Stability and boundedness criteria for certain second-order nonlinear neutral stochastic functional differential equations

A. T. Ademola, Shiping Wen, Yuming Feng, Wei Zhang, Leszek RutkowskiProyecciones (Antofagasta) 2024 · DOI

Proyecciones Journal of Mathematics

Stability and Boundedness Criteria for Certain Second-Order Nonlinear Neutral Stochastic Functional Differential Equations

Abstract

This paper presents stochastic stability and stochastic boundedness to certain second-order nonlinear neutral stochastic differential equations. The second-order differential equation is weakened to a neutral stochastic system of first-order equations and used together with a second-order quadratic function to obtain perfect Lyapunov-Krasovskii functional. This functional is adapted and applied to obtain criteria on the nonlinear functions to ensure novel results on stochastic stability and stochastic asymptotic stability of the zero solution. Furthermore, when the forcing term is nonzero, fresh results on stochastic boundedness and uniform stochastic boundedness of solutions are obtained. The results of this paper are original, new, essentially improving, complementing, and simplifying several related ones in the literature. Two special cases of the theoretical results are supplied to demonstrate the applicability of the hypothetical results.

Introduction

In the modeling of complex real-world systems—ranging from predator-prey populations to high-frequency trading algorithms—dynamics are rarely instantaneous or linear. Most sophisticated systems involve acceleration (second-order dynamics), non-linear interactions, and inherent environmental noise (stochasticity). Furthermore, many systems exhibit "neutral" delays, where the current state depends not only on past states but also on the past rate of change (the derivative).

While stability for simpler differential equations is well-documented, neutral stochastic functional differential equations present significant analytical challenges. This research, published in the Proyecciones Journal of Mathematics, addresses this gap by establishing explicit mathematical criteria to predict when these noisy, delayed systems will remain stable or stay within safe operating bounds.

Methodology

The authors utilize the Lyapunov Direct Method, a cornerstone of stability theory that avoids the often-impossible task of solving complex differential equations directly. Instead, the method involves identifying a "Lyapunov functional"—a mathematical representation of the system's "energy."

Key methodological steps include:

  1. System Transformation: The complex second-order equation is decomposed into a more manageable system of two first-order equations.
  2. Functional Construction: The researchers designed a novel Lyapunov-Krasovskii functional. This functional is specifically engineered to account for both the time delay and the influence of random noise.
  3. Itô’s Calculus: By applying Itô’s formula, the authors differentiate their functional within a random environment to prove that, under specific conditions, the average energy of the system dissipates over time, ensuring stability.

Key Insights and Results

Addressing the "Neutral" Challenge

Unlike standard delay models that only consider past positions, this work handles delays in velocity (the neutral term). This makes the model far more applicable to mechanical and biological systems where momentum and reaction times are critical.

Stability and Boundedness Criteria

  • Stochastic Asymptotic Stability: Theorems 3.1 and 3.2 establish the conditions under which a system will eventually settle to zero despite continuous noise. This occurs if the nonlinear functions and noise intensity satisfy specific mathematical inequalities.
  • Stochastic Boundedness: For systems influenced by an external force, Theorems 4.1 and 4.2 prove that if the force is bounded, the system's response will also remain within a predictable range rather than diverging to infinity.

Practical Application

The theoretical results are "backwards compatible" with classical linear systems, such as the Routh-Hurwitz criteria. To demonstrate practical utility, the authors provide numerical simulations. For example, they calculate specific maximum allowable delays (e.g., $\pi \approx 0.0621$ units of time) for a system with defined damping and noise parameters to remain stable.

Conclusion and Future Directions

While the criteria provided are "sufficient" (offering a guaranteed safe zone for system parameters), they may be conservative. A system might remain stable even if it falls slightly outside these bounds. Future research is expected to extend these techniques to third-order neutral stochastic equations, further expanding the toolkit for engineers and mathematicians working with complex, non-deterministic systems.

Summary updated

Abstract

This paper presents stochastic stability and stochastic boundedness to certain second-order nonlinear neutral stochastic differential equations. The second-order differential equation is weakened to a neutral stochastic system of first-order equations and used together with a second-order quadratic function to obtain perfect Lyapunov-Krasovskii functional. This functional is adapted and applied to obtain criteria on the nonlinear functions to ensure novel results on stochastic stability and stochastic asymptotic stability of the zero solution. Furthermore, when the forcing term is nonzero, fresh results on stochastic boundedness and uniform stochastic boundedness of solutions are obtained. The results of this paper are original, new, essentially improving, complementing, and simplifying several related ones in the literature. Two special cases of the theoretical results are supplied to demonstrate the applicability of the hypothetical results.
BibTeX
@inproceedings{ademola2024stability,
  author = {A. T. Ademola and Shiping Wen and Yuming Feng and Wei Zhang and Leszek Rutkowski},
  title = {{Stability and boundedness criteria for certain second-order nonlinear neutral stochastic functional differential equations}},
  booktitle = {Proyecciones (Antofagasta)},
  year = {2024},
  doi = {10.22199/issn.0717-6279-6146},
}
Figure