Abstract
This paper proposes a decentralized dynamic event-triggered control method of aero-engine T-S fuzzy system, aiming to achieve both exponential stabilization of aero-engine networked control systems (NCSs) and reduction in network communication loads. Firstly, a novel decentralized dynamic event-triggered mechanism (DETM) is developed to regulate data transmissions in each communication channel independently. The closed-loop model is then established, by considering network-induced delays, dynamic quantization effects, external disturbance, and parameter perturbation. Stability criteria are derived using the Lyapunov–Krasovskii method, and a collaborative design method based on linear matrix inequalities (LMIs) is proposed for controller, quantizers, and DETM. Lastly, a parameter tuning method based on the iL-SHADE algorithm is proposed for better feasibility of the obtained LMIs. Simulation results show that the proposed method has good robustness to multi-uncertainty conditions and can effectively reduce communication resource wastage while ensuring well control performance across a wide range of flight envelopes.
Similar content being viewed by others
Data availability
Data sharing is not applicable to this article as no new data were created or analyzed in this study.
References
Li, R., Nguang, S.K., Guo, Y., et al.: Networked control system design for turbofan aeroengines with aging and deterioration. Complexity 2018, 1–13 (2018)
Zhou, B., Xie, S., Wang, F., et al.: Multi-step predictive compensated intelligent control for aero-engine wireless networked system with random scheduling. J. Franklin Inst. 357(10), 6154–6174 (2020)
Zhang, X., Han, Q., Ge, X., et al.: Networked control systems: a survey of trends and techniques. IEEE/CAA Journal of Automatica Sinica 7(01), 1–17 (2019)
Xu, Z., Li, X., Stojanovic, V.: Exponential stability of nonlinear state-dependent delayed impulsive systems with applications. Nonlinear Anal. Hybrid Syst. 42, 101088 (2021)
Wei, T., Li, X., Stojanovic, V.: Input-to-state stability of impulsive reaction–diffusion neural networks with infinite distributed delays. Nonlinear Dyn. 103(2), 1733–1755 (2021)
Belapurkar, R.K. and Yedavalli, R.K. Study of model-based fault detection of distributed aircraft engine control systems with transmission delays. In: 47th AIAA/ASME/SAE/ASEE Joint Propulsion Conference. 2011. San Diego, California.
Zhang, L., Xie, S., Yu, Z., et al.: Aero-engine DCS fault-tolerant control with Markov time delay based on augmented adaptive sliding mode observer. Asian J. Contr. 22(2), 788–802 (2018)
Wang, W., Xie, S., Peng, J., et al.: Dual-terminal event-triggered dynamic output feedback H∞ control for aero engine networked control systems with quantization. Int. J. Robust Nonlinear Control 33(2), 697–719 (2023)
Wang, H., Wang, D.: Turbofan aero-engine full flight envelope control using H∞ loop-shaping control technique. Int. J. Adv. Mechatr. Syst. 3(1), 65–76 (2011)
Takagi, T., Sugeno, M.: Fuzzy identification of systems and its applications to modeling and control. IEEE Trans. Syst. Man Cybern. 15(1), 116–132 (1985)
Li, R., Guo, Y., Nguang, S.K., et al.: Takagi-Sugeno fuzzy model identification for turbofan aero-engines with guaranteed stability. Chin. J. Aeronaut. 31(6), 1206–1214 (2018)
Wang, H., Xie, S., Wang, W., et al.: Investigation of unmeasured parameters estimation for distributed control systems. Complexity 2020, 1–15 (2020)
