NumericalEvaluationof Effective Transport Properties of Random Cell Models: Two-Point Probability Approach

摘要

This paper presents a computationally efficient method for evaluating effective transport properties of two-phase random composite materials. It consists in determining, based on a two point probability function, the appropriate volume size of composite and then performing for such a volume a sequence of realizations and numerical calculations to obtain Monte Carlo estimate of the effective transport coefficient. The methodology proposed is tested by considering a particular two-dimensional random cell structure. It is shown that there exists a threshold value of the sample size below which, if it is used, the Monte Carlo estimate does not give a correct prediction of effective transport coefficient value. On the other hand, any samples greater than the critical one can be utilized providing that a sufficient number of realizations are performed.