respectively. The column specimen was heated above 700 °C by a gas furnace, and the axial force was applied to that under the condition when the specimen temperature is approximately constant. All of the specimens exhibited the stable residual strength after the buckling (the unprotected column specimen) or local buckling (the protected column specimen). It is well known that the high strength bolt possesses the lower strength than the mild steel at elevated temperature, however, the bolted connection in the specimen did not fracture. This is because the loaded axial force was transmitted through the metal touch between the column cross sections at the column connection, therefore the bolts were not sheared by that. On the other hand, the column connection was subjected to the bending moment owing to the overall buckling behavior, however, the bending moment reduced by the local buckling occurred in the center of column specimen and the fracture of column connection was avoided. In particular, the column connection designed by using the small numbers of the flange bolts, which possessed the lower bending strength than the required design strength at the ambient temperature, did not fracture and exhibited the stable residual strength after the overall buckling. It is possible to conclude that there is little possibility of bolt-fracture for the column connection at the fire based on the experimental results. The maximum axial force of the column specimen could be evaluated by the design value of overall buckling and the yield strength considering the effect of steel strain rate at the elevated temperatures.

　On the other hand, it is necessary to conduct the composite loading experiments at the elevated temperatures, regarding the column connection subjected to both axial force and bending moment, to examine the possibility of bolt-fracture owing to the bending moment. Furthermore, the evaluation of bending strength involving the fracture of bolted connection for the wide flange shape in the weak axis, which has not been clarified in the fire resistance field of steel structures, is required in the future research.