A Method of Hopfield Neural Network Based Nearest Neighbor Mode Solving Logistics Distribution

Abstract

To solve the problem of invalid and not optimal result for Hopfield neural network to logistics distribution, a nearest neighbor mode Hopfield neural network algorithm based on improved-loop is constructed. The solution of logistics distribution is initialized by nearest neighbor matrix, to seek optimized solution of logistics distribution with the energy function evolvement of Hopfield neural network. To verify this algorithm and the result of simulation, which is compared to other well-known algorithms, indicated that , nearest neighbor mode Hopfield neural network algorithm would avoid invalid result, and better than other well-known algorithm both in convergence rate and quality of optimization.