New stability and stabilization conditions for stochastic neural networks of neutral type with Markovian jumping parameters☆
Introduction
It is well known that many physical systems are subject to frequent unpredictable structural changes, such as random failures [1], repairs of components [2], sudden environment disturbances [3], abrupt variation of the operating point on a nonlinear plant [4], etc. Markovian jump systems are often used to describe such systems. Generally speaking, a Markovian jump system is a hybrid system with state vector that has two components x(t) and r(t), where x(t) denotes the state and r(t) is a continuous-time Markov chain, which is usually regarded as the mode. In its operation, the jump system will switch from one mode to another in a random way, which is determined by a continuous-time Markov chain r(t), based on a Markov chain with finite state space . Therefore, it is important to research the dynamics behaviors of systems with Markovian jumping parameters.
Recently, a large number of stability and stabilization results on stochastic systems with Markovian jumping parameters have been reported in the literature, see [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30] and the references therein. In [5], [6], Ji and Chizeck and Mariton studied the stability of linear jump equations; Boukas and Yang studied the exponential stability problem for stochastic systems with Markovian jumping parameters in [7]. Mao investigated the exponential stability for general nonlinear stochastic differential equations with Markovian jumping parameters [8]. The robustly exponential stability of stochastic neural networks with mixed time delays and Markovian switching was investigated in [9]. Balasubramaniam and Rakkiyappan [10] studied the globally asymptotic stability for a class of Markovian jumping stochastic Cohen–Grossberg neural networks with discrete interval and distributed delays. Very recently, the exponential stability for several new classes of Markovian jump stochastic neural networks with mixed time delays with/without impulse control were discussed in [11], [12], [13]. The globally asymptotic stability for a class of stochastic recurrent neural networks with Markovian jumping parameters and continuously distributed delays were investigated in [14]. In the area of control, the linear quadratic regulator problem were studied in [15], [16]. Mariton and Bertrand [17] extended those quadratic regulator to the case of output feedback. The robust stability and stabilization of linear jump systems were studied in [18]. Ghosh et al. [19] developed a dynamic programming approach to the optimal control of general stochastic differential equations with Markovian switching. Recently, Yuan and Lygeros [20] considered the problem of exponential stabilization by the state feedback for a class of stochastic differential equations with Markovian jumping parameters. In [27], the trajectory tracking control problem of a class of non-affine stochastic nonlinear switched systems with the non-lower triangular form under arbitrary switching was considered. Very recently, Li and Yang [30] studied the adaptive event-triggered control problem of nonlinear continuous-time systems in strict-feedback form.
On the other hand, neural networks of neutral type have been widely discussed since this model can be extensively used in the area of computer aided design, circuit analysis, real time simulation of mechanical systems, chemical process simulation, power systems, population dynamics and automatic control. For instance, Park and Kwon [31] studied the delay-dependent stability criterion for a class of neural networks of neutral-type with interval time-varying delays by the Lyapunov stability theory and linear matrix inequality approach. In [32], the authors obtained some delay-dependent conditions for a class of neutral type neural networks with both discrete and unbounded distributed delays. Park et al. [33] proved the globally asymptotic stability of equilibrium for a class of continuous bidirectional associative memory neural networks of neutral type. Mahmoud and Ismail [34] derived the robustly exponential stability condition for a class of neutral-type delayed neural networks via using the Lyapunov–Krasovskii functional and the integral inequality. In [35], the authors considered the adaptive pinning synchronization problem for neutral-type neural networks with stochastic perturbation. However, the results in [31], [32], [33], [34], [35] did not consider neural networks of neutral type with Markovian jumping parameters. The delay-dependent stability analysis for stochastic neural networks of neutral type with Markovian jumping parameters was considered in [36], [37]. The almost surely exponential stability for neutral stochastic neural networks with constant delay was discussed by using the semi-martingale convergence theorem in [38], [39]. However, the almost surely exponential stability analysis for stochastic systems with neural networks of neutral type with Markovian jumping parameters has not yet been investigated in [38], [39], which prompt our present research. In addition, to the best of our knowledge, a few little stabilization results on stochastic neural networks of neutral type with Markovian jumping parameters were reported [40], [41]. Therefore, it may exist large space to develop novel stabilization conditions for stochastic systems with neural networks of neutral type with Markovian jumping parameters. All of those motivate this research.
