Litcius/Paper detail

Early fault detection based on empirical mode decomposition method

Akash Patel, Piyush Shakya

2020Procedia CIRP14 citationsDOIOpen Access PDF

Abstract

Vibration signal analysis is a widely used condition monitoring technique. Though the extraction of information from the raw vibration signals is difficult because of its non-stationary and non-linear nature, the Empirical mode decomposition has proven to be an effective method. In this method, the raw vibration signal is decomposed into the various intrinsic mode functions. Further, the energy content of these intrinsic modes and the spectral entropy’s associated with each intrinsic mode is analyzed. The method is validated with the available data sets. Based on the results obtained, this method has proven to work better for early fault detection.

Topics & Concepts

Hilbert–Huang transformVibrationMode (computer interface)Entropy (arrow of time)SIGNAL (programming language)Computer scienceEnergy (signal processing)Fault (geology)Fault detection and isolationSignal processingPattern recognition (psychology)AlgorithmBiological systemControl theory (sociology)EngineeringElectronic engineeringAcousticsArtificial intelligenceMathematicsPhysicsStatisticsDigital signal processingQuantum mechanicsProgramming languageGeologyBiologyControl (management)SeismologyActuatorOperating systemMachine Fault Diagnosis TechniquesStructural Health Monitoring TechniquesStructural Integrity and Reliability Analysis
Early fault detection based on empirical mode decomposition method | Litcius