A new framework of sensor selection for developing a fault detection system based on data-envelopment analysis
Several methods have been proposed to identify which sensor sets are optimal for finding and localizing faults under different conditions for mechanical equipment. In order to preserve acceptable performance while minimizing costs, it is crucial to identify the most effective set of sensors available. Nevertheless, some sensor sets can behave differently in fault detection because of uncertainty in the measurement data contaminated by noise interference. This paper develops new sensor selection models using Data Envelopment Analysis (DEA), which has proven helpful as a management approach for determining an optimal number of sensors, associated costs, and sensor performance in the fault diagnosis. We propose four linear optimization models for sensor selection to design the fault detection system. The validity of the presented models is demonstrated using a vibration dataset collected from a gearbox. Based on the case study results, the proposed methods are effectively superior to the other comparison sensor selection methods in fault detection accuracy.
Code (0)
등록된 구현이 없습니다.
Tasks
Fault DetectionFault DiagnosisManagementMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
Enhancing Fault Detection in CO2 Refrigeration Systems: Optimal Sensor Selection and Robustness Analysis Using Tree-Based Machine Learning
This study investigates the reliability and robustness of data-driven Fault Detection and Diagnosis (FDD) models for CO2 refrigeration systems (CO2-RS) in supermarkets, focusing on optimal sensor selection and resilience…
Fault DetectionCASU2Net: Cascaded Unification Network by a Two-step Early Fusion for Fault Detection in Offshore Wind Turbines
This paper presents a novel feature fusion-based deep learning model (called CASU2Net) for fault detection in offshore wind turbines. The proposed CASU2Net model benefits of a two-step early fusion to enrich features in …
Fault DetectionTime SeriesTime Series AnalysisNeural Associative Skill Memories for safer robotics and modelling human sensorimotor repertoires
Modern robots face challenges shared by humans, where machines must learn multiple sensorimotor skills and express them adaptively. Equipping robots with a human-like memory of how it feels to do multiple stereotypical m…
Fault DetectionA Framework for the Automated Parameterization of a Sensorless Bearing Fault Detection Pipeline
This study proposes a framework for the automated hyperparameter optimization of a bearing fault detection pipeline for permanent magnet synchronous motors (PMSMs) without the need of external sensors. A automated machin…
AutoMLFault DetectionHyperparameter OptimizationMultiple profiles sensor-based monitoring and anomaly detection
Generally, in an advanced manufacturing system hundreds of sensors are deployed to measure key process variables in real time. Thus it is desirable to develop methodologies to use real-time sensor data for on-line syst…
Anomaly Detection