Mean square delay dependent-probability-distribution stability analysis of neutral type stochastic neural networks

Area of Science:

• Control Systems Engineering
• Computational Neuroscience
• Applied Mathematics

Background:

• Stochastic neural networks with time-delays are crucial in modeling complex dynamical systems.
• Ensuring stability in the presence of interval time-varying and random delays is a significant challenge.
• Robustness against uncertainties is essential for practical applications.

Purpose of the Study:

• To investigate the mean square delay-dependent stability of neutral-type stochastic neural networks with time-delays.
• To develop a novel sufficient condition for global robust asymptotic stability.
• To analyze the impact of interval time-varying and randomly occurring delays.

Main Methods:

• Utilizing a novel Lyapunov-Krasovskii functional.
• Employing a stochastic analysis approach.
• Formulating stability conditions in the form of linear matrix inequalities (LMIs).

Main Results:

• A new sufficient condition for global robust asymptotic stability in the mean-square sense was derived.
• The condition guarantees stability for all admissible uncertainties.
• The analysis accounts for interval time-varying and randomly occurring delays.

Conclusions:

• The proposed method provides a reliable framework for stability analysis of delayed stochastic neural networks.
• Numerical examples validate the theoretical findings and demonstrate the calculation of maximum allowable delay bounds.
• The results contribute to the robust control design for complex neural network systems.