A new extreme learning machine (ELM) classifier for multi-classification task was designed based on the structural features of the ELM classifier, and an improved classification algorithm based on ELM (One-Class-PCA-ELM) was purposed. The classification algorithm was realized as follows. PCA method was utilized to process the fault data for dimensionality reduction as well as removing noise and redundant information. Then the training data were allocated into different categories according to their respective class labels and the corresponding classification model was constructed for each training data category, obtaining One-Class-PCA-ELM model. An unclassified fault data was constructed into the trained One-Class-PCA-ELM model, getting its class label and making classification process completed. Experimental results show that the proposed algorithm maintains the fast training speed of ELM, and has high classification accuracy and ideal classification stability.
QIU Ri hui, LIU Kang ling, TAN Hai long, LIANG Jun. Classification algorithm based on extreme learning machine and its application in fault identification of Tennessee Eastman process. JOURNAL OF ZHEJIANG UNIVERSITY (ENGINEERING SCIENCE), 2016, 50(10): 1965-1972.
［1］ 马贺贺.基于数据驱动的复杂工业过程故障检测方法研究［D］.上海：华东理工大学,2013.
MA Hehe. Fault detection of complex ｉndustrial ｐrocesses ｂased on ｄatadriven ｍethods ［D］. Ｓｈａｎｇｈａｉ：East China University of Science and Technology, 2013.
［2］ 石向荣.面向过程监控的非线性特征提取方法研究［D］.杭州：浙江大学, 2014.
SHI Xiangrong. Study on ｎonlinear ｆeature ｅxtraction for ｐrocess ｍonitoring ［D］. Ｈａｎｇｚｈｏｕ: Zhejiang University, 2014.
［3］ HUANG G B, ZHU Q Y, SＩＥＷ C K. Extreme learning machine: a new learning scheme of feedforward neural networks ［J］. Proceeding of International Joint Conference Neural Networks, 2004(2): 985-990.
［4］ SＩＥＷ C K, HUANG G B, ZHU Q Y. Extreme learning machine: ｔheory and applications ［J］. Neurocomputing, 2006, 70: 489501.
［5］ 遇炳杰,朱永利.加权极限学习机在变压器故障诊断中的应用［J］.计算机工程与设计,2013(12): 4340-4344.
YU Bingjie, ZHU Yongli. Transformer fault diagnosis using weighted extreme learning machine ［J］. Computer Engineering and Design, 2013(12): 4340-4344.
［6］ 周江嫚,黄清秀,彭敏放,等.基于差分进化优化ELM的模拟电路故障诊断［J］.计算机工程与应用,2014,08: 221-225.
ZHOU Jiangman, HUANG Qingxiu, PENG Minfang, et al. Analog circuit fault diagnosis based on differential evolution and extreme learning machine ［J］. Computer Engineering and Applications, 2014, 50(8): 221-225.
［7］ 赵立杰,袁德成,柴天佑. 基于多分类概率极限学习机的污水处理过程操作工况识别［J］. 化工学报,2012(10): 3173-3182.
ZHAO Lijie, YUAN Decheng, CHAI Tianyou. Identification of wastewater operational conditions based on multiclassification probabilistic extreme learning machine ［J］. CIESC Journal, 2012(10): 3173-3182.
［8］ 李彬,李贻斌,荣学文.ELMRBF神经网络的智能优化策略［J］.山东大学学报：理学版,2010, 05: 48-51.
LI Bin, LI Yibin, RONG Xuewen. Intelligent optimization strategy for ELMRBF neural networks ［J］. Journal of Shandong University: Natural Science, 2010, 05: 48-51.
［9］ 李栋,张文宇.基于ELM和FOA的股票价格预测［J］.计算机工程与应用,2014,18: 14-18.
LI Dong, ZHANG Wenyu. Stock price prediction based on ELM and FOA ［J］. Computer Engineering and Applications, 2014, 18: 14-18.
［10］ 陈爱军.最小二乘支持向量机及其在工业过程建模中的应用［D］.杭州：浙江大学, 2006.
CHEN Aijun. The study of ｌeast ｓquares ｓupport ｖector ｍachine and ｉts ａpplication in ｉndustrial ｐrocess ｍodeling ［D］. Ｈａｎｇｚｈｏｕ: Zhejiang University, 2006.
［11］ PEARSON K. On lines and planes of closest fit to systems of points in space ［J］. Philosophical Magazine, 1901,2(11): 559572.
［12］ JOLLIFFE I T. Principal component analysis ［M］. New York: Springer,2002.
［13］ 李玉鑑,王利娟.基于PCA的眉毛识别方法研究［J］.计算机工程与科学,2008(11): 28-30.
LI Yujian, WANG Lijuan. Eyebrow ｒecognition ｂased on ｐrincipal ｃomponent ａnalysis ［J］. Computer Engineering ａｎｄ Science, 2008(11): 28-30.
［14］ DOWNS J J, VOGEL E F. A ｐlantwide ｉndustrial ｐrocess ｃontrol ｐroblem ［J］. Computers ａｎｄ Chemical Engineering, 1993, 17(3): 245-255.
