查詢結果分析
來源資料
相關文獻
- 適應性RBF類神經網路控制器應用於未知非線性離散系統
- An Adaptive Neural Network Position Controller for a PM Synchronous Servo Motor Drive
- Adaptive Output Tracking of a Class of Unknown Nonlinear Systems Using Neural Networks
- 可變排量泵之多重類神經網路適應控制
- 類神經網路應用於適應性控制之研究與設計
- 智能控制技術在特殊環境控制系統上之應用
- 直接類神經網路應用於負載變動直流伺服馬達速度調節適應控制及收斂性分析
- 智慧控制技術在特殊環境控制系統上之應用
- 類神經網路智能控制技術在空調通風系統上之應用
- 智慧型線性超音波馬達位置追蹤控制
頁籤選單縮合
題 名 | 適應性RBF類神經網路控制器應用於未知非線性離散系統=Adaptive RBF Neural Controller Design for a Class of Unknown Nonlinear Discrete-Time Systems |
---|---|
作 者 | 洪瑞鴻; 王致舜; 廖德祿; | 書刊名 | 技術學刊 |
卷 期 | 13:3 1998.09[民87.09] |
頁 次 | 頁521-528 |
分類號 | 312.2、312.2 |
關鍵詞 | 類神經網路; RBF網路; 適應控制; Neural networks; Radial basis function network; Adaptive control; |
語 文 | 中文(Chinese) |
中文摘要 | 本論文對某類未知非線性離散系統提出以適應性RBF類神經網路(Radial Basis Function Networks)與單層前向網路結合為基礎,並結合單層前向網路之one-step-ahead控 制器,以達到輸出追蹤之目的。所設計的控制系統同時保有適應控制理論與類神經網路特性 的優點。傳統的RBF類神經網路對於輸入信號變動非常敏感,因此本文提出K means clustering 的調變方式,以因應激發函數的中心點與寬度因輸入訊號的變動所產生的變化,並利用 least-square algorithm和dead zone函數調變網路權值,使整個閉迴路系統達到控制目的。最 後以電腦模擬的結果印証本文所提出的控制方式。 |
英文摘要 | In this paper, an one-step-ahead controller based on an adaptive RBF neural network combined with a one-layer feedforward network is derived for output tracking of a class of unknown nonlinear discrete time systems. The proposed controller design ingeniously takes the merits of both adaptive control techniques and neural networks' approximation capability. For conventional RBF networks, the locations of both centers and widths will directly affect the accuracy and quality of the overall network approximation. In this paper, the locations of the RBF network centers and widths are adaptively selected by the K means clustering algorithm based on the input signal space. In addition, the connection weights of the neural network are updated according to the least-square algorithm with dead zone function. The control objective of the closed-loop system can be achieved by the proposed control scheme. Finally, some simulation results are given to demonstrate the effectiveness of the proposed control scheme. |
本系統中英文摘要資訊取自各篇刊載內容。