A new projection neural network for convex nonlinear second-order cone programming with an application in force optimization problems

被引：0
|
作者：
Mu, Xuewen [1 ]
Zhao, Bingcong [1 ]
Liu, Yitong [1 ]
机构：
[1] Xidian Univ, Sch Math & Stat, Xian 710071, Peoples R China
来源：
INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS | 2025年
关键词：
Convex nonlinear second-order cone programming; Projection neural network; Force optimization problem; Lyapunov stability; CONTRACTION METHODS; CONVERGENCE; FRAMEWORK;
D O I：
10.1007/s13042-025-02538-8
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
In this paper, a new projection neural network (PNN) is proposed for the convex nonlinear second-order cone programming (CNSOCP) with linear constraints. The KKT system of the CNSOCP with linear constraints is equivalent to the cone projection equation. Based on the natural residual function, a new PNN is proposed by using the descent direction of the Lyapunov function. The new PNN is derived from a projection descent direction of the prediction-correction projection and contraction method. Furthermore, the Lyapunov stability and global convergence of the proposed PNN are proved. The numerical results show the new PNN is efficient for CNSOCP problems with linear constraints and two force optimization problems.
引用
收藏
页数：9
相关论文
共 50 条
12345