Authors
Chen-Rui Wang, Yong He, Kang-Zhi Liu, Wen-Hu Chen, Chuan-Ke Zhang
Published in
ISA transactions. Nov 05, 2026. Epub Nov 05, 2026.
Abstract
This study addresses the design of H∞ state estimation for delayed neural networks by proposing enhanced estimation techniques with reduced conservatism. First, two augmented Lyapunov-Krasovskii (L-K) functionals are constructed: a Legendre-polynomial-based augmented quadratic functional employing a set of m state vectors, and an augmented double-integral functional incorporating both the state and its derivative information. Then, to obtain tighter estimation bounds for the derivative of the augmented L-K functionals, a refined free-matrix-based Bessel-Legendre integral inequality is introduced. It directly estimates the coupled integral term of the state and its derivative, and it is constructed in coordination with the Legendre integral states of the augmented functional, so that the inequality and the functional work together to reduce conservatism. Moreover, to resolve the negative definiteness issue caused by high-order delay polynomials arising from the m-order augmented L-K functional and the application of the integral inequality, variable-augmented free-weighting matrices are employed. These innovations yield less conservative H∞ state estimation results and enable the design of suitable estimator gains for engineering systems subject to time-varying delays and external disturbances. Finally, numerical examples demonstrate the effectiveness of the proposed methods.
PMID:
43009379
Bibliographic data and abstract were imported from PubMed on 11 Nov 2026.
Read full publication at:
Please sign in
to see all details.
Advertisement
Stats
- Recommendations n/a n/a positive of 0 vote(s)
- Views 3
- Comments 0