作者: Juri Belikov , Eduard Petlenkov , Kristina Vassiljeva , Sven Nõmm
DOI: 10.1007/978-3-642-42054-2_28
关键词:
摘要: The paper describes a unified algorithm for both parametric and structural identification. approach combines three typical techniques such as neural networks, statistics genetic algorithm. A specific structure of the network is used that allows to design controller directly from parameters identified model. control strategy based on reference model discussed. Finally, proposed solution illustrated by numerical example.