Abstract:High-precision fault diagnosis is important for the widely installed complex industrial product, the wind turbine. However, intelligent monitoring is difficult due to the fuzzy boundaries and individual different variations of the unseen single or simultaneous-fault of such intricate equipment. To solve this problem, this study proposes an ensemble fault diagnostic framework for simultaneous and coupling failure. First, this paper develops novel signal processing methods for effective feature learning and mapping from the non-stationary and nonlinear raw vibrational signals. The adapted variational mode decomposition is introduced based on the particle swarm optimization that applies the minimum mean envelope entropy to optimize the parameters settings. Second, the novel ensemble extreme learning machine-based network is proposed to isolate the faults that applies one extreme learning machine network to count the number of fault scenarios, and the other one to identify the specific single or simultaneous-fault labels. With this scheme, the self-adaptive ensemble extreme learning machine-based fault diagnostic framework is more accurate and faster than the prevailing probabilistic classifier-based methods, as the proposed method does not rely on empirically specified decision-making threshold and generates all the candidate fault labels at the same time. Finally, this study builds the test platform and compares the overall results with the existing feature analysis methods and classifiers. The experimental results verify that the proposed framework detects both single and simultaneous-fault accurately and quickly.

On Prognosis of Wind Turbine Faults Based on Nonlinear Mixed Vibration Signals: A PSO Based EMD and KICA Combined Approach

Particle Swarm Optimization-Based Empirical Mode Decomposition-Kernel Independent Component Analysis Joint Approach for Diagnosing Wind Turbine Gearbox with Multiple Faults

ICOPE-15-C022 Fault Diagnosis for Steam Turbine Flow Passage Using EMD-KPCA Feature Extraction and Support Vector Machine

Condition Monitoring And Fault Diagnosis Of Rotating Machines With Blind Source Separation

Diagnosis of non-linear mixed multiple faults based on underdetermined blind source separation for wind turbine gearbox: simulation, testbed and realistic scenarios

Sparse Component Analysis-Based Under-Determined Blind Source Separation for Bearing Fault Feature Extraction in Wind Turbine Gearbox

Study on novel signal processing and simultaneous-fault diagnostic method for wind turbine

On The Use Of Eemd And Svm Based Approach For Bearing Fault Diagnosis Of Wind Turbine Gearbox

Empirical Mode Decomposition for Blind Separation of Modulation Fault Source Signals

Fault Detection of Wind Turbines by Subspace Reconstruction-Based Robust Kernel Principal Component Analysis

A Rolling Bearing Fault Diagnosis Method Based on EEMD-WSST Signal Reconstruction and Multi-Scale Entropy

Fault feature extraction and diagnosis of rolling bearings based on wavelet thresholding denoising with CEEMDAN energy entropy and PSO-LSSVM

A Fault Diagnosis Method for Rotating Bearings Based on EWT Multi-Scale Entropy and PSO Algorithm to Optimize SVM

Piecewise Hybrid System with Cross-Correlation Spectral Kurtosis for Fault Diagnosis in Rolling Bearing of Wind Power Generator

Fault Detection of Bearing Systems through EEMD and Optimization Algorithm

Fault Diagnosis of Wind Turbine Bearings Based on CEEMDAN-GWO-KELM

Application of Variational Mode Decomposition to Feature Isolation and Diagnosis in a Wind Turbine

Research on the Rotor Fault Diagnosis Method Based on QPSO-VMD-PCA-SVM

Application of the Ensemble Empirical Mode Decomposition and Hilbert Transform to Pedestal Looseness Study of Direct-Drive Wind Turbine

Intelligent Mechanical Fault Diagnosis Based on Multiwavelet Adaptive Threshold Denoising and MPSO

Data-Driven Diagnosis of Nonlinearly Mixed Mechanical Faults in Wind Turbine Gearbox