Abstract：Aiming at the problems of small samples and dataset imbalance in fault diagnosis, an optimized cost-sensitive support vector machine (CS-SVM) model was proposed and an improved fly optimization algorithm (FOA) was applied to select the best regularized constants C_+,C_-, and the kernel function parameter g, by increasing the penalties for the misclassification of fault dataset, so as to raise the diagnostic accuracy of the fault samples.The proposed method was verified by combining the stochastic resonance and KPCA feature extraction methods, taking the experimental data of a IMS aviation bearing as an example.The results show that the CS-SVM can effectively deal with unbalanced samples of small fault class in bearing fault diagnosis, and improve the diagnosis accuracy.It can also be further expanded to other fields of fault diagnosis.
何大伟, 彭靖波, 胡金海, 李腾辉, 贾伟州. 基于改进FOA优化的CS-SVM轴承故障诊断研究[J]. 振动与冲击, 2018, 37(18): 108-114.
HE Dawei, PENG Jingbo, HU Jinhai, LI Tenghui JIA Weizhou. Bearing fault diagnosis based on a modified CS-SVM model optimized by an improved FOA algorithm. JOURNAL OF VIBRATION AND SHOCK, 2018, 37(18): 108-114.
[1] 屈梁生. 机械故障的全息诊断原理[M]. 北京：科学出版社，2007
[2] E L Ding, H Fennel, S X Ding. Model-based diagnosis of sensor faults for ESP systems[J]. Control Engineering Practice, 2004, 12(7)：847- 856.
[3] SIMON D L，LITT J S. A data filter for identifying steady-state operating points in engine flight data for condition monitoring applications[J]. Journal of Engineering for Gas Turbines and power. 2011，133( 7) ：1- 8.
[4] 段礼祥，谢骏遥，王凯等. 基于不同工况下辅助数据集的齿轮箱故障诊断[J].振动与冲击，2017，36(10)：104-108.
DUAN Li-xiang，XIE Jun-yao，WANG Kai,et al. Gearbox diagnosis based on auxiliary monitoring datasets of different working conditions[J]. Journal of vibration and shock, 2017, 36(10)：104-108.
[5] 段礼祥，郭晗，王金江. 数据集不均衡下的设备故障程度识别方法研究[J].振动与冲击，2016，35(20)：178-182.
DUAN Li-xiang，Guo Han，WANG Jin-jiang. A mechanical fault severity identification method under unbalanced datasets[J]. Journal of vibration and shock. 2016，35(20)：178-182.
[6] Han J，Kamber M. Data Mining: Concepts and Techniques[M]. San Francisco CA: Morgan Kaufmann，2001.
[7] Gangsar P，Tiwari R. Taxonomy of Induction-Motor Mechanical-Fault Based on Time-Domain Vibration Signals by Multiclass SVM Classifiers[J]. Intelligent Industrial Systems，2016，2(3):269-281.
[8] Ridha Ziani，Ahmed Felkaoui，Rabah Zegadi. Bearing fault diagnosis using multiclass support vector machines with binary particle swarm optimization and regularized Fisher’s criterion[J]. Journal of Intelligent Manufacturing，2017，28(2):405-417.
[9] 陶新民， 刘福荣， 童智靖， 等．不均衡数据下基于 SVM 的故障检测新算法[J]．振动与冲击，2010，29(12) : 8－12．
TAO Xin-min，LIU Fu-rong，TONG Zhi-jing， et al．A new fault detection method of unbalanced data based on SVM[J]．Journal of Vibration and Shock，2010，29(12):8－12．
[10] WANG Xi-ping，TAN Wen-xue，WU Hua-rui．An innovative SVM for wheat seed quality estimation[J]．Journal of Information & Computational Science，2015，12 (1) : 223 －233．
[11] 万建武，杨明，陈银娟.代价敏感的半监督Laplacian支持向量机[J].电子学报，2012，40(7):1410-1415.
