A series of studies exploring major factors influencing1 earthquake disasters from direct to indirect, immediate countermeasures, and restoration processes is made for all municipalities damaged in the 1982 Urakawa-oki earthquake, Hokkaido, Japan. A continuation of the factor analysis in the previous paper on earthquake disaster occurrence, this paper proceeds deals with the post-earthquake countermeasures and subsequent restoration processes, so as to elucidate how the administrative organization of municipalities were actually involved. The major results obtained are summarized as follows. i) The number of municipalities in which post-earthquake countermeasures were carried out increases with increasing seismic intensity from IV to VI in the JMA scale. The seismic intensity as an index of the severity of earthquake input seems to control the priority in the adoption and execution sequence of countermeasures, such as emergency calls for municipality staff, orders for evacuation of the affected areas, setup of the headquarters, and supply of water and food to the affected population. ii) The time length necessary for the restoration of the physical damage to the structures of residential houses and commercial buildings is well correlated with the degree of the individual damage. As is expected, their restoration duration becomes longer with increasing seismic intensity. Contrary to this, in life line functional systems such as electricity and water supply the necessary time for restoration is affected not only by the seismic intensity but also by the actual number of damaged facilities in the area, because they are repaired under a restriction of a limited number of available laborers. From this series of studies it is clear that the earthquake disasters in city areas are much more complicated than those in less populated town and village areas and therefore the restoration in city areas takes a longer time, it should be kept in mind that the sophisticated functional systems convenient in everyday life in city areas make earthquake damage complicated and their recovery prolonged.

View full abstract

The response analysis of containment buildings in nuclear power plants which are composed of reinforced concrete shear walls was made. The analytical model is a two-mass system which consists of a slab mat with ground springs and a superstructure reduced into a single-mass system. The energy input to the superstructure was estimated and was expressed in a simple empirical formula.

View full abstract

This paper presents the formulas for calculating the load-displacement relation of reinforced concrete reactor buildings based on the results obtained from load tests using reduced scale model tests. Since the second half of the 1970s, from the viewpoint of the rationalization and economy of the designs of reactor buildings, evaluation of building safety was tried to confirm that the response value of the reactor building has sufficient margin for the final stage. For that purpose, the properties of the building up to final stage should be understood. The relation between the lateral load and displacement is an important measure to understand these properties. In dynamic response analysis of earthquake-resistant design, the relation between lateral load and lateral displacement are idealized by envelope curves and hysteresis loops. Lateral displacement consists of shear and bending components. Shear displacement is obtained from the proposed relation between shear stress τ and shear strain γ, and the bending displacement is obtained from the relation between moment M and curvature φ. Formulas for calculating the τ〜γ envelope curves were proposed in Part 1 and those for calculating the M 〜 φ envelope curves, in Part 2. This paper describes the algorithms for constructing hysteresis loops using third order and multi-linear functions. Under cyclic load at identical displacement to that previously experienced, load-displacement plots will be stable loops. As Ogawa proposed, normalized stable loops can be idealized with a pair of third order functions (Ref. 1). The coefficients of the third order functions for shear displacement and bending displacement are determined from the experimental data. When the velocity changes from negative to positive, for example, the trace forms a branced curves. The function for forming the branched curve are also proposed. As the third order function model was not practical for response analysis of multi-degree-of-freedom systems, the version of multi-linear model was prepared. . Lateral load-lateral displacement curves which were obtained from load tests were compared with those of the individual specimens obtained from the proposed formulas. The computed curves show good agreement with the experimental ones. (Ref. 1; Ogawa, J., et al., "Experimental Study on Dynamic Characteristics of R/C Frame", Summaries of Technical Papers of AIJ 1968 (in Japanese))

View full abstract

The objective of this research is to make clear the elasto-plastic flexural deformation behavior, fracture mode and fatigue fracture criteria under cyclic bending of steel profile encased reinforced concrete beam-columns experimentally. Tests are carried out under alternately repeated cyclic bending with various constant axial loads. Fracture modes are different according to the axial load levels and deflection amplitudes. In case of lower axial load level N = 0, 1/6 N_0, fractures are caused by the tear off of flange plates for all deflection amplitudes (fracture type F). On the contrary, in case of higher axial load level AT = 2/3 N_0, fracture are caused by the reduction of moment resistance due to crashing down of concrete and the occurrences of the local buckling at the flange plates for all deflection amplitudes (fracture type B). In case of medium axial load level N = 1/3 N_0, the fracture mode is type B for larger deflection amplitudes δ_a=±30mm, ±20 mm and combined type B and F for smaller deflection amplitude δ_a= ±15 mm. All specimen except N = 0N_0 lose lateral and axial resistances at the fracture. The relationship between the deflection amplitudes and the cycles until fracture in Fig. 13 shows very clearly the approximately linear correlation in log-log scale and the influences of axial load levels and deflection amplitudes upon the low cycle fatigue limits of steel profile encased reinforced concrete beam-columns under cyclic bending.

