一类广义变分不等式问题的神经网络模型  

A Neural Network Model for a Class of Generalized Variational Inequality Problem

作　　者：韦丽兰[1] 覃建波[1] 

机构地区：[1]钦州学院物理与电子工程学院,钦州535099

出　　处：《应用泛函分析学报》2014年第4期341-345,共5页Acta Analysis Functionalis Applicata

基　　金：桂教科研[2011]14号(201106LX524)

摘　　要：基于解的充分必要条件,提出一类广义变分不等式问题的神经网络模型.通过构造Lyapunov函数,在适当的条件下证明了新模型是Lyapunov稳定的,并且全局收敛和指数收敛于原问题的解.数值试验表明,该神经网络模型是有效的和可行的.Based on the necessary and sufficient conditions of the solution, this paper proposed a neural network model for a class of generalized variational inequality problem. Under proper conditions, the new model is proved to be Lyapunov stable and convergent to a solution of the original problem by introducing the Lyapunov functions and globally exponential stability of the neural network model for generalized variational inequality problem is studied. The efficiency and feasibility of the proposed model is confirmed by numerical experiments.

关 键 词：广义变分不等式问题 神经网络模型 Lyapunov稳定 指数稳定 

分 类 号：O211.2[理学—概率论与数理统计]