WAN Jian-wu，YANG Ming，CHEN Yin-juan. Cost Sensitive Semi-Supervised Laplacian Support Vector Machine[J]. ACTA ELECTRONICA SINICA，2012，40(7)：1410-1415.
[12] 唐明珠，阳春华，桂卫华.基于改进QBC和CS-SVM的故障检测[J].控制与决策，2012,27（10）：1489-1493.
TANG Ming-zhu，YANG Chun-hua，GUI Wei-hua. Fault detection based on modified QBC and CS-SVM[J]. Control and Decision，2012,27（10）：1489-1493.
[13] 吴松林.基于改进代价敏感支持向量机的铜闪速熔炼工况评估模型研究及应用[D].中南大学，长沙，2012.
[14] Vapnik V N，张学工[译]. 统计学习理论的本质[M]. 北京: 清华大学出版社，2000．
[15] 周波. 基于集成学习的不平衡数据分类的研究及应用[D]. 大连:大连理工大学, 2014.
[16] 潘文超. 果蝇最佳化演算法[M]. 台北：沧海书局，2011.
PAN W C. Fruit Fly Optimization Algorithm[M]. Taipei: Tsang Hai Publishing, 2011
[17] PAN W C. A new evolutionary computation approach: Fruit fly optimization algorithm[C] //2011 Conference of Digital Technology and Innovation Management, 2011.
[18] Pan W C．A new fruit fly optimization algorithm: taking the financial distress model as an example[J].Knowledge Based System，2012，29: 69-74．
[19] 赵伟. 基于IFOA算法的SVM参数优化及其应用[J]. 计算机系统应用. 2015，24(6)：207- 210 .
ZHAO Wei. SVM Parameters Optimizing Based on Improved Diminishing Step Fruit Fly Optimization Algorithm and Its Application[J]. Computer Systems &Application，2015，24(6)：207- 210 .
[20] 徐跃州，张欣. 基于混沌果蝇算法的WSN优化布局[J].计算机工程与设计. 2015，36(4)：900-905.
XU Yue-zhou，ZHANG Xin. Layout optimization for wireless sensor network based on chaotic fruit fly algorithm[J]. Computer Engineering and Design. 2015，36(4)：900-905.
[21] NSF I/UCR Center for Intelligent Maintenance Systems (IMS–www.imscenter.net) with support from Rexnord Corp. in Milwaukee, WI.
[22] 任立通，胡金海，谢寿生等. 基于随机共振预处理的振动故障特征提取研究[J]. 振动与冲击. 2014，33(2)：143-146.
REN Li-tong，HU Jin-hai，XIE Shou-sheng， et. al. Vibration fault feature extraction based on stochastic resonance pretreatment[J]. Journal of vibration and shock. 2014，33(2)：143-146.
[23] 刘蕴哲，胡金海，任立通等. 基于特征选择与概率神经网络的轴承故障诊断研究[J]. 机械传动. 2016，40(10)：48-53
LIU Yun-zhe，HU Jin-hai，REN Li-tong，et al. Study on the Bearing Fault Diagnosis based on Feature Selection and Probabilistic Neural Network[J].Journal of Mechanical Transmission. 2016，40(10)：48-53.
[24] 胡金海，谢寿生，侯胜利等.核函数主元分析及其在故障特征提取中的应用[J]. 振动、测试与诊断. 2007，27(1)： 48-52.
HU Jin-hai，XIE Shou-sheng，HOU Sheng-li，et al. Kernel Principal Component Analysis and Its Application to Fault Feature Extraction[J]. Journal of Vibration, Measurement& Diagnosis. 2007，27(1)：48-52.
[25] Akbani R，Kwek S，Japkowicz N. Applying support vector machines to imbalanced datasets [C]//Proceedings of the International Conference on Artificial Intelligence，2004：214-220
[26] 金鑫，李玉鑑. 平衡支持向量机的惩罚因子选择方法[J]. 计算机工程与应用. 2011,47(33)：129-133.
JIN Xin，LI Yu-jian. Error-cost selection for biased support vector machines[J]. Computer Engineering and Applications. 2011,47(33)：129-133.
