HOMOGENIZED PARAMETERS OF THE FIBRE- REINFORCED COMPOSITES SUBJECTED TO THE STOCHASTIC AGING PROCESSES Prof. M. Kamiński, Ph.D., D.Sc. Chair of Mechanics of Materials, Technical University of Łódź Al. Politechniki 6, 90-924 Łódź, POLAND Email: Marcin.Kaminski@p.lodz.pl http://kmm.p.lodz.pl/pracownicy/Marcin_Kaminski/index.html SUMMARY The main aim of this paper is to present an application of the generalized stochastic perturbation technique to model the stochastic aging processes of the fibre-reinforced periodic composite materials in terms of their effective properties. The perturbation technique is used here in its Response Function Method to detect for various time moments of the aging process the analytical interrelations between the effective elasticity tensor components and the composite parameters exposed to the aging phenomena. Numerical procedure based on the FEM code MCCEFF and the system MAPLE is used here to model aging of the glass-epoxy composites and may find a general application in the reliability analysis of various composite structures. Keywords: fibre-reinforced composites, homogenisation method, stochastic ageing, Finite Element Method, Monte-Carlo simulation 1. INTRODUCTION A computational modeling of the stochastic processes [5] based on the Monte-Carlo simulation method is known for their large time consumption because the entire generalizations of random populations in different discrete time moments are evaluated and used in various Finite Element Method models. More optimal strategy would be based on the observation of the probabilistic moments of the examined processes in the same time moments as before performed with at least the comparable accuracy. The generalized stochastic perturbation technique based on the Taylor expansion of the uncertain parameters and the state functions may be useful in this case, to determine for instance the probabilistic moments of the effective elasticity tensor. However, we need to have the analytical description for the basic moments of the stochastic process(es) being modeled to assure the sufficient input for the perturbation-based analysis. This process does not need to have the Gaussian realizations for the whole history, however the lack of correlation between various parameters simplifies significantly the analysis. Contrary to the second order second moment implementations of the perturbation technique, the hierarchical equations are not solved for the increasing order approximants for the probabilistic output. Now, the Response Function Method (RFM)