Probability Based Determination of Slab Thickness for Residential Apartment Building to Floor Vibration

Abstract

Floor vibrations in residential buildings are typically induced by the movements of jumping and walking of occupants, which often annoy other occupants. Such floor vibration can be controlled by a proper floor design rather than using additional vibration reduction devices. Current design codes (ACI 318M-08, KCI-M-07-003) specify slab thickness that satisfies serviceability requirement for deflection. Even though slab thickness satisfies the requirement for deflection, it cannot guarantee its satisfactory performance against occupant induced floor vibration. To help control floor vibrations, several foreign criteria have been developed mostly based on human perception tests. The main objective of this study is to develop the floor vibration criteria for domestic residential apartment buildings. For the propose this study performed (1) the experimental investigation for the acceleration thresholds of appropriate floor vibration due to heel drop impacts and walking activities to provide vibration limit, (2) the evaluation of the serviceability to floor vibration for domestic residential apartment building to define dynamic characters of floor structure and possible problems with floor vibration, and (3) the probabilistic simulation analysis to determine the thickness of flat slabs that satisfies both deflection and occupant induced floor vibration serviceability. Firstly, shaking table tests were conducted simulating heel impact induced- and walking induced-floor vibrations for various combinations of frequencies, damping ratios and acceleration amplitudes. Four different perception levels are considered appropriate criteria for Korean residential buildings; imperceptible, slightly perceptible, distinctly perceptible, and strongly perceptible. Twenty Korean test subjects were used. It is observed that test subjects are more sensitive to walking-induced vibrations than those induced by heel impacts having the same peak accelerations except for slightly perceptible level, in which acceleration threshold does not vary with respect to heel drop and walking induced vibrations. This study also shows that the acceleration thresholds proposed as a result of this study using Korean testing subjects are lower than those proposed by other researchers in different countries. Next, this study conducted floor vibration test using heel drop impact and evaluated vibration serviceability of conventional floor in apartment building. Heel drop test was also conducted to define load-time function of heel drop impact. Finally this study proposed the thickness of flat plate in residential apartment building with bearing wall that satisfies occupant induced floor vibration serviceability. Since there are large uncertainties associated with heel drop impact load, slab weight, modulus of elasticity, and damping ratios, a probabilistic method was used to propose the slab thickness. To compute probability considering multiple random variables having different distribution functions, this study adopted Monte-Carlo simulation. To reduce the computational difficulties associated with the Monte-Carlo simulation, an equivalent single degree of freedom (ESDF) system was used to estimate the peak acceleration demand due to occupant induced floor vibration.