東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

Classification of transients in powe...

Chen, Jin.

FindBook

Google Book

Amazon

博客來

Classification of transients in power systems.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Classification of transients in power systems./
作者:	Chen, Jin.
面頁冊數:	201 p.
附註:	Source: Masters Abstracts International, Volume: 41-05, page: 1485.
Contained By:	Masters Abstracts International41-05.
標題:	Engineering, Electronics and Electrical. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=MQ76913
ISBN:	0612769135

Classification of transients in power systems.
Chen, Jin.

Classification of transients in power systems. - 201 p.

Source: Masters Abstracts International, Volume: 41-05, page: 1485.

Thesis (M.Sc.)--University of Manitoba (Canada), 2002.

This thesis presents a method of feature extraction and classification of power system transients. The system developed in this thesis uses both the wavelet transform and multifractal analysis as tools to analyze power system transients for feature extraction, as well as probabilistic neural network that is used as a classifier for identifying various types of power transients associated with power system faults and switching.

ISBN: 0612769135Subjects--Topical Terms:

626636
Engineering, Electronics and Electrical.

Classification of transients in power systems.
LDR:02604nmm 2200289 4500 001 1819279
005 20061004161556.5
008 130610s2002 eng d
020 $a 0612769135
035 $a (UnM)AAIMQ76913
035 $a AAIMQ76913
040 $a UnM $c UnM
100 1 $a Chen, Jin. $3 1908568
245 1 0 $a Classification of transients in power systems.
300 $a 201 p.
500 $a Source: Masters Abstracts International, Volume: 41-05, page: 1485.
500 $a Adviser: W. Kinsner.
502 $a Thesis (M.Sc.)--University of Manitoba (Canada), 2002.
520 $a This thesis presents a method of feature extraction and classification of power system transients. The system developed in this thesis uses both the wavelet transform and multifractal analysis as tools to analyze power system transients for feature extraction, as well as probabilistic neural network that is used as a classifier for identifying various types of power transients associated with power system faults and switching.
520 $a This system has three major modules: (i) transients simulation, (ii) feature extraction of the transients, and (iii) features classification. A wavelet transform is used to generate multiresolution time-frequency features and multifractal analysis is conducted to generate a variance fractal dimension trajectory directly used as features of the power transient. The combination of these features characterizes the power transient in good and compact representations for neural network analysis. Finally, the modelled transient is classified using the multi-sigma PNN classifier that uses a separate sigma weight for each input variable.
520 $a Performance of the proposed classification system is evaluated based on simulated transients. The experimental results show that this identification system is precise and fast in discriminating the type of a transient event. More specifically, the system was trained with 250 out of 750 simulated pure transients belonging to five classes of disturbances. Testing the system with the remaining 500 transients yielded a correct classification rate around 99%. The average processing time for completing feature extraction and classification was about 2 second per transient with Pentium (R) III of 600 MHz. Experiments also show that the classification system is robust in a noisy environment.
590 $a School code: 0303.
650 4 $a Engineering, Electronics and Electrical. $3 626636
690 $a 0544
710 2 0 $a University of Manitoba (Canada). $3 1018707
773 0 $t Masters Abstracts International $g 41-05.
790 1 0 $a Kinsner, W., $e advisor
790 $a 0303
791 $a M.Sc.
792 $a 2002
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=MQ76913