【摘要】：目前,電力系統(tǒng)正在成為綜合國力的重要標志。但是,由于電力系統(tǒng)越來越復(fù)雜和運行人員操作失誤、人為破壞和自然不可抗拒力的因素,大規(guī)模停電事故時有發(fā)生。當電力系統(tǒng)發(fā)生故障后,如何迅速的、精確的確定故障點,以便快速地修復(fù)故障、恢復(fù)供電是繼電保護故障診斷研究課題的重點。 電力系統(tǒng)的故障診斷關(guān)鍵問題在于如何快速、準確的識別故障元件。鑒于現(xiàn)代電力系統(tǒng)的龐大、故障信息復(fù)雜和多樣。本文在國內(nèi)外相關(guān)研究成果的基礎(chǔ)上,依據(jù)電力系統(tǒng)中繼電保護開關(guān)的變位和繼電保護動作的信息,提出了基于支持向量機的繼電保護故障診斷模型。根據(jù)支持向量機算法原理,通過對多種參數(shù)優(yōu)化算法的學習與研究,提出了基于粒子群優(yōu)化算法的支持向量機故障診斷模型。在詳細總結(jié)了母線、輸電線路、變壓器的繼電保護動作原理后,應(yīng)用繼電保護故障診斷模型對數(shù)據(jù)資源進行挖掘分析,運用當今主流框架,設(shè)計并實現(xiàn)了繼電保護故障診斷系統(tǒng)。該系統(tǒng)可以準確、快速的對電網(wǎng)故障進行診斷。系統(tǒng)展示內(nèi)容豐富,分析結(jié)果直觀易懂。 該系統(tǒng)可以是繼電保護裝置更加準確、合理的診斷電網(wǎng)故障,能更好的反應(yīng)電網(wǎng)運行的情況,從而為電網(wǎng)運行人員的決策提供支持。通過對以往電網(wǎng)故障的實驗表明,該系統(tǒng)具有良好的實用性和高效性。
[Abstract]:At present, power system is becoming an important symbol of comprehensive national strength. However, large-scale power outages often occur due to the increasing complexity of the power system and the operation errors of operators, man-made damage and natural irresistible force. When the power system fails, how to quickly and accurately determine the fault point in order to quickly repair the fault and restore the power supply is the focus of relay protection fault diagnosis research. The key problem of power system fault diagnosis is how to identify fault components quickly and accurately. In view of the large scale of modern power system, the fault information is complex and diverse. On the basis of relevant research results at home and abroad, this paper presents a fault diagnosis model of relay protection based on support vector machine (SVM), according to the information of the shift of relay protection switch and the action of relay protection in power system. According to the principle of support vector machine (SVM) algorithm, a support vector machine (SVM) fault diagnosis model based on particle swarm optimization (PSO) is proposed through the study and research of multi-parameter optimization algorithms. After summarizing the principle of relay protection operation of busbar, transmission line and transformer in detail, the fault diagnosis model of relay protection is used to mine and analyze the data resources, and the mainstream frame is used today. The fault diagnosis system of relay protection is designed and implemented. The system can diagnose the fault of power network accurately and quickly. The system display content is rich, the analysis result is intuitive and easy to understand. The system can be a more accurate and reasonable fault diagnosis system for relay protection devices, and can better reflect the operation of the power grid, thus providing support for the decision-making of the power network operators. The experiments on the previous power network faults show that the system has good practicability and high efficiency.
【學位授予單位】：華北電力大學
【學位級別】：碩士
【學位授予年份】：2014
【分類號】：TM77

【參考文獻】

相關(guān)期刊論文 前10條

1 強迎春;魏廣華;劉峗;呂春蓮;;設(shè)備投資項目綜合效益評價模型及應(yīng)用[J];長沙航空職業(yè)技術(shù)學院學報;2010年02期

2 李盼池,肖紅,許少華,劉顯德;支持向量機在模式分類中的應(yīng)用[J];大慶石油學院學報;2003年02期

3 連可;黃建國;王厚軍;龍兵;;一種基于遺傳算法的SVM決策樹多分類策略研究[J];電子學報;2008年08期

4 何敏;張志利;劉輝;趙鍇;張永鑫;;故障診斷技術(shù)方法綜述[J];國外電子測量技術(shù);2006年05期

5 顧雪平，，張文勤，高曙，盛四清，楊以涵;基于人工神經(jīng)網(wǎng)絡(luò)和專家系統(tǒng)結(jié)合運用的電網(wǎng)故障診斷技術(shù)的研究[J];河北省科學院學報;1995年Z1期

6 陳玉東,施頌椒,翁正新;動態(tài)系統(tǒng)的故障診斷方法綜述[J];化工自動化及儀表;2001年03期

7 林圣;何正友;錢清泉;;輸電網(wǎng)故障診斷方法綜述與發(fā)展趨勢[J];電力系統(tǒng)保護與控制;2010年04期

8 倪麗萍;倪志偉;李鋒剛;潘永剛;;基于蟻群算法的SVM模型選擇研究[J];計算機技術(shù)與發(fā)展;2007年09期

9 李敏君;楊昕卉;王榮芝;王淑玉;;基于遺傳算法的故障診斷研究[J];微計算機信息;2006年16期

10 徐玉秀,邢剛,原培新;基于專家系統(tǒng)與神經(jīng)網(wǎng)絡(luò)集成的故障診斷的應(yīng)用研究[J];振動與沖擊;2001年01期

本文編號：2435351