Less conservative stability criteria for general neural networks through novel delay-dependent functional
作者：
Highlights：
• This paper proposes two new LKFs to relieve the conservatism of stability criteria for neural networks with time-varying delays. The proposed LKFs are inspired from Wirtinger-based integral inequality.
• Since they include the time-varying delays, more information on time-varying delays have been utilized in stability conditions.
• By constructing novel augmented LKFs and utilizing the augmented zero equality approach, an enhanced stability condition is derived in Theorem 3.
• At this time, the free variable related to the augmented zero equalities introduced in Theorems 1 and 2 are eliminated, thus reducing computational complexity in Theorem 3.
摘要
•This paper proposes two new LKFs to relieve the conservatism of stability criteria for neural networks with time-varying delays. The proposed LKFs are inspired from Wirtinger-based integral inequality.•Since they include the time-varying delays, more information on time-varying delays have been utilized in stability conditions.•By constructing novel augmented LKFs and utilizing the augmented zero equality approach, an enhanced stability condition is derived in Theorem 3.•At this time, the free variable related to the augmented zero equalities introduced in Theorems 1 and 2 are eliminated, thus reducing computational complexity in Theorem 3.
论文关键词：Stability,Zero equalities,Time-varying delay,Neural networks
论文评审过程：Received 2 June 2021, Revised 16 September 2021, Accepted 18 December 2021, Available online 3 January 2022, Version of Record 3 January 2022.
论文官网地址：https://doi.org/10.1016/j.amc.2021.126886
