用于线性噪声时变凸二次规划的归零神经网络

doi:10.11959/j.issn.1000-436x.2023075

通信学报 ›› 2023, Vol. 44 ›› Issue (4): 226-233.doi: 10.11959/j.issn.1000-436x.2023075

• 学术通信 • 上一篇

用于线性噪声时变凸二次规划的归零神经网络

李建锋¹^,², 刘哲宇¹^,², 荣洋², 李展³, 廖柏林², 屈林曦², 刘志杰², 林琨煌²

¹ 吉首大学数学与统计学院，湖南吉首 416000
² 吉首大学计算机科学与工程学院，湖南吉首 416000
³ 斯旺西大学计算机科学学院，斯旺西 SA28PP

修回日期:2023-03-22 出版日期:2023-04-25 发布日期:2023-04-01
作者简介:李建锋（1979- ），男，湖南张家界人，博士，吉首大学教授、硕士生导师，主要研究方向为人工智能、数字图像处理、计算机视觉、医学图像处理
刘哲宇（1996- ），男，湖南益阳人，吉首大学硕士生，主要研究方向为神经网络
荣洋（1998- ），男，河南商丘人，吉首大学硕士生，主要研究方向为新一代人工智能、神经网络、计算机视觉等
李展（1987- ），男，湖南郴州人，博士，斯旺西大学高级讲师、博士生导师，主要研究方向为神经网络、机器人
廖柏林（1981- ），男，湖南常宁人，博士，吉首大学教授、硕士生导师，主要研究方向为人工智能、神经网络、非线性控制等
屈林曦（1998- ），男，湖南张家界人，吉首大学硕士生，主要研究方向为神经网络、计算机视觉
刘志杰（1995- ），男，河南三门峡人，吉首大学硕士生，主要研究方向为医学图像处理、计算机视觉
林琨煌（1998- ），男，福建泉州人，吉首大学硕士生，主要研究方向为医学图像处理
基金资助:
国家自然科学基金资助项目(61962023);国家自然科学基金资助项目(62066015);湖南省自然科学基金资助项目(2020JJ4511);湖南省教育局研究基金资助项目(20A396);吉首大学科学研究基金资助项目(Jdy20063)

Zeroing neural network for time-varying convex quadratic programming with linear noise

Jianfeng LI¹^,², Zheyu LIU¹^,², Yang RONG², Zhan LI³, Bolin LIAO², Linxi QU², Zhijie LIU², Kunhuang LIN²

¹ Department of Mathematics and Statistics, Jishou University, Jishou 416000, China
² Department of Computer Science and Engineering, Jishou University, Jishou 416000, China
³ Department of Computer Science, Swansea University, Swansea SA28PP, UK

Revised:2023-03-22 Online:2023-04-25 Published:2023-04-01
Supported by:
The National Natural Science Foundation of China(61962023);The National Natural Science Foundation of China(62066015);Hunan Natural Science Foundation(2020JJ4511);The Research Foundation of Education Bureau of Hunan Province(20A396);The Scientific Re-search Foundation of Jishou University(Jdy20063)

摘要/Abstract

摘要：

针对线性噪声可能会对现有的归零神经网络（ZNN）模型求解时变二次规划（TVQP）问题产生负面影响，从而导致模型收敛缓慢、准确率降低的问题，提出了一种双重积分增强ZNN（DIEZNN）模型。为了解决线性噪声的干扰，在 ZNN 原有公式基础上引入双重积分，设计了一个激活函数去除线性噪声的影响。理论分析证实了 DIEZNN 模型具有收敛性和良好的噪声抑制能力。实验结果表明，与传统的梯度神经网络和其他变量 ZNN模型相比，DIEZNN模型收敛更快、精度更高，并且能够有效地解决线性噪声的影响。

关键词: 归零神经网络, 时变二次规划, 线性约束, 噪声干扰

Abstract:

Aiming at the problem that linear time-varying noise may have a negative impact on the existing zeroing neural network model to solve TVQP problem, resulting in slow convergence and low accuracy of the model, a double integral enhancement zeroing neural network was proposed.To solve the problem of linear time-varying interference of the noise, the double integral was introduced based on the original ZNN design formula, and a activation function was designed to eliminate the effects of linear time-varying noise.Theoretical analysis proved that the DIEZNN model had convergence and good noise suppression ability.The experimental results show that compared with the traditional gradient neural network and other variable ZNN models, the proposed DIEZNN model has faster convergence and higher accuracy, and can effectively solve the linear time-varying noise.

