基于改进ABC算法优化的LSSVM多分类器组机械故障诊断模型

中国机械工程 ›› 2013, Vol. 24 ›› Issue (16): 2157-2164.

基于改进ABC算法优化的LSSVM多分类器组机械故障诊断模型

李鑫滨;陈云强;张淑清

燕山大学工业计算机控制工程河北省重点实验室,秦皇岛,066004

出版日期:2013-08-25 发布日期:2013-08-23
基金资助:
国家自然科学基金资助项目(61172095,51075349);河北省自然科学基金资助项目(F2012203138)
National Natural Science Foundation of China（No. 61172095,51075349）;
Hebei Provincial Natural Science Foundation of China（No. F2012203138）

Mechanical Fault Diagnosis Model Based on IABC Algorithm Optimized Multiple LSSVM Classifier Group

Li Xinbin;Chen Yunqiang;Zhang Shuqing

Yanshan University, Key Lab of Industrial Computer Control Engineering of Hebei Province,Qinhuangdao,Hebei,066004

Online:2013-08-25 Published:2013-08-23
Supported by:

National Natural Science Foundation of China（No. 61172095,51075349）;
Hebei Provincial Natural Science Foundation of China（No. F2012203138）

摘要/Abstract

摘要：

为了提高复杂机械故障诊断的确诊率,提出了一种基于改进人工蜂群算法(improved artificial bee colony,IABC)优化LSSVM多分类器组的故障诊断模型。该模型利用多特征提取方法,获取了较为完备的时频域特征信息,同时选择具有较强搜索能力和快速收敛性的IABC算法优化了LSSVM分类器的参数,提高了分类效率,在诊断决策层,利用评估矩阵进行了多分类器诊断结果的融合决策。通过与传统方法的对比表明:该诊断模型不仅能获取完备的故障特征信息,而且能更快地获取LSSVM最优分类参数；同时,基于评估矩阵的融合决策能够充分考虑各子分类器的性能差异,保证了诊断决策的高效精确。多种数据仿真表明,该诊断模型适用于机械故障诊断。


关键词: 时频域特征, 改进人工蜂群算法, LSSVM多分类器组, 评估矩阵

Abstract:

In order to improve the fault diagnosis accuracy,the paper presented a fault diagnosis model based on IABC optimized multiple LSSVM classifier group.It used multiple feature extraction methods to get the complete time-frequency domain features.Then IABC was utilized to optimize LSSVM parameters,because it had strong searching ability and fast convergence to improve classification efficiency.And in the diagnostic decisions stage,the criterion matrix was utilized to make the decision of the multiple classifiers.Through the comparison with the traditional methods,the results show that it can obtain complete features information,and can get LSSVM parameters more quickly and effectively.The fusion decision based on criterion matrix fully considering the differences of sub-classifiers performance,and it ensures the higher accuracy of diagnostic decisions.Finally,the various simulation results show that the algorithm can be used in mechanical fault diagnosis well. 

Key words: time-frequency domain feature, improved artificial bee colony(IABC), multiple LSSVM classifier group, criterion matrix

中图分类号:

TP274
TH133

李鑫滨, 陈云强, 张淑清. 基于改进ABC算法优化的LSSVM多分类器组机械故障诊断模型[J]. 中国机械工程, 2013, 24(16): 2157-2164.

LI Xin-Bin, CHEN Yun-Jiang, ZHANG Chu-Qing. Mechanical Fault Diagnosis Model Based on IABC Algorithm Optimized Multiple LSSVM Classifier Group[J]. China Mechanical Engineering, 2013, 24(16): 2157-2164.

