文献类型：期刊文献

英文题名：An accelerated neural dynamics model for solving dynamic nonlinear optimization problem and its applications

作者：Fu, Dongyang[1];Si, Yang[1];Wang, Difeng[2];Xiong, Yizhen[1]

机构：[1]Guangdong Ocean Univ, Sch Elect & Informat Engn, Zhanjiang 524088, Peoples R China;[2]Minist Nat Resources, Inst Oceanog 2, State Key Lab Satellite Ocean Environm Dynam, Hangzhou 310012, Peoples R China

年份：2024

卷号：180

外文期刊名：CHAOS SOLITONS & FRACTALS

收录：SCI-EXPANDED(收录号：WOS:001173747200001)、、EI(收录号：20240615508665)、Scopus(收录号：2-s2.0-85183949444)、WOS

基金：This research was funded in part by the National Key Research and Development Program of China (No. 2022YFC3103101) , the Key Special Project for Introduced Talents Team of Southern Marine Science and Engineering Guangdong Laboratory, China (GML2021GD0809) , the National Natural Science Foundation of China (No. 42206187) .

语种：英文

外文关键词：Dynamic nonlinear optimization; Accelerated neural dynamics (AND); Acoustic localization

外文摘要：Zeroing neural dynamics (ZND) model is a powerful tool for solving dynamic problems. This study presents an accelerated neural dynamics (AND) model by solving a dynamic nonlinear optimization (DNO) problem. Different from the classical activation function (AF), the AND model describes a novel accelerated convergence strategy that designs a nonlinear dynamic variable according to the error paradigm. Additionally, the AND model can be converted into a paradigm-based dynamical mode, which provides a quantification of the convergence time. Notably, the AND model shows outstanding robustness to various perturbations in the computational environment. The superiority of the AND model is further validated by comparing different models. Subsequently, the model's practicality is shown through the utilization of acoustic-based time difference of arrival (TDOA) localization.