Dissertation > Excellent graduate degree dissertation topics show

Pulse Variable Delay static dynamic behavior of the neural network

Author: SunBianNi
Tutor: GaoXingBao
School: Shaanxi Normal University
Course: Computational Mathematics
Keywords: Static Neural Networks Varying Delay Pulse Asymptotic stability Exponential Stability Delay dependence
CLC: TP183
Type: Master's thesis
Year: 2011
Downloads: 13
Quote: 0
Read: Download Dissertation

Abstract

In recent years, the neural network has been widely used in the associative memory, pattern recognition, signal processing, combinatorial optimization, image processing, intelligent control and many other fields. Different selected according to the basic variables of the system, the neural network can be divided into local neural network and static neural network of local neural networks in the field of optimization calculation and associative memory made successful applications and static neural networks have significant advantages in solving variational inequalities optimization problem in real life, the operation of the neural network may by both the one hand, due to the the limited conversion speed of neurons amplifier, often by Delays Delay may cause oscillation or instability of the system. On the other hand, the voltage mutation will produce the wrong circuit This is typical pulse phenomenon, it can affect the transient behavior of the neural network. study neural network, consider the impact of delay and pulse is very necessary. therefore, this article will study at the same time by the impact of delay and pulse static the dynamical behavior of the neural network., the first review of the development and characteristics of the neural network analysis of the time delay and the pulse of the neural network. Secondly, influenced by the pulse and time delay neural network research in recent years and results, and summarizes the main findings of this paper. second chapter, the prior knowledge, including definitions, theorems and inequalities. then stability theory, and finally gives the pulse with variable delays affect the static neural network model. third chapter, the asymptotic stability of static neural network research varying delay case containing pulse by constructing appropriate Lyapunov functional and using linear matrix inequalities method to prove the model delay dependent conditions under the global asymptotic stability due to taking into account the impact of delay and pulse, and thus concluded more universal, applicable to a wider range. Finally, a simulation example illustrates the results for a wide range of conservative small, easy to verify features. exponential stability of static neural network research varying delay in the pulse-containing case. constructed two Lyapunov functional and using linear matrix inequalities, ensure that the neural network model Two sufficient conditions for global exponential stability of delay dependence under conditions compared with the existing results, the stability criterion given easy to verify, less conservative, applicable to a wide range of features. Finally, simulation example the correctness of the results obtained.

Related Dissertations

CLC: > Industrial Technology > Automation technology,computer technology > Automated basic theory > Artificial intelligence theory > Artificial Neural Networks and Computing
© 2012 www.DissertationTopic.Net Mobile