Enhanced dynamic security assessment for power system under normal and fake tripping contingencies.
Recently, power system networks have become more dependent on new technologies especially in using a communication network to enhance the overall performance of system operation. The communication network facilities are applied to send and receive data and commands through the wide-area power networ...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English English English |
| Published: |
2019
|
| Subjects: | |
| Online Access: | http://eprints.uthm.edu.my/115/1/24p%20QUSAY%20A.%20SALIH.pdf http://eprints.uthm.edu.my/115/2/QUSAY%20A.%20SALIH%20COPYRIGHT%20DECLARATION.pdf http://eprints.uthm.edu.my/115/3/QUSAY%20A.%20SALIH%20WATERMARK.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Recently, power system networks have become more dependent on new technologies especially in using a communication network to enhance the overall performance of system operation. The communication network facilities are applied to send and receive data and commands through the wide-area power network. However, this dependency has opened a new threat of fake tripping contingency towards the power system operation. This challenge has motivated this study to ensure that all analytical tools applied during power system operation are not affected under fake tripping contingency, especially on dynamic security assessment (DSA) classifier. To address this challenge, this study aims to invistigate the impact of fake tripping contingency on the power system security via DSA classifier, then develop a novel hybrid approach for DSA classifier based on advanced feature selection technique for decision tree (DT) classifier and finally evaluate the performance of DSA classifier under normal and fake tripping contingencies, in terms of accuracy and computational time. The hybrid logistic model tree (hybrid LMT) approach proposed in this study combines the symmetrical uncertainties (SU) algorithm and the logistic model tree (LMT) algorithm. The training dataset is built by applying all possible contingencies during normal and fake tripping scenarios to the test system models. The effectiveness of the proposed approach is demonstrated on modified IEEE 9-, 14-, and 30-bus test system models due to the limitations in the simulator program. The results indicate that the hybrid LMT accurately assesses the dynamic security status of the system under normal and fake tripping contingencies with short time frame. The results show that the proposed method has 98.4126%, 98.3606%, and 99.537% accuracy and requires 22.22%, 23.529 % and 25.27% less computational time as compared to the conventional LMT algorithm in assessing the dynamic security status of the IEEE 3-machıne 9-bus, the IEEE 5-machıne 14-bus, and the IEEE 6-machıne 30-bus test system models, respectively. In summary, the results obtained in this study offer accurate and high-speed information for the dynamic
security state, which makes DSA classifier able to provide vital information for protection and control applications to keep the power system in a secure and reliable state. |
|---|