Song, P., Yang, Q., Wen, G., et al.: Fuzzy H∞ robust control for T-S aero-engine systems with network-induced factors under round-robin-like protocol. Aerosp. Sci. Technol. 137, 108258 (2023)
Song, X., Sun, P., Song, S., et al.: Event-driven NN adaptive fixed-time control for nonlinear systems with guaranteed performance. J. Franklin Inst. 359(9), 4138–4159 (2022)
Zhang, K., Zhao, T., Dian, S.: Dynamic output feedback control for nonlinear networked control systems with a two-terminal event-triggered mechanism. Nonlinear Dyn. 100(3), 2537–2555 (2020)
Kanakalakshmi, S., Sakthivel, R., Karthick, S.A., et al.: Finite-time decentralized event-triggering non-fragile control for fuzzy neural networks with cyber-attack and energy constraints. Eur. J. Control. 57, 135–146 (2021)
Ge, X., Han, Q., Ding, L., et al.: Dynamic event-triggered distributed coordination control and its applications: a survey of trends and techniques. IEEE Trans. Syst. Man Cyber. Syst. 50(9), 3112–3125 (2020)
Ahmad, I., Ge, X., Han, Q.: Decentralized dynamic event-triggered communication and active suspension control of in-wheel motor driven electric vehicles with dynamic damping. IEEE/CAA J. Autom. Sin. 8(5), 971 (2021)
Liberzon, D.: Hybrid feedback stabilization of systems with quantized signals. Automatica 39(9), 1543–1554 (2003)
Csernak, G., Gyebroszki, G., Stepan, G.: Multi-baker map as a model of digital PD control. Int. J. Bifurcat. Chaos 26(02), 1650025 (2016)
Hu, S., Yue, D.: Event-triggered control design of linear networked systems with quantizations. ISA Trans. 51(1), 153–162 (2012)
Tan, Y., Xiong, M., Du, D., et al.: Observer-based robust control for fractional-order nonlinear uncertain systems with input saturation and measurement quantization. Nonlinear Anal. Hybrid Syst 34, 45–57 (2019)
Wang, J., Zhou, M., Liu, C.: Stochastic stability of Markovian jump linear systems over networks with random quantization density and time delay. Phys. A 509, 1128–1139 (2018)
Chen, G., Chen, Y., Zeng, H.: Event-triggered H∞ filter design for sampled-data systems with quantization. ISA Trans. 101, 170–176 (2020)
Chang, X., Xiong, J., Li, Z., et al.: Quantized static output feedback control for discrete-time systems. IEEE Trans. Industr. Inf. 14(8), 3426–3435 (2018)
Liu, K., Fridman, E., Johansson, K.H.: Dynamic quantization of uncertain linear networked control systems. Automatica 59, 248–255 (2015)
Li, B., Wang, Z., Han, Q., et al.: Input-to-state stabilization in probability for nonlinear stochastic systems under quantization effects and communication protocols. IEEE Trans. Cybern. 49(9), 3242–3254 (2019)
Li, Z., Xiong, J.: Event-triggered fuzzy filtering for nonlinear networked systems with dynamic quantization and stochastic cyber attacks. ISA Trans. 121, 53–62 (2022)
Wang, Y., Xia, Y., Ahn, C.K., et al.: Exponential stabilization of takagi-sugeno fuzzy systems with aperiodic sampling: an aperiodic adaptive event-triggered method. IEEE Trans. Syst. Man Cybern. Syst. 49(2), 444–454 (2019)
Qi, Y., Tang, Y., Ke, Z., et al.: Dual-terminal decentralized event-triggered control for switched systems with cyber attacks and quantization. ISA Trans. 110, 15–27 (2021)
Yang, H., Li, P., Xia, Y., et al.: H∞ static output feedback for low-frequency networked control systems with a decentralized event-triggered scheme. IEEE Trans. Cybern. 51(8), 4227–4236 (2021)
Brest, J., Maucec, M.S. and Boskovic, B. iL-SHADE improved L-SHADE algorithm for single objective real-parameter optimization. In: 2016 IEEE Congress on Evolutionary Computation 2016.
