リンク式流体慣性ダンパによる層間変形制御機構の動力学特性に関する研究

抄録

 This study concerns an analytical method to investigate the dynamic characteristics of a Linked Fluid Inertia Mass Damper (LFIMD). The damper consists of two cylinders, piston rods, and link tubes. Two piston rods are linked and move following the oil flow in the link tubes connecting two oil chambers. It can change the stroke ratio of the two piston rods to arbitrary values by adjusting the cross-sectional area of the linked oil chamber. In this study, we use a hydraulic link mechanism as a story deflection control system. This mechanism can adjust the story deflection ratio when installed across the upper and lower stories.
 In the first part, we study the influence of the link stiffness on the dynamic characteristics of a two-story structure. First, we show that the eigenvectors converge to arbitrary values by installing a link mechanism with sufficiently large link stiffness. In addition, the link mechanism does not change the natural period in the 1st mode regardless of link stiffness, but the second mode natural period becomes shorter. Therefore, in the case of the link stiffness is sufficiently high and the story deflection is linked, we can calculate the response of the entire structural system without considering the influence of the 2nd mode. Next, we derive the theoretical solution for harmonic oscillation. The theoretical solution shows that the dynamic characteristics of the link mechanism depends on the apparent total stiffness that is different from the real stiffness of the main structure. Moreover, the apparent stiffness of the story changes when the piston displacement ratio of the linked damper is changed. The apparent stiffness exhibited by the linked damper is the opposite sign and same absolute value in the upper and lower stories. The additional stiffness of the link mechanism is determined based on the transfer relationship of the shear forces between the linked stories, and we call it the redistribution of shear forces by link mechanism. Basically, the link mechanism redistributes the shear force from the relatively strong layer to relatively weak layer, resulting in the link mechanism adding the apparent positive stiffness to the weak layer and the negative stiffness to the strong layer. By organizing the theoretical solution and the generalized eigenvalue program, if the eigenvalues of the original structure are known, it is possible to estimate the redistribution of shear forces without any time history response analyses.
 Finally, we conducted shaking table test using specimens with different stiffness distributions along the upper and lower stories. We used two models; the first one has nearly uniform deflection in each story (Specimen A), the second one is so designed as to show deflection concentration the lower story (Specimen B). From the redistribution method for shear forces, it was predicted that the redistribution of shear forces in specimen A was small, and the shear forces were redistributed from upper layer to lower layer in specimen B. The test results showed that the story deflection was as expected in the case of the specimen without the damper and that it is quite easy to make the story deflection uniform by using the link mechanism. Next, we evaluated the apparent stiffness, apart from the effects of the inertia mass and the link mechanism, exerted by the damper, using a dynamic model constructed based on experimental results. It is shown that the negative stiffness by inertia mass effect is constant regardless of the stiffness distribution, and that the positive or negative stiffness by link mechanism is relatively high and dominant. We confirmed the validity of the theoretical assumed dynamic characteristics of the link mechanism.