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

Pattern recognition methodology for ...

Wang, Min.

FindBook

Google Book

Amazon

博客來

Pattern recognition methodology for network-based diagnostics of power quality problems.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Pattern recognition methodology for network-based diagnostics of power quality problems./
作者:	Wang, Min.
面頁冊數:	141 p.
附註:	Source: Dissertation Abstracts International, Volume: 65-04, Section: B, page: 2025.
Contained By:	Dissertation Abstracts International65-04B.
標題:	Engineering, Electronics and Electrical. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3131244
ISBN:	0496784889

Pattern recognition methodology for network-based diagnostics of power quality problems.
Wang, Min.

Pattern recognition methodology for network-based diagnostics of power quality problems. - 141 p.

Source: Dissertation Abstracts International, Volume: 65-04, Section: B, page: 2025.

Thesis (Ph.D.)--University of Washington, 2004.

Advanced systems for automatic classification of power quality (PQ) disturbances are highly desired for both utilities and commercial customers. This dissertation presents a DSP-based system for classifying voltage and current waveform events that are related to a variety of PQ problems. The feature extraction process of disturbance waveforms is to project a PQ signal onto a low-dimension time-frequency representation (TFR), which is deliberately designed for maximizing the separability between classes. The technique of designing an optimized TFR from time-frequency ambiguity plane is for the first time applied to power engineering applications. A distinct TFR is designed for separating each individual class. The classifiers include a Heaviside-function linear classifier and neural networks with feedforward structures. The flexibility of this method allows classification of a very broad range of power quality events.

ISBN: 0496784889Subjects--Topical Terms:

626636
Engineering, Electronics and Electrical.

Pattern recognition methodology for network-based diagnostics of power quality problems.
LDR:03532nmm 2200313 4500 001 1819143
005 20061004161511.5
008 130610s2004 eng d
020 $a 0496784889
035 $a (UnM)AAI3131244
035 $a AAI3131244
040 $a UnM $c UnM
100 1 $a Wang, Min. $3 1254867
245 1 0 $a Pattern recognition methodology for network-based diagnostics of power quality problems.
300 $a 141 p.
500 $a Source: Dissertation Abstracts International, Volume: 65-04, Section: B, page: 2025.
500 $a Chair: Alexander V. Mamishev.
502 $a Thesis (Ph.D.)--University of Washington, 2004.
520 $a Advanced systems for automatic classification of power quality (PQ) disturbances are highly desired for both utilities and commercial customers. This dissertation presents a DSP-based system for classifying voltage and current waveform events that are related to a variety of PQ problems. The feature extraction process of disturbance waveforms is to project a PQ signal onto a low-dimension time-frequency representation (TFR), which is deliberately designed for maximizing the separability between classes. The technique of designing an optimized TFR from time-frequency ambiguity plane is for the first time applied to power engineering applications. A distinct TFR is designed for separating each individual class. The classifiers include a Heaviside-function linear classifier and neural networks with feedforward structures. The flexibility of this method allows classification of a very broad range of power quality events.
520 $a The algorithm was implemented on a DSP-based hardware system and tested with 860 real world waveform events that cover five classes, achieving an average recognition rate of 98%. A TI DSP processor TMS320VC5416 and ADC daughter card THS1206EVM are used. Significant optimization efforts have been under taken using C and assembly code to achieve real-time processing capability.
520 $a Detection of incipient fault activities in power systems will enable better condition-based maintenance and prevent catastrophic failures of power infrastructure. The dissertation presents a new detection algorithm based on wavelet decomposition and Wiener filter in wavelet-domain. Evaluation results based on the simulated tree contacts data from EPRI-PEAC show the promise of developing an intelligent system which monitors the heath status of system components and detects the causes of degradation.
520 $a Based on the highly distributed characteristics of modern power systems, the dissertation proposes a new monitoring concept---ubiquitous and collaborative monitoring. The goal is to design a ubiquitous PQ monitoring framework, which requires a large number of small, low-cost, and easy-to-deploy PQ sensors at numerous locations and fuses distributed sensor information to provide highly accurate monitoring results. The architecture design of ubiquitous PQ sensor is presented. New usage models of individual and networked PQ sensors are also proposed.
520 $a The dissertation also presents the development of a standard benchmark dataset for PQ event waveform classification, as well as a web-based diagnostic system for online monitoring service and real data exchange.
590 $a School code: 0250.
650 4 $a Engineering, Electronics and Electrical. $3 626636
690 $a 0544
710 2 0 $a University of Washington. $3 545923
773 0 $t Dissertation Abstracts International $g 65-04B.
790 1 0 $a Mamishev, Alexander V., $e advisor
790 $a 0250
791 $a Ph.D.
792 $a 2004
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3131244