Key words: zeroing neural network, time-varying quadratic programming, linear constraint, noise disturbance

中图分类号:

TP183

李建锋, 刘哲宇, 荣洋, 李展, 廖柏林, 屈林曦, 刘志杰, 林琨煌. 用于线性噪声时变凸二次规划的归零神经网络[J]. 通信学报, 2023, 44(4): 226-233.

Jianfeng LI, Zheyu LIU, Yang RONG, Zhan LI, Bolin LIAO, Linxi QU, Zhijie LIU, Kunhuang LIN. Zeroing neural network for time-varying convex quadratic programming with linear noise[J]. Journal on Communications, 2023, 44(4): 226-233.

图/表 6

图1

图2

图3

图4

图5

图6

参考文献 15

[1]	NAZEMI A . A capable neural network framework for solving degenerate quadratic optimization problems with an application in image fusion[J]. Neural Processing Letters, 2018,47(1): 167-192.
[2]	王建建, 何枫, 吴子轩 ,等. 改进区间可接受度的证券投资组合区间二次规划模型[J]. 中国管理科学, 2018,26(9): 11-18.
	WANG J J , HE F , WU Z X ,et al. Interval quadratic programming model for portfolio selection with improved interval acceptability degree[J]. Chinese Journal of Management Science, 2018,26(9): 11-18.
[3]	ZHANG Y N , YI C F . Zhang neural networks and neural-dynamic method[M]. New York: Nova Science Publishers,Inc., 2011.
[4]	张明换, 朱雅光, 姬子恒 . 基于二次规划的四足机器人机身轨迹优化研究[J]. 国外电子测量技术, 2022,41(9): 101-108.
	ZHANG M H , ZHU Y G , JI Z H . Quadratic programming based body trajectory optimization of quadruped robot[J]. Foreign Electronic Measurement Technology, 2022,41(9): 101-108.
[5]	FLETCHER R . A general quadratic programming algorithm[J]. IMA Journal of Applied Mathematics, 1971,7(1): 76-91.
[6]	ZHANG Y , WANG J . A dual neural network for convex quadratic programming subject to linear equality and inequality constraints[J]. Physics Letters A, 2002,298(4): 271-278.
[7]	HU X L , WANG J . Design of general projection neural networks for solving monotone linear variational inequalities and linear and quadratic optimization problems[J]. IEEE Transactions on Systems,Man,and Cybernetics Part B(Cybernetics), 2007,37(5): 1414-1421.
[8]	LI W B , MA X , LUO J W ,et al. A strictly predefined-time convergent neural solution to equality- and inequality-constrained time-variant quadratic programming[J]. IEEE Transactions on Systems,Man,and Cybernetics:Systems, 2021,51(7): 4028-4039.
[9]	ZHANG Y N , RUAN G Q , LI K N ,et al. Robustness analysis of the Zhang neural network for online time-varying quadratic optimization[J]. Journal of Physics A:Mathematical and Theoretical, 2010,43(24): 245202.
[10]	ZHANG Z J , YANG S , ZHENG L N . A penalty strategy combined varying-parameter recurrent neural network for solving time-varying multi-type constrained quadratic programming problems[J]. IEEE Transactions on Neural Networks and Learning Systems, 2021,32(7): 2993-3004.
[11]	BOYD S P , VANDENBERGHE L . Convex optimization[M]. Cambridge: Cambridge University Press, 2004.
[12]	NOCEDAL J , WRIGHT S J . Numerical optimization[M]. New York: Springer New York, 1999.
[13]	HUANG X , LOU X , CUI B . A novel neural network for solving convex quadratic programming problems subject to equality and inequality constraints[J]. Neurocomputing, 2016,214: 23-31.
[14]	JIN L , ZHANG Y N , LI S . Integration-enhanced Zhang neural network for real-time-varying matrix inversion in the presence of various kinds of noises[J]. IEEE Transactions on Neural Networks and Learning Systems, 2016,27(12): 2615-2627.
[15]	ZHANG Y N , CHEN K , LI X Z ,et al. Simulink modeling and comparison of Zhang neural networks and gradient neural networks for time-varying Lyapunov equation solving[C]// Proceedings of 2008 Fourth International Conference on Natural Computation. Piscataway:IEEE Press, 2008: 521-525.

用于线性噪声时变凸二次规划的归零神经网络

Zeroing neural network for time-varying convex quadratic programming with linear noise

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 6

参考文献 15

相关文章 2

Metrics

推荐阅读 0

[1]	李宇翀，罗兴国，钱叶魁，赵鑫. RMPCM：一种基于健壮多元概率校准模型的全网络异常检测方法[J]. 通信学报, 2015, 36(11): 201-212.
[2]	龚文飞,孙昕,吴嗣亮. 小快拍情况下的稳健GPS干扰抑制方法[J]. 通信学报, 2010, 31(8A): 143-147.