Partial discharge detection and location technique based on segmented correlation trimmed mean algorithm for power cable
Power cable may suffer from insulation degradation after a certain period of time because of environment, mechanical and electrical factors. Partial discharge (PD) at void or cavity of power cable’s insulation will lead to the power system breakdown in the near future. Nowadays, many PD location dev...
Saved in:
| Format: | Thesis |
|---|---|
| Language: | English |
| Subjects: | |
| Online Access: | http://dspace.unimap.edu.my:80/xmlui/bitstream/123456789/78039/1/Page%201-24.pdf http://dspace.unimap.edu.my:80/xmlui/bitstream/123456789/78039/2/Full%20text.pdf http://dspace.unimap.edu.my:80/xmlui/bitstream/123456789/78039/3/Chai%20Chang%20Yii.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Power cable may suffer from insulation degradation after a certain period of time because of environment, mechanical and electrical factors. Partial discharge (PD) at void or cavity of power cable’s insulation will lead to the power system breakdown in the near future. Nowadays, many PD location devices had been invented to estimate PD location on power cable. New technology has enabled PD estimation to evolve from offline PD estimation to online PD estimation. Advanced signal processing technique can be implemented in those devices in order to estimate PD location accurately. In this thesis, segmented correlation trimmed mean (SCTM) algorithm is proposed to estimate PD location on medium voltage (MV) power cable. The algorithm uses segmented correlation
technique and trimmed mean data filtering technique to enhance the accuracy of the
estimated PD location. Two experiments have been performed to test the program
execution time and accuracy against noise of the algorithm. The algorithm had been tested
in Matrix Laboratory (MATLAB) environment which consists modelled PD signals and
different levels of white Gaussian noise (WGN) and discrete spectral interference (DSI).
Discrete wavelet transform (DWT) de-nosing technique has been used for noise
suppression. The first experiment is performed by increasing the sampling number of
measured signal while recording the program execution time of the algorithm. The second
experiment is performed by increasing the level of WGN and DSI while recording the
maximum percentage error of the estimated PD location. The results from both
experiments are compared with the existing multi-end correlation (MEC) algorithm. The
results shown that the SCTM algorithm has longer time but lower maximum percentage
error of the estimated PD location than MEC algorithm. In conclusion, SCTM algorithm
is more suitable to apply in PD location estimation system for power cable due to its lower
maximum percentage error. |
|---|