Xu, W., Huang, J., Qin, J., et al.: Limit protection design in turbofan engine acceleration control based on scheduling command governor. Chin. J. Aeronaut. 34(10), 67–80 (2021)
Sheng, H., Zhang, T., Jiang, W.: Full-range mathematical modeling of turboshaft engine in aerospace. Int. J. Turbo Jet Eng. 33(4), 309 (2016)
Liu, X., Luo, C., Xiong, L.: Compensators design for bumpless switching in aero-engine multi-loop control system. Asian J. Contr. 24(5), 2665–2678 (2021)
Imani, A., Montazeri-Gh, M.: A multi-loop switching controller for aircraft gas turbine engine with stability proof. Int. J. Control Autom. Syst. 17(6), 1359–1368 (2019)
Zhang, J., Peng, C., Du, D., et al.: Adaptive event-triggered communication scheme for networked control systems with randomly occurring nonlinearities and uncertainties. Neurocomputing 174, 475–482 (2016)
Hu, S., Yue, D., Cheng, Z., et al.: Co-design of dynamic event-triggered communication scheme and resilient observer-based control under aperiodic DoS attacks. IEEE Trans. Cybernet. 51(9), 4591–4601 (2021)
Sun, X., Yang, D., Zong, G.: Annular finite-time H∞ control of switched fuzzy systems: a switching dynamic event-triggered control approach. Nonlinear Anal. Hybrid Syst. 41, 101050 (2021)
Jiang, X., Han, Q., Liu, S., et al.: A new H∞ stabilization criterion for networked control systems. IEEE Trans. Autom. Control 53(4), 1025–1032 (2008)
Kim, J.-H.: Further improvement of Jensen inequality and application to stability of time-delayed systems. Automatica 64, 121–125 (2016)
Song, P., Yang, Q., Wen, G., et al.: Observer-based output feedback control for networked systems with dual-channel event-triggered sampling and quantization. J. Franklin Inst. 359(14), 7365–7392 (2022)
Choi, T.J., Ahn, C.W.: An improved LSHADE-RSP algorithm with the Cauchy perturbation: iLSHADE-RSP. Knowl Based Syst. 215, 106628 (2021)
Zhou, B., Xie, S., Hui, J.: H-infinity control for T-S aero-engine wireless networked system with scheduling. IEEE Access 7, 115662–115672 (2019)
Park, P., Ko, J.W., Jeong, C.: Reciprocally convex approach to stability of systems with time-varying delays. Automatica 47(1), 235–238 (2011)
Acknowledgements
This work is supported by the National Major Science and Technology Project of China (Grant Numbers [J2019-V-0003-0094]), and the Natural Science Basic Research Program of Shaanxi (Grant Numbers [2021JQ-359]).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflicts of interest
The authors have no relevant financial or non-financial interests to disclose.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendices
Appendix 1
Proof of Theorem 1
Choose the following Lyapunov–Krasovskii function as
where \(V_{1} \left( t \right) = {\varvec{x}}^{T} \left( t \right){\varvec{Px}}\left( t \right)\) with \({\varvec{P}} > 0\), \(V_{2} \left( t \right) = \int_{{t - \tau_{{\text{M}}} }}^{t} {e^{{2\lambda \left( {s - t} \right)}} {\varvec{x}}^{T} \left( s \right){\varvec{Qx}}\left( s \right)ds}\) with \({\varvec{Q}} > 0\), \(V_{3} \left( t \right) = \int_{{ - \tau_{{\text{M}}} }}^{0} {\int_{t + s}^{t} {e^{{2\lambda \left( {v - t} \right)}} {\dot{\varvec{x}}}^{T} \left( v \right){\varvec{R}}{\dot{\varvec{x}}}\left( v \right)dv} ds}\) with \({\varvec{R}} > 0\), \(V_{4} \left( t \right) = \sum\limits_{i = 1}^{2} {\eta_{i} \left( t \right)}\).
This together with Lemma 1 yields
where \(\lambda_{{\text{M}}} = \lambda_{\max } \left( {\varvec{P}} \right) + \tau_{{\text{M}}} \lambda_{\max } \left( {\varvec{Q}} \right) + \frac{1}{2}\tau_{{\text{M}}}^{2} \lambda_{\max } \left( {\varvec{R}} \right)\).
Calculating the derivative of \(V_{l} \left( t \right)\left( {l = 1,2,3,4} \right)\) along the trajectory of system (18) yields
where \(\Gamma_{1} = - \tau_{{\text{M}}} e^{{ - 2\lambda \tau_{{\text{M}}} }} \int_{{t - \tau_{{\text{M}}} }}^{t - \tau \left( t \right)} {{\dot{\varvec{x}}}^{T} \left( s \right){\varvec{R}}{\dot{\varvec{x}}}\left( s \right)ds}\), \(\Gamma_{2} = - \tau_{{\text{M}}} e^{{ - 2\lambda \tau_{{\text{M}}} }} \int_{t - \tau \left( t \right)}^{t} {{\dot{\varvec{x}}}^{T} \left( s \right){\varvec{R}}{\dot{\varvec{x}}}\left( s \right)ds}\).
