关键词: 状态监测, 核主成分分析, Renyi熵, 特征提取

Abstract: Aiming at the problem that the conventional Kernel Principle Component Analysis (KPCA) approaches depended on experiences to select parameters too much,a state monitoring method based on integrated entropy of KPCA was proposed,which combined conventional KPCA with the information entropy to determine the parameters by information measuring in the high dimensional space.The kernel principal component was extracted with Renyi entropy estimator.Operating state was monitored by forming synthesis statistics variables.The simulation on the TE process and the compressor system of a enterprise were applied to prove the better capability of processing nonlinear data and the more accuracy rate on detecting the fault or abnormal of the proposed approach comparing with conventional methods of KPCA.

Key words: status monitoring, kernel principal component analysis, Renyi entropy, feature extraction