Random vibration of hysteretic systems under Poisson white noise excitations

Abstract Hysteresis widely exists in civil structures, and dissipates the mechanical energy of systems. Research on random vibration hysteretic systems, however, is still insufficient, particularly when excitation non-Gaussian. In this paper, radial basis function (RBF) neural network (RBF-NN) method adopted as a numerical to investigate Bouc-Wen system under Poisson white noise excitations. The solution reduced generalized Fokker-Planck-Kolmogorov (GFPK) equation expressed terms RBF-NNs with Gaussian activation functions, whose weights are determined by minimizing loss GFPK residual constraint associated normalization condition. A steel fiber reinforced ceramsite concrete (SFRCC) column loaded studied an example illustrate process. effects several important parameters both stochastic response evaluated, obtained results compared those Monte Carlo simulations (MCSs). show that RBF-NN can accurately predict stationary considerable high computational efficiency.