To handle the nonlinear integral terms in \(\dot{V}_{3} \left( t \right) + 2\lambda V_{3} \left( t \right)\), we utilize Lemma 2 and reciprocally convex combination inequality [45] to obtain
Follow to \(\dot{V}_{4} \left( t \right) + 2\lambda V_{4} \left( t \right)\), owing to \(\rho_{i} > 2\lambda\), combining (19) and (20), it can be deduced that
where \(v_{i} = 1 + \rho_{i} - 2\lambda\). It follows that
Define \(J = \dot{V}\left( t \right) + 2\lambda V\left( t \right) - \gamma^{2} {\varvec{w}}^{T} \left( t \right){\varvec{w}}\left( t \right) + {\varvec{y}}^{T} \left( t \right){\varvec{y}}\left( t \right)\). Combining (32)–(35), one obtains
where
\({{\boldsymbol{\varSigma}}}_{0} = \sum\limits_{m = 1}^{r} {\sum\limits_{n = 1}^{r} {\mu_{m} \mu_{n} \left\{ \begin{gathered} {\varvec{e}}_{1}^{T} \left[ {2\lambda {\varvec{P}} + {\text{sym}}\left\{ {{\varvec{P}}\left( {{\varvec{A}}_{m} + \Delta {\varvec{A}}_{m} } \right)} \right\} + {\varvec{Q}}} \right]{\varvec{e}}_{1} + \sigma {\varvec{e}}_{2}^{T} {\boldsymbol{\nu \varPhi e}}_{2} - e^{{ - 2\lambda \tau_{{\text{M}}} }} {\varvec{e}}_{3}^{T} {\varvec{Qe}}_{3} \hfill \\ - \left( {1 - \sigma } \right){\varvec{e}}_{8}^{T} {\boldsymbol{\nu \varPhi e}}_{8} - \gamma^{2} {\varvec{e}}_{9}^{T} {\varvec{e}}_{9} + {\text{sym}}\left\{ {{\varvec{e}}_{1}^{T} {\varvec{PB}}_{m} {\varvec{K}}_{n} \left( {{\varvec{e}}_{2} + {\varvec{e}}_{8} + {\varvec{e}}_{10} } \right) + {\varvec{e}}_{1}^{T} {\varvec{Pe}}_{9} } \right\} \hfill \\ + {\text{sym}}\left\{ {{\varvec{e}}_{1}^{T} {\varvec{PB}}_{m} {\varvec{e}}_{11} + \sigma {\varvec{e}}_{2}^{T} {\boldsymbol{\nu \varPhi e}}_{8} } \right\} - {{\boldsymbol{\varTheta}}} + \tau_{{\text{M}}}^{2} {{\boldsymbol{\varLambda}}}_{mn}^{T} {\varvec{R\varLambda }}_{mn} + {\varvec{e}}_{1}^{T} {\varvec{C}}_{m}^{T} {\varvec{C}}_{m} {\varvec{e}}_{1} \hfill \\ \end{gathered} \right\}} }\).
From (12), we can get whenever \(\left\| {\frac{{{\varvec{x}}_{i} \left( {\tilde{t}_{k}^{i} h} \right)}}{{\mu_{x}^{i} }}} \right\| \le M_{x}^{i}\) and \(\left\| {\frac{{{\varvec{u}}_{\text{c}} \left( t \right)}}{{\mu_{u} }}} \right\| \le M_{u}\), then
Without loss of generality, the following variables are defined
where \(\kappa_{x}^{i} \ge 1\), \(0 < \kappa_{u} \le 1\) are scalars to be designed. It is easy to verify that (39) can guarantee the precondition of (37), and (40) can guarantee that of (38).
By combining (13), (37) and (39), we can get
where \(\kappa_{x} = \max \left\{ {\kappa_{x}^{1} ,\kappa_{x}^{2} } \right\}\), \(\Delta_{x} = \max \left\{ {\Delta_{x}^{1} ,\Delta_{x}^{2} } \right\}\), \(M_{x} = \min \left\{ {M_{x}^{1} ,M_{x}^{2} } \right\}\).
Then, the inequality (41) can be rewritten as
where \({{\boldsymbol{\varSigma}}}_{1} = \left( {{\varvec{e}}_{2} + {\varvec{e}}_{8} } \right)^{T} \left( {{\varvec{e}}_{2} + {\varvec{e}}_{8} } \right) - \frac{{M_{x}^{2} }}{{\kappa_{x}^{2} {\Delta }_{x}^{2} }}{\varvec{e}}_{10}^{T} {\varvec{e}}_{10}\).
