ニューラルネットワークによる目標値自己生成型制御系の設計と磁気ディスク装置への応用

概要

論文の詳細を見る
This paper introduces a reference signal self-organizing control system using neural networks for nonlinear systems. First, we introduce a control structure which consists of a series combination of both a plant with a feedback loop and a neural network with a feedforward loop. In this structure, the desired signal for the neural network is plant input. The simulation results indicate that the neural network of the proposed system can compensate for the effect of nonlinear terms in the plant. This structure can be extended to a closed loop structure which is a two-degree-of-freedom control system with a reference filter. In this structure, the neural network functions as the reference filter and it organizes a new reference signal to the closed loop system. Experimental results of the track-following control system of magnetic disk drives show that the proposed control system is effective in following the reference signal accurately.
社団法人日本機械学会の論文
1994-08-25