Motivated by the aforementioned discussions, the objective of this paper is to further study the stability and stabilization for a new class of stochastic neural networks of neutral type with both Markovian jump parameters and mixed time delays. The main contributions are summarized as follows.
- (1)
By employing the semi-martingale convergence theorem, some delay-dependent sufficient criteria to guarantee the globally exponential stability in mean square and almost surely exponential stability conditions are both given.
- (2)
Based on the obtained less conservative stability conditions, the state feedback controller is designed by the linear matrix inequality technique.
- (3)
Compared with some existing stability criteria, our derived criteria are not only less conservative but also easily checked.
Notation: Throughout this paper, the following notations will be used. and denote, respectively, the n-dimensional Euclidean space and the set of all n × n real matrices. The superscript T denotes the transposition and the notation X ≥ Y (respectively, X > Y), where X and Y are symmetric matrices, means that is positive semi-definite (respectively, positive definite). λmax (X) and λmin (X) denote the maximum and minimum eigenvalues of the corresponding matrix X, respectively. . col{⋅⋅⋅} denotes a column vector. E{ · } stands for the mathematical expectation operator. In is the n × n identity matrix. | · | denotes the absolute value of the real number and the norm of a vector or a matrix.
Section snippets
Problem description and preliminaries
Let {r(t), t ≥ 0} be a right-continuous Markov process on the probability space taking values in a finite state space with producer given bywhere Δt > 0 and πij ≥ 0 is the transition rate from i to j, if i ≠ j while for each mode i.
In this paper, we will consider the following stochastic neural networks of neutral type with both Markovian switching and mixed time delays:
Main results
In this section, by constructing an appropriate Lyapunov–Krasovskii functional, the exponential stability in mean square and the almost surely exponential stability for Markovian jumping stochastic neural networks of neutral type with mixed time delays (3) will be studied. As a matter of convenience, define a new state variable:and a new disturbance variable:Thus, system
Numerical examples
In this section, the following examples are provided to show the effectiveness of the theoretical results. Example 1 Consider the two-dimensional stochastic neural networks of neutral type with both Markovian jumping parameters and mixed time delays (3) with the following parameters as in [47]:
Conclusion
In this paper, we have introduced and studied the globally exponential stability in mean square and almost surely exponential stability for a new class of stochastic neural networks of neutral type with both Markovian jump parameters and mixed time delays. The obtained results are shown to be much less conservative by constructing a new Lyapunov–Krasovskii functional and with the help of the semi-martingale convergence theorem. Our delay-dependent sufficient conditions obtained are less
References (47)
- et al.
Exponential stabilizability of stochastic systems with Markovian jumping parameters
Automatica
(1999) Stability of stochastic differential equations with Markovian switching
Stoch. Process. Appl.
(1999)- et al.
Delay-dependent robust stability analysis for Markovian jumping stochastic Cohen–Grossberg neural networks with discrete interval and distributed time-varying delays
Nonlinear Anal. Hybrid Syst.
(2009) - et al.
Ergodic control of switching diffusions
SIAM J. Control Optim.
(1997) - et al.
Mean-square exponential stability for a class of discrete-time nonlinear singular Markovian jump systems with time-varying delay
J. Franklin Inst.
(2014) - et al.
Stability analysis for neutral Markovian jump systems with partially unknown transition probabilities
J. Frankl. Inst.
(2012) - et al.
Delay-dependent robust exponential stability of Markovian jumping reaction–diffusion Cohen–Grossberg neural networks with mixed delays
J. Frankl. Inst.
(2012) - et al.
Stability of switched positive linear systems with average dwell time switching
Automatica
(2012) - et al.
Intelligent tracking control for a class of uncertain high-order nonlinear systems
IEEE Trans. Neural Netw. Lear. Syst.
(2016) - et al.
Observer-based fuzzy adaptive event-triggered control co-design for a class of uncertain nonlinear systems
IEEE Trans. Fuzzy Syst.
