数学物理学报(英文版) ›› 2025, Vol. 45 ›› Issue (3): 1188-1204.doi: 10.1007/s10473-025-0323-3

• • 上一篇    下一篇

PERIODICITY AND FIXED-TIME STABILIZATION OF DISCONTINUOUS NEURAL NETWORKS WITH MIXED DELAYS: UNBOUNDED DELAY-DEPENDENT CRITERIA

Ziwei WANG1, Lin SUN1, Fanchao KONG2,†, Rathinasamy SAKTHIVEL3   

  1. 1. School of Finance, Hunan University of Finance and Economics, Changsha 410205, China;
    2. School of Mathematics and Statistics, Anhui Normal University, Wuhu 241000, China;
    3. Department of Applied Mathematics, Bharathiar University, Coimbatore 641046, India
  • 收稿日期:2023-08-11 修回日期:2024-05-15 出版日期:2025-05-25 发布日期:2025-09-30

PERIODICITY AND FIXED-TIME STABILIZATION OF DISCONTINUOUS NEURAL NETWORKS WITH MIXED DELAYS: UNBOUNDED DELAY-DEPENDENT CRITERIA

Ziwei WANG1, Lin SUN1, Fanchao KONG2,†, Rathinasamy SAKTHIVEL3   

  1. 1. School of Finance, Hunan University of Finance and Economics, Changsha 410205, China;
    2. School of Mathematics and Statistics, Anhui Normal University, Wuhu 241000, China;
    3. Department of Applied Mathematics, Bharathiar University, Coimbatore 641046, India
  • Received:2023-08-11 Revised:2024-05-15 Online:2025-05-25 Published:2025-09-30
  • Contact: Fanchao KONG, E-mail: fanchaokong88@ahnu.edu.cn
  • About author:Ziwei WANG, E-mail: hnu517@163.com; Lin SUN, E-mail: 43356972@qq.com; Rathinasamy SAKTHIVEL, E-mail: krsakthivel0209@gmail.com
  • Supported by:
    Social Science Fund of Hunan province (Grant No.22JD074) and the Research Foundation of Education Bureau of Hunan province (Grant No.22B0912).

摘要: In this paper, a class of discontinuous neutral-type neural networks (NTNNs) with proportional delays is considered. The targets of the paper are to study the problem of periodic solutions and fixed-time (FXT) stabilization of the addressed neural networks. In order to complete the targets, based on set-valued map, differential inclusions theory, coincidence theorem and Hölder inequality technique, some new proportional delay-dependent criteria shown by the inequalities are derived. Based on the fact of the existence of solution, further by applying the FXT stability lemmas and equivalent transformation, the zero solution of closed-loop system achieves FXT stabilization and the corresponding settling-times are estimated. Some previous related works on NTNNs are extended. Finally, one typical example is provided to show the effectiveness of the established results.

关键词: Fixed-time stabilization, Periodic solutions, Discontinuous neural systems, Differential inclusions theory, Proportional delays

Abstract: In this paper, a class of discontinuous neutral-type neural networks (NTNNs) with proportional delays is considered. The targets of the paper are to study the problem of periodic solutions and fixed-time (FXT) stabilization of the addressed neural networks. In order to complete the targets, based on set-valued map, differential inclusions theory, coincidence theorem and Hölder inequality technique, some new proportional delay-dependent criteria shown by the inequalities are derived. Based on the fact of the existence of solution, further by applying the FXT stability lemmas and equivalent transformation, the zero solution of closed-loop system achieves FXT stabilization and the corresponding settling-times are estimated. Some previous related works on NTNNs are extended. Finally, one typical example is provided to show the effectiveness of the established results.

Key words: Fixed-time stabilization, Periodic solutions, Discontinuous neural systems, Differential inclusions theory, Proportional delays

中图分类号: 

  • 34A36