基于IAFSA-SVM的岸電箱斷路器故障診斷
楊奕飛1,焦文文1,何祖軍1,張發(fā)平2,郭江2
(1 江蘇科技大學(xué) 電子信息學(xué)院,江蘇 鎮(zhèn)江 212003;2 江蘇中智海洋工程裝備有限公司,江蘇 鎮(zhèn)江 212000)
摘 要:斷路器的故障診斷對岸電系統(tǒng)的穩(wěn)定運(yùn)行有重要意義。針對人工魚群算法和其他智能算法在優(yōu)化支持向量機(jī)參數(shù)時,存在易陷入局部最優(yōu)、泛化能力差等問題,通過自適應(yīng)調(diào)整步長和引入全局隨機(jī)行為,提出基于改進(jìn)人工魚群算法優(yōu)化支持向量機(jī)參數(shù)的故障診斷模型。將斷路器合閘線圈電流信號中的時間和電流信號作為特征量,采用改進(jìn)人工魚群算法對支持向量機(jī)的參數(shù)尋優(yōu),以提升支持向量機(jī)的故障分類性能。仿真結(jié)果顯示,該算法在樣本數(shù)量小的情況下仍具有良好的分類性能,能夠準(zhǔn)確對斷路器進(jìn)行故障分類。
關(guān)鍵詞:支持向量機(jī);改進(jìn)人工魚群算法;岸電箱;斷路器故障診斷
中圖分類號:TM561 文獻(xiàn)標(biāo)識碼:A 文章編號:1007-3175(2019)08-0057-05
Circuit Breaker Fault Diagnosis of Shore Connection Box Based on Improved
Artificial Fish Swarm Algorithm and Support Vector Machine
YANG Yi-fei1, JIAO Wen-wen1, HE Zu-jun1, ZHANG Fa-ping2, GUO Jiang2
(1 School of Electronics and Information, Jiangsu University of Science and Technology, Zhenjiang 212003, China;
2 Jiangsu Zhongzhi Marine Engineering Equipment Co., Ltd, Zhenjiang 212000, China)
Abstract: The fault diagnosis of the circuit breaker is of great significance to the stable operation of the shore power system. For the artificial fish swarm algorithm and other intelligent algorithms, when optimizing the parameters of support vector machine, there were problems such as easy to fall into local optimum and poor generalization ability. By adaptively adjusting the step size and introducing global random behavior, this paper proposed an improved artificial fish swarm algorithm to optimize the fault diagnosis model of the support vector machine parameters. To improve the fault classification performance of the support vector machine, the time and current signals extracted from the current signals of the circuit breaker closing coil were used as the characteristic variables, and the improved artificial fish swarm algorithm was adopted to optimize the parameters of the support vector machine. The simulation results show that this algorithm can accurately judge the fault type of the circuit breaker with good classification performance under the conditions of small quantity of samples.
Key words: support vector machine; improved artificial fish swarm algorithm; shore power box; circuit breaker fault diagnosis
參考文獻(xiàn)
[1] 張蓮,王磊,禹紅良,等. 改進(jìn)量子神經(jīng)網(wǎng)絡(luò)高壓斷路器故障診斷方法研究[J]. 重慶理工大學(xué)學(xué)報( 自然科學(xué)),2018,32(9):157-162.
[2] 胡曉光,孫來軍,紀(jì)延超. 基于線圈電流和觸點(diǎn)狀態(tài)的斷路器故障分析[J]. 電力自動化設(shè)備,2006,26(8):5-7.
[3] 趙書濤,王亞瀟,孫會偉,等. 基于自適應(yīng)權(quán)重證據(jù)理論的斷路器故障診斷方法研究[J]. 中國電機(jī)工程學(xué)報,2017,37(23):7040-7046.
[4] 黃浩. 基于相空間重構(gòu)理論的滾動軸承故障診斷研究[D]. 武漢:武漢科技大學(xué),2014.
[5] 李志杰,王健. 基于多分類SVM的車牌字符識別算法研究[J]. 物流工程與管理,2016,38(5):260-263.
[6] 李君,梁昔明. 人工魚群算法收斂速度改進(jìn)優(yōu)化仿真[J]. 計算機(jī)仿真,2018,35(1):232-238.
[7] 張艷. 基于粒子群優(yōu)化支持向量機(jī)的變壓器故障診斷和預(yù)測[D]. 成都:西華大學(xué),2011.
[8] 費(fèi)騰,張立毅,陳雷. 混合Levy變異與混沌變異的改進(jìn)人工魚群算法[J]. 計算機(jī)工程,2016,42(7):146-152.
[9] 袁金麗,李奎,郭志濤,等. 基于SVM與合分閘線圈電流參數(shù)的高壓斷路器機(jī)械故障診斷[J]. 高壓電器,2011,47(3):26-30.
[10] 齊巖磊,陳娟,楊祺,等. 基于SVM的葛根素提取軟測量系統(tǒng)的設(shè)計[J]. 電子測量與儀器學(xué)報,2012,26(8):726-731.