کد مقاله کد نشریه سال انتشار مقاله انگلیسی نسخه تمام متن
6863025 1439402 2018 14 صفحه PDF دانلود رایگان
عنوان انگلیسی مقاله ISI
Delay-dependent dynamical analysis of complex-valued memristive neural networks: Continuous-time and discrete-time cases
ترجمه فارسی عنوان
تجزیه و تحلیل دینامیکی وابسته به تاخیر از شبکه های عصبی مرکب ارزش پیچیده: موارد مداوم و زمان گسسته
کلمات کلیدی
ممریستور، شبکه عصبی پیچیده ارزشمند، توابع فعال متوقف شده، نابرابری ماتریس، ثبات وابسته به تاخیر،
موضوعات مرتبط
مهندسی و علوم پایه مهندسی کامپیوتر هوش مصنوعی
چکیده انگلیسی
This paper considers the delay-dependent stability of memristive complex-valued neural networks (MCVNNs). A novel linear mapping function is presented to transform the complex-valued system into the real-valued system. Under such mapping function, both continuous-time and discrete-time MCVNNs are analyzed in this paper. Firstly, when activation functions are continuous but not Lipschitz continuous, an extended matrix inequality is proved to ensure the stability of continuous-time MCVNNs. Furthermore, if activation functions are discontinuous, a discontinuous adaptive controller is designed to acquire its stability by applying Lyapunov-Krasovskii functionals. Secondly, compared with techniques in continuous-time MCVNNs, the Halanay-type inequality and comparison principle are firstly used to exploit the dynamical behaviors of discrete-time MCVNNs. Finally, the effectiveness of theoretical results is illustrated through numerical examples.
ناشر
Database: Elsevier - ScienceDirect (ساینس دایرکت)
Journal: Neural Networks - Volume 101, May 2018, Pages 33-46
نویسندگان
, , , ,