基于粒子群优化的SVM供水管道泄漏诊断方法PSO-SVM based leakage diagnosis method of water supply pipeline
王学渊,陈志刚,钟新荣,卢宁
摘要(Abstract)：
供水管道泄漏会造成水资源浪费和经济损失,传统支持向量机泄漏诊断模型中存在参数选择不确定的问题,导致其分类结果不稳定。提出将粒子群进化算法应用于泄漏诊断支持向量机模型中的参数优化选择,利用粒子群群体智能优化搜索从全局迅速地迭代出合理的支持向量机的惩罚参数以及核参数,使建立的PSO-SVM管道泄漏诊断模型达到最优。实验测试表明,通过结合粒子群算法全局搜索收敛速度快的优点,有效地解决了支持向量机模型中两个重要参数优化选择的问题,提升了支持向量机分类的准确率和效率。
关键词(KeyWords)： 供水管道;泄漏诊断;支持向量机;粒子群算法;参数优化;PSO-SVM
基金项目(Foundation): 国家自然科学基金(51004005);; 住建部资助项目(2016-K4-081);; 北京市优秀人才培养资助项目(2013D005017000013);; 北京市属高等学校高层次人才引进与培养计划项目;; 北京市教育委员会科技计划一般项目(KM201610016017)~~
作者(Author): 王学渊,陈志刚,钟新荣,卢宁
DOI: 10.16652/j.issn.1004-373x.2018.07.036
参考文献(References)：
- [1]文玉梅,张雪园,文静,等.依据声信号频率分布和复杂度的供水管道泄漏辨识[J].仪器仪表学报,2014,35(6):1223-1229.WEN Y M,ZHANG X Y,WEN J,et al.Identification of water pipeline leakage based on acoustic signal frequency distribution and complexity[J].Chinese journal of scientific instrument,2014,35(6):1223-1229.
- [2]唐秀家.管道系统泄漏检测神经网络与模式识别方法[J].核科学与工程,1998(3):33-40.TANG X J.Fluid pipeline system leak detection based on neural network and pattern recognition[J].Chinese journal of nuclearence&engineering,1998(3):33-40.
- [3]陈志刚,张来斌,梁伟,等.复杂工况下热油管道泄漏识别与定位方法研究[J].西南石油大学学报,2008,30(6):157-160.CHEN Z G,ZHANG L B,LIANG W,et al.Leak detection and positioning for hot oil pipeline under complicated conditions[J].Journal of Southwest Petroleum University,2008,30(6):157-160.
- [4]VAPNIK V N.The nature of statistical learning theory[M].2nd ed.New York:Springer,2000.
- [5]CHEN Peng,YUAN Lifen,HE Yigang,et al.An improved SVM classifier based on double chains quantum genetic algorithm and its application in analogue circuit diagnosis[J].Neurocomputing,2015,211:202-211.
- [6]张亮,陈志刚,杨建伟,等.基于决策树与多元支持向量机的齿轮箱早期故障诊断方法[J].计算机测量与控制,2016(1):12-15.ZHANG L,CHEN Z G,YANG J W,et al.Early fault diagnosis of gearbox based on multiclass support vector machine and decision tree[J].Computer measurement&control,2016(1):12-15.
- [7]TUNGADIO D H,JORDAAN J A,SITI M W.Power system state estimation solution using modified models of PSO algorithm:comparative study[J].Measurement,2016,92:508-523.
- [8]梁伟,张来斌,王朝晖.基于数据挖掘的负压波特征参数优化方法[J].石油机械,2008,36(12):38-42.LIANG W,ZHANG L B,WANG Z H.Study of optimization method of characteristic parameters of negative pressure wave based on data mining[J].China petroleum machinery,2008,36(12):38-42.
- [9]YIN Shen,YIN Jiapeng.Tuning kernel parameters for SVM based on expected square distance ratio[J].Information sciences,2016(2):370-371.
- [10]LIU Chao,YIN Shiqun,ZHANG Miao,et al.An improved grid search algorithm for parameters optimization on SVM[J].Applied mechanics and materials,2014,644:2216-2219.