Interval Self-Organizing Map for Nonlinear System Identification and Control

Autor:	Luzhou Liu, Jian Xiao, Long Yu
Rok vydání:	2008
Předmět:	Self-organizing map Nonlinear system Inverse system Artificial neural network Nonlinear system identification Robustness (computer science) Control theory Unsupervised learning Gradient descent Algorithm Mathematics
Zdroj:	Lecture Notes in Computer Science ISBN: 9783540877318 ISNN (1)
DOI:	10.1007/978-3-540-87732-5_10
Popis:	The self-organizing map (SOM) is an unsupervised neural network which projects high-dimensional data onto a low-dimensional. A novel model based on interval self-organizing map(ISOM) whose weights are interval numbers presented in this paper differ from conventional SOM approach. Correspondingly, a new competition algorithm based on gradient descent algorithm is proposed according to a different criterion function defined in this paper, and the convergence of the new algorithm is proved. To improve the robustness of inverse control system, the inverse controller is approximated by ISOM which is cascaded with the original to capture composite pseudo-linear system. Simulation results show that the inverse system has superior performance of tracking precision and robustness.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::57f3f83fea6c2577f824a561dae9ba10 https://doi.org/10.1007/978-3-540-87732-5_10 Zobrazit plný text záznamu