Condition monitoring scheme via one-class support vector machine and multivariate control charts
- PDF / 713,894 Bytes
- 8 Pages / 595.22 x 842 pts (A4) Page_size
- 44 Downloads / 177 Views
DOI 10.1007/s12206-020-2203-z
Journal of Mechanical Science and Technology 34 (0) 2020 Original Article DOI 10.1007/s12206-020-2203-z Keywords: · Condition-based maintenance · Hotelling’s T 2 chart · Nonparametric control chart · One class support vector machine · Prognostics and health management
Correspondence to: Suk Joo Bae [email protected]
Citation: Mun, B. M., Lim, M., Bae, S. J. (2020). Condition monitoring scheme via oneclass support vector machine and multivariate control charts. Journal of Mechanical Science and Technology 34 (0) (2020) ?~?. http://doi.org/10.1007/s12206-020-2203-z
Condition monitoring scheme via one-class support vector machine and multivariate control charts Byeong Min Mun, Munwon Lim and Suk Joo Bae Department of Industrial Engineering, Hanyang University, Seoul, Korea
Abstract
A condition-based maintenance (CBM) has been widely employed to reduce maintenance cost by predicting the health status of many complex systems in prognostics and health management (PHM) framework. Recently, multivariate control charts used in statistical process control (SPC) have been actively introduced as monitoring technology. In this paper, we propose a condition monitoring scheme to monitor the health status of the system of interest. In our condition monitoring scheme, we first define reference data set using one-class support vector machine (OC-SVM) to construct the control limit of multivariate control charts in phase I. Then, parametric control chart or non-parametric control chart is selected according to the results from multivariate normality tests. The proposed condition monitoring scheme is applied to sensor data of two anemometers to evaluate the performance of fault detection power.
Received April 27th, 2020 Revised
May 12th, 2020
Accepted May 12th, 2020
1. Introduction
† This paper was presented at ICMR2019, Maison Glad Jeju, Jeju, Korea, November 27-29, 2019. Recommended by Guest Editor Insu Jeon
Prognostics and health management (PHM) has been widely used to monitor the health status of the operating system. Its future health status can be efficiently predicted using advance sensing technology and artificial intelligence (AI) for the purpose of fault isolation or reliability prediction [1]. PHM methodology targets to provide prognostic information or knowledge on the monitoring system to prevent catastrophic failures of the system by predicting the time to failure or estimating remaining useful life during operation, resulting in significantly reducing total operational cost. In applying PHM concept to maintenance, a condition-based maintenance (CBM) has an increasing attention in many complex systems (e.g., power plants, large transportation vehicles) as a predictive maintenance approach. CBM activities have been conducted based on the prediction of remaining useful life until a failure occurs, thus it can greatly reduce maintenance cost caused by unnecessary preventive maintenance tasks [2]. In many practical applications, univariate or multivariate control charts have been
Data Loading...