Considering the property of Euclidean norm, we have
From (14), (22), (38), (40), and (43), one has
Further, we can rewrite the inequality (44) to
where \({{\boldsymbol{\varSigma}}}_{2} = \left( {{\varvec{e}}_{2} + {\varvec{e}}_{8} } \right)^{T} \left( {{\varvec{e}}_{2} + {\varvec{e}}_{8} } \right) - \frac{{s\kappa_{u} M_{u}^{2} }}{{\Delta_{u}^{2} }}\left( {1 + \frac{{\kappa_{x} \Delta_{x} }}{{M_{x} }}} \right)^{ - 2} {\varvec{e}}_{11}^{T} {\varvec{e}}_{11}\).
Define \(\kappa_{1} = \frac{1}{{\kappa_{x}^{2} }}\), \(\kappa_{2} = \kappa_{u} \left( {1 + \frac{{\kappa_{x} \Delta_{x} }}{{M_{x} }}} \right)^{ - 2}\). Since \(\kappa_{x}^{i} \ge 1\), \(0 < \kappa_{u} \le 1\), it is easy to obtain
\(0 < \kappa_{1} \le 1\)
Then, according to Lemma 3, it can be deduced that \({{\boldsymbol{\xi}}}^{T} \left( t \right){{\boldsymbol{\Sigma}}}_{0} {{\boldsymbol{\xi}}}\left( t \right) < 0\) holds if
Noting that \( \sum\nolimits_{{m = 1}}^{r} {\sum\nolimits_{{m = 1}}^{r} {\mu _{m} \mu _{n} {\text{ }}\left\{ {{\boldsymbol{\varOmega }}_{{mn}} } \right\}} } = \sum\nolimits_{{m = 1}}^{r} {\mu _{m}^{2} } \left\{ {{\boldsymbol{\varOmega }}_{{mm}} } \right\} + \sum\nolimits_{{m = 1}}^{r} {\sum\nolimits_{{n = m + 1}}^{r} {\mu _{m} \mu _{n} } } \left\{ {{\boldsymbol{\varOmega }}_{{mn}} + {\boldsymbol{\varOmega }}_{{nm}} } \right\} \), we can obtain that (47) can be guaranteed by
Based on Schur complement, (23) and (24) is equivalent to
respectively. Therefore, combining (36), (47), and (48), we can obtain that if (23) and (24) hold, then \(J \le 0\). It implies
Under zero conditions, the integration of both sides of (51) from \(0\) to \(+ \infty\) yields
With condition \(w\left( t \right) \equiv 0\), we can get \(\dot{V}\left( t \right) \le - 2\lambda V\left( t \right), t \in \Omega_{{t_{k} }}\) from (51). Thus, we have
From (31) and (53), it follows that
Letting \(\beta = \sqrt {{{\lambda_{{\text{M}}} } \mathord{\left/ {\vphantom {{\lambda_{{\text{M}}} } {\lambda_{\min } \left( {\varvec{P}} \right)}}} \right. \kern-0pt} {\lambda_{\min } \left( {\varvec{P}} \right)}}}\), (54) follows that
Then from (52), (55) and Definition 1, we naturally obtain that system (18) is exponentially stable while having \(H\infty\) performance level \(\gamma\). This completes the proof.
Appendix 2
Proof of Theorem 2
Using the Schur complement to (22) leads to
with \({\varvec{Y}}_{m} = {\varvec{K}}_{m} {\varvec{X}}\), performing congruence transformation to (56) with \({\text{diag}}\left\{ {{\varvec{X}},\,{\varvec{I}}} \right\}\) yields
It is known that \(\left( {{{\boldsymbol{\Delta}}} - \frac{1}{\vartheta }{\varvec{X}}} \right){{\boldsymbol{\Delta}}}^{ - 1} \left( {{{\boldsymbol{\Delta}}} - \frac{1}{\vartheta }{\varvec{X}}} \right) \ge 0\) always holds for any matrix \({{\boldsymbol{\Delta}}} > 0\) and scalar \(\vartheta > 0\), which implies that
Let \({{\boldsymbol{\Delta}}}\) stands for \(s{\varvec{I}}\), (58) becomes \(- {\varvec{X}}\frac{1}{s}{\varvec{X}} \le \vartheta_{1}^{2} s{\varvec{I}} - 2\vartheta_{1} {\varvec{X}}\). Then, the inequality (57) can be guaranteed by (26).
