Abstract:In this paper, the stabilization is studied for a complex dynamic model which involves nonlinearities, uncertainty, and Lévy noises. This paper also discusses the controller discretization and presents a new algorithm to obtain the upper bound for the sample interval through which the exponential stability of the discrete system can still be guaranteed. Firstly, an integral sliding surface is designed to obtain the sliding mode dynamics for the considered stochastic Lévy process. By using Lyapunov theory, generalized Itô formula and some inequality techniques, the exponential stability is proved in the sense of mean square for sliding mode dynamics. The reachability of the sliding mode surface is also ensured by designing a sliding mode control law. Secondly, the continuous-time controller is discretized from the point of control cost, and the squared difference is analyzed for the states before and after the discretization. Different from those classical stochastic differential equations driven by Brownian motions, the noise is supposed to be Lévy type and the squared difference is analyzed in different cases. Furthermore, we obtain the largest sampling interval through which the discretized controller can still stabilize the Lévy process driven stochastic system. Finally, a simulation for a drill bit system is given to demonstrate the results under the algorithms.

Refraction strategies in stochastic control: optimality for a general Lévy process model

On stochastic control under Poisson observations: optimality of a barrier strategy in a general Lévy model

Optimality of a refraction strategy in the optimal dividends problem with absolutely continuous controls subject to Parisian ruin

De Finetti's control problem with a concave bound on the control rate

Optimal Multiple Stopping with Negative Discount Rate and Random Refraction Times under Levy Models

An Optimal Multibarrier Strategy for a Singular Stochastic Control Problem with a State-Dependent Reward

Optimality of a barrier strategy in a spectrally negative Lévy model with a level-dependent intensity of bankruptcy

Stochastic Near-Optimal Controls for Path-Dependent Systems

Alternative approach to the optimality of the threshold strategy for spectrally negative Levy processes

Stochastic optimal control of Lévy tax processes with bailouts

Long Run Stochastic Control Problems with General Discounting

Optimal withdrawals in a general diffusion model with control rates subject to a state-dependent upper bound

On optimal control of reflected diffusions

Discrete-Time Approximation of Risk-Averse Control Problems for Diffusion Processes

Optimal control of diffusion processes: $\infty$-order variational analysis and numerical solution

On a Class of Singular Stochastic Control Problems for Reflected Diffusions

Discrete Control of Nonlinear Stochastic Systems Driven by Lévy Process

Optimal control of conditioned processes with feedback controls

On optimal joint reflective and refractive dividend strategies in spectrally positive Lévy models

Stochastic control problems with state-reflections arising from relaxed benchmark tracking