Please use this identifier to cite or link to this item:
標題: 適應性幅射基底類神經網路於線型直流無刷馬達之運動控制與倒單擺控制
Adaptive Motion and Inverted Pendulum Control with RBF Neural Network for a Linear DC Brushless Motor
作者: 程代琛
Chen, Dai-Cheng
關鍵字: Adaptive Control;適應性控制;Linear Motor;Radial Basis Function Neural Network;Backstepping Control;Inverted Pendulum;線型馬達;幅射基底函數類神經網路;倒逆步控制;倒單擺
出版社: 電機工程學系
本文旨在提出一線型直流無刷馬達之適應控制器,並發展一套以個人電腦為基礎之控制器實現技術。本文利用簡化的線型直流無刷馬達數學模為基礎,並假設馬達漣波力和非線性行為摩擦力為有界的條件,結合RBF NN網路,提出整合的非線性適應NN控制器來達成速度追蹤控制;其次以適應倒逆步RBF NN控制器達成線型馬達位置追蹤控制;最後,則以適應倒逆步RBF NN法來完成倒單擺之垂直姿態控制。 電腦模擬及實驗結果皆顯示本文所提出的控制法,對於具明顯非線性行為的線型直流無刷馬達的速度追蹤及位置追蹤為可行及有效的控制策略,對於倒單擺的控制同樣也有良好的效果。

This thesis aims at proposing two adaptive motion control methods with radial- basis-function neural network (RBF NN) for a linear DC brushless motor, and one adaptive control method for an inverted pendulum mounted on the linear motor. A simplified model of a linear motor with a current driver is adopted herein. Based on some upper bound assumptions of the motor ripple force and the nonlinear friction force, an adaptive RBF NN control is proposed to achieve velocity tracking control. With the same control concept, an adaptive backstepping RBF NN control is presented to achieve position tracking. Moreover, the adaptive backstepping RBF NN control approach is also utilized to regulate the inverted pendulum mounted on the linear motor. Computer simulation and experimental results are conducted to show the feasibility and effectiveness of the proposed control schemes for velocity tracking, position tracking and inverted pendulum control.
Appears in Collections:電機工程學系所

Show full item record

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.