一类求解约束非线性规划问题的神经网络模型(英文)

A Kind of Neural Network Model for Nonlinear Programming Problems with Constraints

提出一类求解闭凸集上非线性规划问题的神经网络模型．理论分析和计算机模拟表明在适当的假设下所提出的神经网络模型大范围指数级收敛于非线性规划问题的解集．本文神经网络所采用的方法属于广义的最速下降法．甚至当规划问题是正定二次时，本文的模型也比己有的神经网络模型简单．

A kind of neural network model for nonlinear programming problems on a closed convex set is presented in this paper. Theoretic analysis and simulation results on the computer show the neural network is globally convergent exponentially to the exact solutions of the programming problems under some appropriate assumptions. The optimization method employed by the neural network falls into the extended gradient method. The model is simpler than the existing neural network models even when it is for positive definite quadratic programming problems.