Neural Networks in Advanced Computational Problems

Abstract

It this chapter we discuss the use of parallel processing for the training of back propagation neural networks. This leads to a discussion of how neural networks and genetic algorithms may be utilised for preprocessing large finite element problems for parallel or distributed finite element analysis. This preprocessing is the partitioning of the finite element mesh into subdomains to ensure load balancing and minimum interprocessor communication during the parallel finite element analysis on a MIMD distributed memory computer.