Received: 4 May 2020 Revised: 31 August 2020 Accepted: 1 September 2020 DOI: 10.1002/eng2.12291
RESEARCH ARTICLE
Early failure detection of paper manufacturing machinery using nearest neighbor-based feature extraction
WonjaeLee 1 Kangwon Seo 1,2
1 Department of Industrial and Manufacturing Systems Engineering, University of Missouri, Columbia, Missouri, 2 Department of Statistics, University of Missouri, Columbia, Missouri, Correspondence Kangwon Seo, Department of Industrial and Manufacturing Systems Engineering, University of Missouri, E3437M Thomas & Nell Lafferre Hall, Columbia, MO 65211, USA. Email: seoka@missouri.edu
Abstract In a paper manufacturing system, it is substantially important to detect machine failure before it occurs and take necessary maintenance actions to prevent an unexpected breakdown of the system. Multiple sensor data collected from a machine provides useful information on the system’s health condition. How- ever, it is hard to predict the system condition ahead of time due to the lack of clear ominous signs for future failures, a rare occurrence of failure events, and a wide range of sensor signals which might be correlated with each other. We present two versions of feature extraction techniques based on the near- est neighbor combined with machine learning algorithms to detect a failure of the paper manufacturing machinery earlier than its occurrence from the mul- tistream system monitoring data. First, for each sensor stream, the time series data is transformed into the binary form by extracting the class label of the near- est neighbor. We feed these transformed features into the decision tree classifier for the failure classification. Second, expanding the idea, the relative distance to the local nearest neighbor has been measured, results in the real-valued feature, and the support vector machine is used as a classifier. Our proposed algorithms are applied to the dataset provided by Institute of Industrial and Systems Engi- neers 2019 data competition, and the results show better performance than other state-of-the-art machine learning techniques. KEYWORDS 1-nearest neighbor, feature extraction, multistream time series classification, rare event prediction, relative distance
1 INTRODUCTION The pulp and paper production requires highly complex and integrated processes by chemical or mechanical means, which include wood preparation, pulping, chemical recovery, bleaching, and papermaking. 1 In the advanced papermak- ing facilities, the systems are continuously monitored so that the operators can manage and control the processes, and detect any possible incidents that might cause an abrupt production break. To do this, a wide range of sensors are deployed in many different parts of manufacturing equipment to measure important process variables and monitor the system sta- This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2020 The Authors. Engineering Reports published by John Wiley & Sons Ltd. Engineering Reports . 2020;e12291. wileyonlinelibrary.com/journal/eng2 1of 19 https://doi.org/10.1002/eng2.12291
Made with FlippingBook - Online catalogs