(2017)
Global stability for neural networks of neutral-type with interval time-varying delays
Chaos Solitons Fract.
A new stability criterion for bidirectional associative memory neural networks of neutral-type
Appl. Math. Comput.
Improved results on robust exponential stability criteria for neutral-type delayed neural networks
Appl. Math. Comput.
Pinning adaptive synchronization of neutral-type coupled neural networks with stochastic perturbation
Adv. Differ. Equ.
Globally exponential stability of stochastic neutral type delayed neural networks with impulsive perturbations and markovian switching
Nonlinear Dyn.
Novel delay-dependent robust stability criteria for neutral stochastic delayed neural networks
Neurocomputing
Adaptive exponential stabilization of neutral-type neural network with Lévy noise and Markovian switching parameters
Neurocomputing
Stochastic Differential Equations and Applications
Stability of Time-delay Systems
Linear Matrix Inequalities in Systems and Control Theory
Stochastic stability analysis for continuous-time fault tolerant control systems
Proceedings of the American Control Conference
Optimal preventive maintenance with repair
IEEE Trans. Rel.
An LQG solution to a control problem with solar thermal receiver
IEEE Trans. Autom. Control
Cited by (35)
Synchronization for stochastic semi-Markov jump neural networks with dynamic event-triggered scheme
2023, Journal of the Franklin InstituteStability analysis of split-step theta method for neutral stochastic delayed neural networks
2023, Journal of Computational and Applied MathematicsCitation Excerpt :Furthermore, mathematical models are represented as some neutral differential equations that rely on delays in states and state derivative in certain systems. Thus, researches have devoted themselves into stability problems of neutral stochastic delayed neural networks (NSDNNs) [14–17]. The follows are the rest arrangements of this paper.
Stochastic stability criteria and event-triggered control of delayed Markovian jump quaternion-valued neural networks
2022, Applied Mathematics and ComputationHidden Markov model-based asynchronous quantized sampled-data control for fuzzy nonlinear Markov jump systems
2022, Fuzzy Sets and SystemsCitation Excerpt :Various successful applications of Markov jump systems can be seen in modern communication technology, such as robotic manipulators, fault tolerance control and so on. Following this, many the related theoretical achievements have been reported [5–8]. It is common knowledge that, in many fields of engineering control, the control issue for nonlinear Markov jump systems is always complicated and challenging.
Stochastic stability and extended dissipativity analysis for delayed neural networks with markovian jump via novel integral inequality
2022, Journal of the Franklin InstituteCitation Excerpt :Therefore, it is necessary to study the dynamical actions of NNs with time-varying delay. Markovian jump phenomenon is an important random and uncertain phenomenon, which can be used to describe the structure or parameter changes of the system due to malfunction of internal components or sudden changes in the external environment [6,7]. Therefore, it has been widely utilized in economic systems, power systems, flight system, network control system and some other related systems [8–10].
Sampled-data synchronization criteria for Markovian jumping neural networks with additive time-varying delays using new techniques
2022, Applied Mathematics and ComputationCitation Excerpt :As a consequence, broad applications of neural networks are found in multifarious areas, such as signal and image processing, financial market, combinatorial optimization, and fixed-point computations [1,2]. On the other hand, an extremely fascinating phenomenon in the investigation in the investigation of distinct aspects of neural networks is that under the restrictions of equipment and the influence of unknown environmental disturbances, the parameters and structure of neural networks will occur unpredictable change [3–8]. In view of this, there exist finite modes in the neural networks, which can switch from one to another at distinct moments subjected to a stochastic process, namely, Markovian process.
- ☆
This work was supported by National Natural Science Foundation under Grant 11461082, 11601474, 61573096, 61472093 and 61463050, supported by the Jiangsu Provincial Key Laboratory of Networked Collective Intelligence under Grant No. BM2017002, the key laboratory of numerical simulation of Sichuan Province, under Grant No. 2017KF002, Yunnan Provincial Department of Education Science Research Fund Project under Grant 2018Y105, the NSF of Yunnan Province No. 2015FB113.