Estimation of fault distance location using artificial neural network
Electricity demand in Malaysia is significantly increasing. Expanding the grid system to cater the new demand leads to several additional problems on the fault detection and protection coordination system. Single line to ground fault is commonly happen in grid system with possibility of 65% to 70% o...
Saved in:
| 主要作者: | |
|---|---|
| 格式: | Thesis |
| 語言: | English |
| 出版: |
2018
|
| 主題: | |
| 在線閱讀: | http://eprints.utm.my/id/eprint/79375/1/MohamadFadliMFKE2018.pdf |
| 標簽: |
添加標簽
沒有標簽, 成為第一個標記此記錄!
|
| id |
my-utm-ep.79375 |
|---|---|
| record_format |
uketd_dc |
| spelling |
my-utm-ep.793752018-10-14T08:45:03Z Estimation of fault distance location using artificial neural network 2018 Saramuji, Mohamad Fadli Adlizam TK Electrical engineering. Electronics Nuclear engineering Electricity demand in Malaysia is significantly increasing. Expanding the grid system to cater the new demand leads to several additional problems on the fault detection and protection coordination system. Single line to ground fault is commonly happen in grid system with possibility of 65% to 70% occurrence in distribution system. Fault detection is currently identified using costly special software. Typically, detection on single line to ground fault is analyzed by the pattern, placement, with and without bus monitoring approaches. Hence, this study is focused to improve the current fault detection approach on a single line to ground fault. The objective of this study is to develop a system that could estimate the faults location using artificial neural network (ANN) by Levenberg Marquardt Backpropagation (LMB) training approach. This ANN approach will be adapted in Matlab Software. A-10 bus will be developed, and faults will be simulated using Power World Software. During the implementation of the ANN, several buses will be added to enhance the capability of the neural network to detect fault distance in the system. To verify the effectiveness of the proposed ANN on the estimation fault distance determination, 21-bus distribution system has been compared for the validation purpose, with consideration of different location of the generation sources. Capability of the proposed approach has been assessed using a curving fitting tool in Matlab in terms of means square error (MSE) and regression plot (R). From the findings, it shows that LMB method can be implemented for location-based fault detection estimation once it was trained with 150 ANN hidden layer. Under the best condition, the deviation between regression of transmission line for 10-Bus single line and 21-Bus quad generation system has been achieved at 0.057%. 2018 Thesis http://eprints.utm.my/id/eprint/79375/ http://eprints.utm.my/id/eprint/79375/1/MohamadFadliMFKE2018.pdf application/pdf en public masters Universiti Teknologi Malaysia, Faculty of Electrical Engineering Faculty of Electrical Engineering |
| institution |
Universiti Teknologi Malaysia |
| collection |
UTM Institutional Repository |
| language |
English |
| topic |
TK Electrical engineering Electronics Nuclear engineering |
| spellingShingle |
TK Electrical engineering Electronics Nuclear engineering Saramuji, Mohamad Fadli Adlizam Estimation of fault distance location using artificial neural network |
| description |
Electricity demand in Malaysia is significantly increasing. Expanding the grid system to cater the new demand leads to several additional problems on the fault detection and protection coordination system. Single line to ground fault is commonly happen in grid system with possibility of 65% to 70% occurrence in distribution system. Fault detection is currently identified using costly special software. Typically, detection on single line to ground fault is analyzed by the pattern, placement, with and without bus monitoring approaches. Hence, this study is focused to improve the current fault detection approach on a single line to ground fault. The objective of this study is to develop a system that could estimate the faults location using artificial neural network (ANN) by Levenberg Marquardt Backpropagation (LMB) training approach. This ANN approach will be adapted in Matlab Software. A-10 bus will be developed, and faults will be simulated using Power World Software. During the implementation of the ANN, several buses will be added to enhance the capability of the neural network to detect fault distance in the system. To verify the effectiveness of the proposed ANN on the estimation fault distance determination, 21-bus distribution system has been compared for the validation purpose, with consideration of different location of the generation sources. Capability of the proposed approach has been assessed using a curving fitting tool in Matlab in terms of means square error (MSE) and regression plot (R). From the findings, it shows that LMB method can be implemented for location-based fault detection estimation once it was trained with 150 ANN hidden layer. Under the best condition, the deviation between regression of transmission line for 10-Bus single line and 21-Bus quad generation system has been achieved at 0.057%. |
| format |
Thesis |
| qualification_level |
Master's degree |
| author |
Saramuji, Mohamad Fadli Adlizam |
| author_facet |
Saramuji, Mohamad Fadli Adlizam |
| author_sort |
Saramuji, Mohamad Fadli Adlizam |
| title |
Estimation of fault distance location using artificial neural network |
| title_short |
Estimation of fault distance location using artificial neural network |
| title_full |
Estimation of fault distance location using artificial neural network |
| title_fullStr |
Estimation of fault distance location using artificial neural network |
| title_full_unstemmed |
Estimation of fault distance location using artificial neural network |
| title_sort |
estimation of fault distance location using artificial neural network |
| granting_institution |
Universiti Teknologi Malaysia, Faculty of Electrical Engineering |
| granting_department |
Faculty of Electrical Engineering |
| publishDate |
2018 |
| url |
http://eprints.utm.my/id/eprint/79375/1/MohamadFadliMFKE2018.pdf |
| _version_ |
1747818213205344256 |