约束非线性规划的神经网络算法

A New Algorithm for Constrained Nonlinear Programming Based on Neural Networks

神经网络具有内在大规模并行运算和快速收敛特性 ,它在最优化技术上的运用近年来受到广泛的重视。本文提出一个新的求解一般约束非线性规划的的神经网络模型 ,它具有全局收敛性和广泛的适用性 ,是求解一般非线性规划问题的新工具。理论分析和模拟计算均表明了模型的有效性。

Artificiel neural networks possess inherent massively parallel processing and fast convergence. Much attention has been paid to its application to optimiation. The present paper proposes a novel neural network model for general constrained nonlinear programming. This model is globally convergent and can be used for different kinds of nonlinear programming problems. It is a new approach to general nonlinear programming. Theoretical anaysis and simulation both demonstrate the efficiency of the model.