Note that the inequalities (23) and (24) can be equivalently expressed as (59) and (60), respectively.
where \({\varvec{H}}_{{G_{m} }} = \left[ {\begin{array}{*{20}c} {{\varvec{G}}_{m}^{T} {\varvec{Pe}}_{1} } & {\tau_{{\text{M}}} {\varvec{G}}_{m}^{T} {\varvec{R}}} & {\varvec{0}} \\ \end{array} } \right]^{T}\), \({\varvec{H}}_{{E_{m} }} = \left[ {\begin{array}{*{20}c} {{\varvec{E}}_{m} {\varvec{e}}_{1} } & {\varvec{0}} & {\varvec{0}} \\ \end{array} } \right]\), \(\begin{gathered} {\tilde{\boldsymbol{\varPsi }}}_{mn} = {\varvec{e}}_{1}^{T} \left[ {2\lambda {\varvec{P}} + {\varvec{A}}_{m}^{T} {\varvec{P}} + {\varvec{PA}}_{m} + {\varvec{Q}}} \right]{\varvec{e}}_{1} + \sigma {\varvec{e}}_{2}^{T} {\boldsymbol{\nu \varPhi e}}_{2} - e^{{ - 2\lambda \tau_{{\text{M}}} }} {\varvec{e}}_{3}^{T} {\varvec{Qe}}_{3} - \left( {1 - \sigma } \right){\varvec{e}}_{8}^{T} {\boldsymbol{\nu \varPhi e}}_{8} - \gamma^{2} {\varvec{e}}_{9}^{T} {\varvec{e}}_{9} \hfill \\ \quad \quad \quad - \frac{{M_{x}^{2} }}{{{\Delta }_{x}^{2} }}{\varvec{e}}_{10}^{T} {\varvec{e}}_{10} - \frac{{sM_{u}^{2} }}{{\Delta_{u}^{2} }}{\varvec{e}}_{11}^{T} {\varvec{e}}_{11} + {\text{sym}}\left\{ {{\varvec{e}}_{1}^{T} {\varvec{PB}}_{m} {\varvec{K}}_{n} \left( {{\varvec{e}}_{2} + {\varvec{e}}_{8} + {\varvec{e}}_{10} } \right) + {\varvec{e}}_{1}^{T} {\varvec{Pe}}_{9} + {\varvec{e}}_{1}^{T} {\varvec{PB}}_{m} {\varvec{e}}_{11} + \sigma {\varvec{e}}_{2}^{T} {\boldsymbol{\nu \varPhi e}}_{8} } \right\},\; \hfill \\ \quad \quad \quad {\tilde{\boldsymbol{\varLambda }}}_{mn} = {\varvec{A}}_{m} {\varvec{e}}_{1} + {\varvec{B}}_{m} {\varvec{K}}_{n} \left( {{\varvec{e}}_{2} + {\varvec{e}}_{8} + {\varvec{e}}_{10} } \right) + {\varvec{e}}_{9} + {\varvec{B}}_{m} {\varvec{e}}_{11} . \hfill \\ \end{gathered}\)
Using Lemma 4, there exist positive scalar \(\varepsilon_{m}\), \(\varepsilon_{n}\)\(\left( {1 \le m < n \le r} \right)\) such that
Therefore, (59) and (60) are equivalent to
where \(\begin{aligned}&{{\boldsymbol{\varPi}}}_{14} = \left[ {\begin{array}{*{20}l} {\varepsilon_{m} {\varvec{e}}_{1}^{T} {\varvec{PG}}_{m} } & {{\varvec{e}}_{1}^{T} {\varvec{E}}_{m}^{T} } \\ {\tau_{{\text{M}}} \varepsilon_{m} {\varvec{RG}}_{m} } & {\varvec{0}} \\ {\varvec{0}} & {\varvec{0}} \\ \end{array} } \right],\\&{\tilde{\boldsymbol{\varPi }}}_{14} = \left[ {\begin{array}{*{20}c} {\varepsilon_{m} {\varvec{e}}_{1}^{T} {\varvec{PG}}_{m} } & {{\varvec{e}}_{1}^{T} {\varvec{E}}_{m}^{T} } & {\varepsilon_{n} {\varvec{e}}_{1}^{T} {\varvec{PG}}_{n} } & {{\varvec{e}}_{1}^{T} {\varvec{E}}_{n}^{T} } \\ {\tau_{{\text{M}}} \varepsilon_{m} {\varvec{RG}}_{m} } & {\varvec{0}} & {\tau_{{\text{M}}} \varepsilon_{n} {\varvec{RG}}_{n} } & {\varvec{0}} \\ {\varvec{0}} & {\varvec{0}} & {\varvec{0}} & {\varvec{0}} \\ \end{array} } \right].\end{aligned}\)
