Classification of power quality disturbances using wavelet and probabilistic neural network / Mohd Nur Aizat Mohd Ali
Power quality is a term used to describe electric power that motivates an electrical load and the load's ability to function properly with that electric power. The effects of the lack of power quality could suffer major loss especially in the business and industries. Appropriate mitigation proc...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2007
|
| Online Access: | https://ir.uitm.edu.my/id/eprint/84730/1/84730.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Power quality is a term used to describe electric power that motivates an electrical load and the load's ability to function properly with that electric power. The effects of the lack of power quality could suffer major loss especially in the business and industries. Appropriate mitigation procedures need to be taken in order to improve the power quality. This project presents the classification of power quality disturbances in electrical power systems. Power quality is monitored and the disturbances waveforms are recorded in order to identify causes and sources of disturbances. The techniques for recognizing and identifying power quality disturbance waveforms are primarily based on visual inspection of the waveform. The application of Wavelet Transform analysis incorporated with Probabilistic Neural Network (PNN) is used to classify the disturbances events. The Wavelet Transform is applied first to the data of power quality disturbances. Then the disturbances data are analyzed using Multi-resolution decomposition. The Wavelet Transform is used to decompose the disturbances signal into smooth and detailed version which consists of the disturbance waveforms. Magnitude from the detail data is considered to do the training samples and then applied to Probabilistic Neural Network (PNN) to recognize and classify the events. The combination methods have successfully recognized he data disturbances and produce the result with accuracy of 95.6%. Although the classification is not perfectly achieved, it is proved that both combination of Wavelet Transform and Probabilistic Neural Network can be used as assistance for classified and improvement of power quality disturbances. |
|---|