参考文献

[1]Yang Junyan, Zhang Youyun, Zhu Yongsheng. Intel- ligent Fault Diagnosis of Rolling Element Bearing Based on Svms and Fractal Dimension[J]. Mechani- cal Systems and Signal Processing, 2007, 21 (5): 2012-2024.
[2]Wu Qiang, Kong Fanrang, He Qingbo, et al. EarlyFault Diagnosis of Rolling Element Bearings Based on Wavelet Transform and Independent Component Analysis[J]. China Mechanical Engineering, 2012, 23(7) :835-840.
[3]Shen Zhongjie, Chen Xuefeng, Zhang Xiaoli, et al. A Novel Intelligent Gear Fault Diagnosis Model Based on EMD and Multi- class TSVM [J]. Measure- ment,2012,45(1) :30-40.
[4]Hajnayeb A, Ghasemloonia A, Khadem S E, et al. Application and Comparison of an ANN- based Feature Selection Method and the Genetic Algo- rithm in Gearbox Fault Diagnosis[J]. Expert Sys- tems with Applications,2011,38(8) : 10205-10209.
[5]Li Bing,Zhang Peilin,Tian Hao,et al. A New Fea- ture Extraction and Selection Scheme for Hybrid Fault Diagnosis of Gearbox [J]. Expert Systems with Applications, 2011,38(8) : 10000-10009.
[6]Smart E, Brown D, Denman J. Combining Multiple Classifiers to Quantitatively Rank the Impact of Ab- normalities in Flight Data[J]. Applied Soft Compu- ting, 2012,12 (8) : 2583-2592.
[7]Haghighi M S,Vahedian A,Yazdi H S. Creating and Measuring Diversity in Multiple Classifier Systems Using Support Vector Data Description[J]. Applied Soft Computing, 2011,11 (8) : 4931-4942.
[8]Zhang Xian, Wang Hongli. LSSVM Based on PSO and Its Applications to Time Series Prediction[J]. China Mechanical Engineering, 2011,22 (21) : 2572- 2576.
[9]Luo Zhigao,Chen Baolei, Pang Chaoli, et al. Rolling Bearing Complex Fault Diagnosis Based on Genetic Algorithm[J]. Journal of Vibration and Shock, 2012,29(6) ： 174-177.
[10]Kadri O,Mouss L H, Mouss M D. Fault Diagnosis of Rotary Kiln Using SVM and Binary ACO[J]. Journal of Mechanical Science and Technology, 2012,26(2) ;601-608.
[11]Liu Zhiwen, Cao Hongrui, Chen Xuefeng, et al. Multi-fault Classification Based on Wavelet SVM with PSO Algorithm to Analyze Vibration Signals from Rolling Element Bearings[J]. Neuroeomput- ing, 2013,99 : 399-410.
[12]Sulaiman M H, Mustafa M W, Shareef H,et al. An Application of Artificial Bee Colony Algorithm with Least Squares Support Vector Machine for.Real and Reactive Power Tracing in Deregulated Power System[J]. International Journal of Electri- cal Power ： Energy Systems,2012,37(1)：67-77.
[13]Aali：ir D, Dogantekin E. An Automatic Diabetes Diagnosis System Based on LDA-Wavelet Sup- port Vector Machine Classifier[J]. Expert Sys- tems with Applications,2011,38(7)：8311：8315.
[14]Wang Xiang, Wang Zhongsheng. Incipient Rub- impact Fault Diagnosis of Rotor System Using Im- proved HHT[J]. China Mechanical Engineering, 2011,22(24) :2937-2940.
[15]Karaboga D,Basturk 13. A Comparative Study of Artifi- cial Bee Colony Algorithm[J]. Applied Mathematics and Computation,2009,214(1) :108-132.
[16]Huang Jian, Hu Xiaoguang, Yang Fan. SupportVector Machine with Genetic Algorithm for Ma- chinery Fault Diagnosis of High Voltage Circuit Breaker [ J ]. Measurement, 2011, 44 (6): 1018- 1027.
[17]Karaboga D, Ozturk C. A Novel Clustering Ap- proach: Artificial Bee Colony (ABC) Algorithm [J]. Applied Soft Computing, 2011,11 (1) : 652- 657.
[18]Xie Chunli, Shao Cheng, Zhao Dandan. Parameters Optimization of Least Squares Support Vector Ma- chines and Its Application[J]. Journal of Computers, e011,6(9),1935-1941.