Define \({\varvec{X}} = {\varvec{P}}^{ - 1}\). Multiplying the left- and right-hand sides of (21) by \({\text{diag}}\left\{ {{\varvec{X}},{\varvec{X}},{\varvec{X}},{\varvec{X}},{\varvec{X}},} \right.\)\(\left. {\varvec{X}} \right\}\), one can obtain (25). Similarly, multiplying the left- and right-hand sides of inequality (62) by \({\text{diag}}\left\{ {{\varvec{X}},\,{\varvec{X}},\,{\varvec{X}},\,{\varvec{X}},\,{\varvec{X}},\,{\varvec{X}},\,{\varvec{X}},\,{\varvec{X}},\,{\varvec{I}},\,{\varvec{X}},\,{\varvec{I}},\,{\varvec{R}}^{ - 1} ,\,{\varvec{I}},\,{\varvec{I}},\,{\varvec{I}},\,{\varvec{I}},\,{\varvec{I}}} \right\}\) and inequality (63) by \({\text{diag}}\left\{ {{\varvec{X}},{\varvec{X}},{\varvec{X}},{\varvec{X}},{\varvec{X}},{\varvec{X}},{\varvec{X}},{\varvec{X}},{\varvec{I}},{\varvec{X}},{\varvec{I}},{\varvec{R}}^{ - 1} ,{\varvec{I}},{\varvec{I}},{\varvec{I}},{\varvec{I}},{\varvec{I}},{\varvec{I}},{\varvec{I}}} \right\}\). Define new matrix variables \({\overline{\varvec{Q}}} = {\varvec{XQX}}\), \({\overline{\varvec{R}}} = {\varvec{XRX}}\), \({\overline{\boldsymbol{\varPhi }}} = \left( {{\boldsymbol{\nu \varPhi }}} \right)^{ - 1}\), \({\overline{\varvec{U}}} = {\text{diag}}\left\{ {{\varvec{X}},{\varvec{X}},{\varvec{X}}} \right\} \cdot {\varvec{U}} \cdot {\text{diag}}\left\{ {{\varvec{X}},{\varvec{X}},{\varvec{X}}} \right\}\). Let \({{\boldsymbol{\Delta}}}\) stands for \({\overline{\varvec{R}}}\), \({\overline{\boldsymbol{\varPhi }}}\), and \({\varvec{I}}\), (58) becomes \(- {\varvec{R}}^{ - 1} = - {\varvec{X\overline{R}}}^{ - 1} {\varvec{X}} \le \vartheta_{2}^{2} {\overline{\varvec{R}}} - 2\vartheta_{2} {\varvec{X}}\), \(- {\varvec{X}}{\overline{\boldsymbol{\varPhi }}}^{ - 1} {\varvec{X}} \le\)\(\vartheta_{3}^{2} {\overline{\boldsymbol{\varPhi }}} - 2\vartheta_{3} {\varvec{X}}\), and \(- {\varvec{XX}} \le \vartheta_{4}^{2} {\varvec{I}} - 2\vartheta_{4} {\varvec{X}}\), respectively. Using Schur complement, (27) is equivalent to \({\varvec{X}}{\overline{\boldsymbol{\varPhi }}}^{ - 1} {\varvec{X}} < \varepsilon_{0} {\varvec{I}}\). Then, by the Schur complement, (28) and (29) can be obtained easily. This completes the proof.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Wang, W., Peng, J., Xie, S. et al. Exponential stabilization of aero-engine T-S fuzzy system with decentralized dynamic event-triggered mechanism. Nonlinear Dyn 111, 21627–21646 (2023). https://doi.org/10.1007/s11071-023-08906-9
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11071-023-08906-9