| URN |
etd-0615115-141425 |
|---|
| Author |
Kai-Hung Lu |
|---|
| Author's Email Address |
No Public. |
|---|
| Statistics |
This thesis had been viewed 5567 times. Download 6 times. |
|---|
| Department |
Electrical Engineering |
|---|
| Year |
2014 |
|---|
| Semester |
2 |
|---|
| Degree |
Ph.D. |
|---|
| Type of Document |
|
|---|
| Language |
zh-TW.Big5 Chinese |
|---|
| Title |
Applications of Intelligent Controllers for FACTS to Transient Stability Study of Large-Scale Renewable Energy Integrated with Power System |
|---|
| Date of Defense |
2015-07-09 |
|---|
| Page Count |
153 |
|---|
| Keyword |
Flexible AC Transmission Systems
Genetic Ant Colony Optimization
Functional Link based Elman Neural Network
Renewable energy system
Functional Link based Novel Recurrent Fuzzy Neural Network
Adaptive Intelligent Control System
|
|---|
| Abstract |
The scale of nonlinearities and uncertainties of Renewable Energy System (RES) cause problems for dynamic control. This dissertation analyzes the dynamic and transient stability of the power system connected with a large RES including the analysis steady-state, dynamic and transient responses. An Adaptive Intelligent Control System (AICS) is proposed in the dissertation for a Static Synchronous Compensator (STATCOM) and a Unified Power Flow Controller (UPFC) to enhance the stability of the RES for power generation. An AICS can be used to increase the stability of the power control and improves the performance. Two models were proposed:
1. Integration of Critic neural network (CNN), Functional Link based Novel Recurrent Fuzzy Neural Network (FLNRFNN) and Genetic Algorithm Hybrid Time Varying Particle Swarm Optimization (GAHTVPSO) algorithm.
2. Integration of CNN, Functional Link based Elman Neural Network (FLENN) and genetic ant colony optimization algorithm (GACO).
The node connecting weights of the FLNRFNN, FLENN and CNN are trained online. The learning rates of the FLNRFNN, FLENN and CNN are usually selected by trial and error method and are time-consuming. The GAHTVPSO and GACO approach were developed to adjust the learning rates of FLNRFNN, FLENN and CNN to improve the learning rate. The validity of these algorithms was demonstrated with many simulations. The simulation results show that AICS can achieve better damping characteristics as well as transient stability in Flexible Alternating Current Transmission Systems (FACTS) applications. |
|---|
| Advisory Committee |
Jyh-Yih Hsu - chair
Yu-Chi Wu - co-chair
Gary-W. Chang - co-chair
Jen-Hao Teng - co-chair
Wei-Min Lin - advisor
|
|---|
| Files |
Indicate in-campus at 5 year and off-campus access at 5 year. |
|---|
| Date of Submission |
2015-07-15 |
|---|
