Non-fragile<mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline" id="d1e255" altimg="si460.svg"><mml:mrow><mml:msub><mml:mrow><mml:mi mathvariant="script">L</mml:mi></mml:mrow><mml:mrow><mml:mn>2</mml:mn></mml:mrow></mml:msub><mml:mo linebreak="goodbreak" linebreakstyle="after">−</mml:mo><mml:msub><mml:mrow><mml:mi mathvariant="script">L</mml:mi></mml:mrow><mml:mrow><mml:mi>∞</mml:mi></mml:mrow></mml:msub></mml:mrow></mml:math>filtering for a class of switched neural networks

Tai, Weipeng; Zuo, Dandan; Zhou, Xuan; Zhou, Jianping; Wang, Zhen

doi:10.1016/j.matcom.2021.01.014

Cited by 17 publications

(6 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When the boundary condition is changed to the Neumann type, the weight learning rule (10) can also be used to ensure the  ∞ stability of SNN (5). However, since Lemma 2 does not hold under the the Neumann boundary condition, (16) needs to be modified to…”

Section: New Existence Conditionmentioning

confidence: 99%

“…Over the last several decades, various dynamic neural networks models, including classical Hopfield networks, 1 switched neural networks (SNNs), 2 stochastic neural networks, 3 competitive neural networks, 4 quaternion‐valued neural networks, 5 fractional neural networks, 6,7 neutral‐type delay neural networks, 8 interval neural networks, 9 memristor‐based neural networks, 10,11 etc., have been constructed and applied successfully to different types of fields 12,13 . Among these models, SNNs, with the development of the understanding of the practical significance of hybrid systems, have gained increasing research interest from the physics and engineering communities 14‐18 …”

Section: Introductionmentioning

confidence: 99%

“…12,13 Among these models, SNNs, with the development of the understanding of the practical significance of hybrid systems, have gained increasing research interest from the physics and engineering communities. [14][15][16][17][18] Parametric uncertainty and external disturbance occur frequently in a practical dynamic system and are often important potential sources of sustained oscillations and instability. 19,20 As a significant branch of modern control theory,  ∞ control is capable of ensuring stabilization with excellent control performance, while less sensitive to statistical knowledge of the uncertainty and disturbance.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Weight learning for ℋ∞$$ {\mathscr{H}}_{\infty } $$ stabilization of uncertain switched neural networks with external disturbance and reaction‐diffusion

Tai

Gao

Zhao

et al. 2023

Adaptive Control & Signal

Self Cite

View full text Add to dashboard Cite

This paper considers the  ∞ stabilization of uncertain switched neural networks with external disturbance and reaction-diffusion. A weight learning rule that ensures the  ∞ stability of the network is proposed utilizing the Lyapunov-Krasovskii functional method. Then, a useful lemma concerning the equivalence of matrix inequities is derived. With the aid of the lemma and Schur complement, a more concise existence condition on the learning rule is developed. Finally, theoretical comparisons and a numerical example are given, which show that the obtained results are extensions and improvements of an existing result.

show abstract

Section: New Existence Conditionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Weight learning for ℋ∞$$ {\mathscr{H}}_{\infty } $$ stabilization of uncertain switched neural networks with external disturbance and reaction‐diffusion

Tai

Gao

Zhao

et al. 2023

Adaptive Control & Signal

Self Cite

View full text Add to dashboard Cite

show abstract

“…But in general, the perturbations in the estimator have been modeled as gain variations in either additive or multiplicative norm-bounded form. 28 Additive gain variations are characterized by a constant bias that is added to the true system state. This type of variation can arise from modeling errors, sensor biases, or other sources of noise.…”

Section: Event-triggering Mechanismmentioning

confidence: 99%

“…Unfortunately, in the existing literature, 9,18,21 the estimator gain variations of the non‐fragile state estimator for delayed NNs under ETM is assumed to satisfy only the additive norm‐bounded conditions. But in general, the perturbations in the estimator have been modeled as gain variations in either additive or multiplicative norm‐bounded form 28 . Additive gain variations are characterized by a constant bias that is added to the true system state.…”

Section: Problem Formationmentioning

confidence: 99%

Event‐triggered nonfragile state estimation for delayed neural networks with additive and multiplicative gain variations

Karnan

Nagamani

2023

Intl J Robust & Nonlinear

View full text Add to dashboard Cite

This article aims to design a nonfragile state estimator for neural networks with time-varying delay under the event-triggered mechanism. This paper addresses the two most common issues that exist in large-scale networks: gain variations in the estimator design and lack of network resources. Based on the nonfragile paradigm and an event-triggered mechanism, an estimator is being developed to address this issue, considering both the additive and multiplicative structured gain variations. Through the Lyapunov-Krasovskii functional method and linear matrix inequality technique, sufficient conditions are derived to guarantee the existence of the proposed state estimator by employing the generalized free-weighting matrix inequality and improved reciprocally convex inequality.Moreover, the obtained linear matrix inequalities provide an explicit expression of the estimator gain matrix and triggering matrix. Eventually, two numerical examples are provided to authenticate the proposed findings.

show abstract

Event‐triggered quantized L2−L∞$$ {\mathfrak{L}}_2-{\mathfrak{L}}_{\infty } $$ filtering for neural networks under denial‐of‐service attacks

Zhou

Chang

2022

Intl J Robust & Nonlinear

View full text Add to dashboard Cite

This article deals with the reliable event‐triggered quantized frakturL2prefix−frakturL∞$$ {\mathfrak{L}}_2-{\mathfrak{L}}_{\infty } $$ filtering issue for neural networks with exterior interference under denial‐of‐service attacks. In order to lighten the load of communication channels and save network resources, a resilient event‐triggered mechanism and a quantization scheme are employed, simultaneously. By applying a piecewise Lyapunov–Krasovskii functional method, sufficient conditions containing limitations of denial‐of‐service attacks are derived to guarantee that the filter error system is exponentially stable as well as possesses a prescribed frakturL2prefix−frakturL∞$$ {\mathfrak{L}}_2-{\mathfrak{L}}_{\infty } $$ disturbance attenuation performance. Then, a co‐design method of the desired quantized frakturL2prefix−frakturL∞$$ {\mathfrak{L}}_2-{\mathfrak{L}}_{\infty } $$ filtering gain matrix and event‐triggering parameter can be obtained provided that the linear matrix inequalities have a feasible solution. Finally, the usefulness of the proposed design method is demonstrated by a numerical example.

show abstract

Non-fragileL2−L∞filtering for a class of switched neural networks

Cited by 17 publications

References 44 publications

Weight learning for ℋ∞$$ {\mathscr{H}}_{\infty } $$ stabilization of uncertain switched neural networks with external disturbance and reaction‐diffusion

Weight learning for ℋ∞$$ {\mathscr{H}}_{\infty } $$ stabilization of uncertain switched neural networks with external disturbance and reaction‐diffusion

Event‐triggered nonfragile state estimation for delayed neural networks with additive and multiplicative gain variations

Event‐triggered quantized L2−L∞$$ {\mathfrak{L}}_2-{\mathfrak{L}}_{\infty } $$ filtering for neural networks under denial‐of‐service attacks

Contact Info

Product

Resources

About