|
A Knowledge Based Neuro-Fuzzy Model and Controller Synthesis of a Highly Nonlinear Dynamics Electrical Machine
|
|
*1 Ebrahim Mattar, 2 M. Akbaba
1, College of Engineering, University of Bahrain, P. O. Box 13184, Bahrain.
2, College of IT, University of Bahrain, P. O. Box 13184, Bahrain.
Email: 1ebmattar@ieee.org , 2itqoffice@uob.edu.bh
|
|
|
|
Abstract
.It is important to drive robotics systems with heavy duty machines. However, analysis and controller synthesis of an electrical rotating machine is considered as hard task to be achieved. This is due to the complicated and nonlinear differntional equations that govern such types of electro-mechanical systems. This article has been conducted to solve the issue of designing linear controllers (even with some Robust characters) for a class of nonlinear electrical motor. Initially a Takagi –Sugeno (T-S) Neuro-Fuzzy models are built while extracting machine sub-linear models. Local state feedback controllers are synthesized using some optimization tools. For designing the controller with some noise rejection characters, an was used, while solving LYAPUNOV candidate function using LMI formulation. The synthesized controller strategy has proven as an effective in terms of solving for optimal system algebraic Riccatti formulation, while relying on Neuro-Fuzzy sub models.
|
|
|
|
Keywords
:
Neuro-fuzzy Modeling ; Nonlinear Dynamics ; Patterns Clustering
|
|
|
|
URL: http://dx.doi.org/10.7321/jscse.v3.n11.2
|
|
|
|
|