A Fault Diagnosis Approach Integrated LPP with AROMF for Process Industry

Yuan Xu, Zixu Wang, Wei Ke, Yan Lin He, Qun Xiong Zhu, Yang Zhang

研究成果: Conference contribution同行評審

摘要

As the data collected in the industrial process presents high-dimensional and nonlinear characteristics, it brings great challenges to the realization of timely and effective fault diagnosis. In this article, a fault diagnosis method is proposed integrated local preserving projections (LPP) with adaptive rank-order morphological filter (AROMF). First, in order to deal with the problem of high-dimensional and non-linearity of data, LPP algorithm is used to extract the required template trend and test trend. Second, AROMF performs morphological transformation on the test trend under the supervision of the template trend to obtain the output trend signal. Third, the iterative total error of the output trend and the corresponding template trend is calculated to classify the fault. Finally, the proposed method is verified by simulation on the Three-phase flow facility (TFF) dataset. The simulation results prove that this method can improve the accuracy of fault diagnosis.

原文English
主出版物標題Proceedings of 2022 IEEE 11th Data Driven Control and Learning Systems Conference, DDCLS 2022
編輯Mingxuan Sun, Zengqiang Chen
發行者Institute of Electrical and Electronics Engineers Inc.
頁面734-739
頁數6
ISBN(電子)9781665496759
DOIs
出版狀態Published - 2022
事件11th IEEE Data Driven Control and Learning Systems Conference, DDCLS 2022 - Emeishan, China
持續時間: 3 8月 20225 8月 2022

出版系列

名字Proceedings of 2022 IEEE 11th Data Driven Control and Learning Systems Conference, DDCLS 2022

Conference

Conference11th IEEE Data Driven Control and Learning Systems Conference, DDCLS 2022
國家/地區China
城市Emeishan
期間3/08/225/08/22

指紋

深入研究「A Fault Diagnosis Approach Integrated LPP with AROMF for Process Industry」主題。共同形成了獨特的指紋。

引用此