Prediction of structures and mechanical properties of composites using a genetic algorithm and finite element method

A combined method of a genetic algorithm and finite element stress analysis has been developed to design the structure of materials. The genetic algor ithm is applied to searching structures that have a desired property by com bining it with the finite element analysis, which is used to predict the el astic modulus and Poisson's ratio. The calculation of the stress analysis i s validated from the comparison with the theory on parallel, series, and ra ndom structures. The combined method was applied to two searches of structu res. One was to find structures that have a desired elastic modulus, respec tively. The calculation successfully found a proper structure for each desi red elastic modulus. The other was the search of the structure that shows a negative Poisson's ratio. A structure having the negative Poisson's ratio was generated by the calculation. Although this original structure would ap pear to have no features, it gave us a good idea for the design of material s by investigating the stress distribution in the original structure. A new structure that consists of a unique and continuous pattern of the higher e lastic component was designed from the calculation. The reason for the nega tive Poisson's ratio is explained by mechanical linkage. (C) 2001 Kluwer Ac ademic Publishers.