一种解决分布式伪凸优化问题的神经网络模型  

A neural network model for solving distributed pseudo-convex optimization problems with convex inequality constraints

作　　者：周顺 喻昕[1] 黄镘潼 覃海华 李浩宇 ZHOU Shun;YU Xin;HUANG Mantong;QIN Haihua;LI Haoyu(School of Computer,Electronics and Information,Guangxi University,Nanning 530004,China)

机构地区：[1]广西大学计算机与电子信息学院,广西南宁530004

出　　处：《广西大学学报（自然科学版）》2023年第5期1181-1192,共12页Journal of Guangxi University（Natural Science Edition）

基　　金：国家自然科学基金项目(61862004)。

摘　　要：针对带有凸不等式约束的分布式伪凸优化问题,基于微分包含思想,提出了一种分布式神经网络模型。该模型根据目标函数与约束条件构造罚函数,使得每个分布式节点能在有限时间内进入可行域且不再离开,最终达成一致并收敛到原始优化问题的最优解。最后,通过2个仿真实验验证了分布式神经网络模型的有效性和准确性。与现有的神经网络模型相比,该模型结构简单,能够解决伪凸问题,并且不需要精确的计算罚参数。A distributed neural network model based on differential inclusion is proposed for distributed pseudo-convex optimization problems with convex inequality constraints.Penalty function is constructed in this model according to the objective function and constraints,so that each node can enter the feasible region within a finite time and never leave,and finally reaches a consensus and converges to the optimal solution of the original optimization problem.Finally,we use two simulation experiments to verify the effectiveness and accuracy of the distributed neural network model.Compared with the existing neural network models,the proposed model has a simple structure,can solve the pseudoconvex problem,and does not require accurate calculation of penalty parameters.

关 键 词：分布式优化 伪凸优化 罚函数 神经网络 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]