遺傳神經(jīng)算法在電力變壓器故障診斷中的應(yīng)用
李銀龍,林志雄
福建省電力有限公司泉州電業(yè)局,福建 泉州 362000
摘 要: 利用遺傳算法的全局搜索性能和BP 算法較強(qiáng)的局部搜索能力,提出一種收斂速度快的改進(jìn)遺傳神經(jīng)混合算法,并應(yīng)用于油中溶解氣體分析的電力變壓器故障診斷中,實(shí)際結(jié)果表明,該算法能對(duì)電力變壓器各種故障進(jìn)行有效分類,并具有較快的收斂速度和較高的診斷精度。
關(guān)鍵詞: 電力變壓器;遺傳算法;人工神經(jīng)網(wǎng)絡(luò);故障診斷
中圖分類號(hào):TM411 文獻(xiàn)標(biāo)識(shí)碼:A 文章編號(hào):1007-3175(2013)05-0047-03
Application of Genetic Neural Algorithm in Power Transformer Fault Diagnose
LI Yin-long, LIN Zhi-xiong
Quanzhou Power Bureau, Power Company in Fujian Province, Quanzhou 362000, China
Abstract: With the global search performance of genetic algorithm and the local search capability of back propagation (BP) algorithm, this paper raised a kind of fast convergence improved genetic mixed algorithm, which was applied in power transformer fault diagnosis to analyze gases dissolved in the oil. The actual result shows that the algorithm can classify effectively for each kind of fault in power transformer, with faster convergence speed and higher diagnosis accuracy.
Key words: power transformer; genetic algorithm; artificial neural network; fault diagnose
參考文獻(xiàn)
[1] 臧宏志,徐建政,俞曉冬. 基于多種人工智能技術(shù)集成的電力變壓器故障診斷[J]. 電網(wǎng)技術(shù),2003,27(3):15-17.
[2] 胡漢梅, 鮑亮亮, 趙海軍. 神經(jīng)網(wǎng)絡(luò)在變壓器故障診斷中典型算法研究[J]. 高壓電器,2008,44(3):217-220.
[3] 原清,賀新鋒,劉湘崇.遺傳算法和神經(jīng)網(wǎng)絡(luò)在導(dǎo)彈測(cè)試設(shè)備故障診斷中的應(yīng)用研究[J].測(cè)試技術(shù)學(xué)報(bào),2002(12):26-31.
[4] 云慶夏.進(jìn)化算法[M].北京:冶金工業(yè)出版社,2000.
[5] 王南蘭,邱德潤(rùn).油中溶解氣體分析的變壓器故障診斷新方法[J].高電壓技術(shù),2006,32(6):35-37.
[6] 張宇,郭晶,周激流.動(dòng)態(tài)變異遺傳算法[J].電子科技大學(xué)學(xué)報(bào),2002,31(3):234-239.
[7] 閻平凡,張長(zhǎng)水.人工神經(jīng)網(wǎng)絡(luò)與模擬進(jìn)化計(jì)算[M].北京:清華大學(xué)出版社,2000.
[8] 楊海馬,劉瑾,張菁.BP神經(jīng)網(wǎng)絡(luò)在變壓器故障診斷中的應(yīng)用[J].變壓器,2009,46(1):67-70.