Abstract:Strong noise interference or compound fault coupling phenomenon may lead to the failure of fault diagnosis technology. This paper focuses on weak feature extraction and compound faults detection for rotating machinery fault diagnosis and proposes adaptive symplectic geometric mode decomposition (SGMD) using cycle kurtosis entropy. Firstly, an index named cycle kurtosis entropy (CKE) is presented to measure the strength of periodic impulses in a signal. The CKE uses the entropy value of calculating all delay cycle kurtosis (CK) to overcome the shortcomings of the CK in adaptive ability and obtain more stable values. Thirdly, CKE is applied to construct an adaptive slip window with optimal length. This process is called the adaptive window segmentation method, which is mainly used to dig out weak fault features in signals. Finally, CKE is used as the component selection criterion to select the components decomposed by SGMD. The selected components are reconstructed to obtain a de-noised signal. Hilbert envelope analysis is applied to the denoised signal to demodulate the fault characteristic frequency. Numerical simulations and experimental investigations using bearings and gears are performed to testify the property of the presented method. The results indicate that the adaptive slip window can enhance the decomposing ability of SGMD under strong noise condition. Moreover, for the strong periodic impulse identification ability, the cycle kurtosis entropy is suitable to determine the optimal components of SGMD. It is expected that the presented method will be effectively used for fault feature extractions in rotating machinery under stationary running conditions.

Fault Diagnosis for the Intermittent Fault in Gyroscopes: A Data-Driven Method

Model-based fault diagnosis method for gyro

Fault Detection in Non-Gaussian Vibration Systems Using Dynamic Statistical-Based Approaches

A New Dual-Mass MEMS Gyroscope Fault Diagnosis Platform

Satellite gyroscope fault diagnosis based on parity relation and empirical mode decomposition

Sensor fault identification and isolation for multivariate non-Gaussian processes

Fault Diagnosis for Attitude Sensors Via a Bank of Extended Kalman Filters

Detecting and Estimating Intermittent Actuator Faults in Linear Stochastic Systems.

Fault Detection for Satellite Gyroscope Using LSTM Networks

A Novel Fault Detection and Isolation Technology for Redundant Inertial Measurement Unit

Fault Detection and Isolation of Satellite Gyroscopes Using Relative Positions in Formation Flying

Satellite Gyroscopes Fault Diagnosis Based on Numerical Differentiation Scheme

Fault Diagnosis for Attitude Sensors Based on Analytical Redundancy and EMD

Gyro motor fault classification model based on a coupled hidden Markov model with a minimum intra-class distance algorithm

Fault Diagnosis for Satellite Attitude Control Systems with Four Flywheels

A Random Forest and Model-Based Hybrid Method of Fault Diagnosis for Satellite Attitude Control Systems

Generalized Vold–Kalman Filtering for Nonstationary Compound Faults Feature Extraction of Bearing and Gear

Detection, Isolation and Diagnosability Analysis of Intermittent Faults in Stochastic Systems

Cycle kurtosis entropy guided symplectic geometry mode decomposition for detecting faults in rotating machinery

A Novel Fault Diagnosis Method for Dual-Angle Measurement System