Steady state security analysis using artificial neural network
Steady state security analysis aims at assessing the risk a contingency would entail for an electrical network operating at a certain point. System operators' expertise and even human intuition in many ways are successful at assessing the risk a contingency would pose to a network. The objectiv...
Saved in:
| Main Author: | |
|---|---|
| Format: | Student Project |
| Language: | en |
| Published: |
2008
|
| Subjects: | |
| Online Access: | https://ir.uitm.edu.my/id/eprint/125637/1/125637.pdf https://ir.uitm.edu.my/id/eprint/125637/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Steady state security analysis aims at assessing the risk a contingency would entail for an electrical network operating at a certain point. System operators' expertise and even human intuition in many ways are successful at assessing the risk a contingency would pose to a network. The objective of the project is to describe how artificial neural networks can be used to bypass the traditional load flow cycle, resulting in significantly faster computation times for online contingency analysis. The cases where operating violations are observed are considered as alert , while the cases for which the load flow algorithm exhibits a diverging algorithmically solution, are considered as emergency. The most important task in real time security analysis is the problem of identifying the critical contingencies from a large list of credible contingencies and ranks them according to their severity. The artificial neural network (ANN)-based approach for contingency ranking. A set of feed forward neural networks are developed to estimate the voltage stability level at different load conditions for the selected contingencies. The effectiveness of the proposed method has been demonstrated through contingency ranking in IEEE 30-bus system. The performance of the developed model is compared with the unified neural network trained with the full feature set. Simulation results show that the proposed method takes less time for training and has good generalization. |
|---|