油圧モータ・負荷系へのニューラルネットの適用 : 第2報,微小動作信号のパルス幅変調による油圧モータ・負荷系の高精度角度制御(機械力学,計測,自動制御)

概要

論文の詳細を見る
In the previous report, we showed the application of the Neural Network for the angular position control of the hydraulic motor/load system with undesirable dead zone. This was on the condition that the command signals were not as small as the magnitudes corresponding to the actuating signals within dead zone. This paper deals with the effectiveness of the Neural Network compensator for the angular position control. The small actuating signals are converted to PWM signals in order to operate on these signals in the region of the inherent undesirable dead zone. A procedure to decide the magnitude of PWM signal, the carrier frequency and the adjusting gain to rectify actuating signals, is discussed by using the simulation analysis. Then, it is experimentally confirmed that the proposed angular position control system shows extremely precise control and robustness even in small command angles using adequate PWM signals and adjusting gain.
一般社団法人日本機械学会の論文
2002-06-25