Direct adaptive control using self recurrent wavelet neural network via adaptive learning rates for stable path tracking of mobile robots

2005 ◽  
Author(s):  
Sung Jin Yoo ◽  
Jin Bae Park ◽  
Yoon Ho Choi
Keyword(s):  
Neural Network ◽  
Adaptive Control ◽  
Mobile Robots ◽  
Adaptive Learning ◽  
Wavelet Neural Network ◽  
Path Tracking ◽  
Direct Adaptive Control ◽  
Learning Rates ◽  
Stable Path

A Self-Constructing Wavelet Neural Network Controller to Mitigate the Subsynchronous Oscillations

2012 ◽  
Vol 135 (2) ◽  
Author(s):  
Mohsen Farahani ◽  
Soheil Ganjefar
Keyword(s):  
Neural Network ◽  
Power Systems ◽  
Adaptive Learning ◽  
Control Process ◽  
Wavelet Neural Network ◽  
Intelligent Control System ◽  
Intelligent Technique ◽  
Neural Network Controller ◽  
Learning Rates ◽  
On Line

This study proposes a new intelligent controller based on self-constructing wavelet neural network (SCWNN) to suppress the subsynchronous resonance (SSR) in power systems compensated by series capacitors. In power systems, the use of intelligent technique is inevitable, because of the uncertainties such as operating condition variations, different kinds of disturbances, etc. Accordingly, an intelligent control system that is an on-line trained SCWNN controller with adaptive learning rates is used to mitigate the SSR. The Lyapunov stability method is used to extract the adaptive learning rates. Hence, the convergence of the proposed controller can be guaranteed. At first, there is no wavelet in the structure of controller. They are automatically generated and begin to grow during the control process. In the whole design process, the identification of the controlled plant dynamic is not necessary according to the ability of the proposed controller. The effectiveness and robustness of the proposed controller are demonstrated by using the simulation results.

Sign in / Sign up

Export Citation Format

Share Document