The paper presents a numerical, simulation-based approach to investigate th
e stability of torsional motion of slender suspension bridges under stochas
tic wind turbulence. The torsional bridge motion is represented by a linear
, single degree of freedom oscillator. Stochastic turbulence in wind veloci
ty is considered in the form of a periodic excitation with random phase mod
ulation. The stability condition refers to the asymptotic sample stability,
for which the necessary and sufficient condition is that the largest Lyapu
nov exponent be negative. Monte Carlo simulation is performed to evaluate t
he largest Lyapunov exponent, and stochastic differential equations of moti
on are integrated in polar coordinates using Euler's scheme. Unlike earlier
analytical approximations, a quadratic noise term is retained in the prese
nt analysis. The turbulence intensity is shown to have a small stabilizing
effect on the bridge stability in a sense that an increase in the turbulenc
e intensity moderately increases the critical mean wind velocity beyond the
deterministic flutter velocity. The stabilization effect is limited to the
case of narrowband detuned excitation. In the proximity of the parametric
resonance frequency, an increase in the bandwidth of the excitation process
tends to stabilize the bridge motion. (C) 1999 Elsevier Science Ltd. All r
ights reserved.