View full abstract

The authors have presented Ultimate Response Analysis by which one can evaluate earthquake responses of structures according to the principle of the ultimate energy equilibrium. In this analysis, response behaviours of structures are classified into the following two extreme typical cases, i.e., monotonic and, cyclic responses. The former is calculated by Pulse Response Analysis (P. R. A. ) and the latter by Finite Resonance Response Analysis (F. R. R. A. ). In this report, an evaluation method of cumulative damages of hysteretic structures by the F. R. R. A. is proposed from the comparison of the results of the F. R. R. A. with those of usual time history dynamic earthquake response analysis (E. R. A. ). The followings are concluded I (1) The time derivative of cumulative damages of the F. R. R. A. (d_<fγ>) is nearly equal or larger than that of the E. R. A. (d_e). This fact means that the evaluation of cumulative damages by the F. R. R. A. is safe side evaluation. (2) The damage duration for F. R. R. A. (t^p_<fγ>) is given by Eq. (29) as product of duration ratio (t^p) and duration of strong motion (t_<06g>)And the number of cyclic response times by F. R. R. A. (n_c) is given by Eq. (30) as division of damage duration (t^p_<fγ> by the time (T_<eq>-T_e). (3) The damage factor by F. R. R. A. (Dfr) is given by Eq. (31) as product of number of cyclic response times (n_c) and damage factor by one cycle (ΔD_<fγ>). In this research, the authors have presented a method of the quantitative evaluation of cumulative damages by F.R.R.A.

View full abstract

This paper deals with the nonlinear random response of single-degree-of-freedom systems having slip-type hysteretic restoring force characteristics. The ground motions are assumed to have the power spectra proposed by Tajimi. The attention is focussed on the time change of the expectation of equivalent natural frequency ratio β, the expectation of cumulative ductility factor λ^^~ and its .standard deviation σ_λ. The approximate solutions for β, λ^^~ and σ_λ are derived on the basis of the theoretical investigation. They are represented in terms of four nondimensional parameters-h_g which appears in the input spectrum, ν which is the ratio of natural frequency in the elastic range ω_o to ω_g which appears in the input spectrum, the nondimensional input intensity ξ and the non-dimensional time τ. β-τ, λ^^~-τ and σ_λ-τ relations with parameters ξ and ν reasonably reflect the nonlinearity of the hysteretic system and the property of the input spectrum. The analytical results indicate that in the inelastic case the spectral value at the initial natural frequency plays a less important role than in the elastic case. The approximate solutions are compared with the digital estimates obtained from the Monte Carlo simulation. The agreements between the both are satisfactory over the wide ranges of related parameters.

View full abstract

The elastic-plastic behavior of beam to column connections may be explained by assuming that the shearing resistance of the panel zone is expressed as an average shearing stress and that the resistance of beam to column connection is expressed as the sume of the bending moment at four corner points of the frame surrounding the panel zone composed of column flange and horizontal stiffene'r and the resistance of the panel. However, to establish a practical evaluation which can always give good agreement with experimental results, it is nessesary to take the influence of the stiffenesses of beam and column menbers surrounding the beam to column connection and the residual stress in panel zone into account. In this paper, it was attempted to estimate these infulences, and the obtained values were compared with test results. As a result of this examination, a practical evaluation method which is simple and gives good agreement with experimental results was obtained.

View full abstract

An exact second-order theory is formulated for predicting the slow drift excitation forces on moored vessels in random seas. It is based upon direct integration of the hydrodynamic pressure on the submerged body surface in conjunction with the consistent perturbation expansion to second order in wave steepness. Green's second identity and Haskind's reciprocal relations are used to derive a formula for the second-order exciting forces due to the second-order waves. This permits the evaluation of the slow drift forces only from the solution to the first-order diffraction problem. As application of the theory, results for the slow drift forces on the semi-submersible are presented, which are evaluated by means of the source-distribution numerical technique. The results based on the present exact theory are compared with the solutions from different simplified approaches. It is concluded that Newman's method using data of mean drift forces in regular waves gives accurate results for the slow drift forces in short to moderate seas, but tends to underestimate the forces in extreme sea states with longer mean wave period.

View full abstract