Abstract
Existing micromechanics analyses for the prediction of composite material properties, which are based upon the assumption of various idealized regular filament packings, indicate the significant influence of the assumed geometry on numerical results obtained.
To represent the random filament packing of actual composite materials more accurately, a method of analyzing random arrays is presented in this paper. Although certain geometric constraints have necessarily been retained for mathematical reasons, it has been dem onstrated that as the degree of randomness is increased, the agree ment between analytical predictions and available experimental data is improved.
Because of the necessary approximations and additional complexities involved in random array analyses, the methods presented here are not intended to replace regular array analyses, but rather to supplement them, indicating the influence of this difficult-to-predict characteristic of real materials.
Get full access to this article
View all access options for this article.