A Feature Selection Approach Hybrid Grey Wolf and Heap-Based Optimizer Applied in Bearing Fault Diagnosis

An effective bearing fault diagnosis model based on machine learning is proposed in this study. The can separate into three stages: feature extraction, selection, and classification. In the stage of multiresolution analysis (MRA) fast Fourier transform (FFT) are applied to extract features from raw signal measured rotating machine. second stage, a powerful selection method utilized selection. new grey wolf optimization (GWO) heap-based optimizer (HBO) with strategies combined. Finally, support vector (SVM) linear discriminant (LDA) used as classifier independently. To verify capability model, four different datasets test study, respectively University California Irvine (UCI) benchmark dataset, Case Western Reserve (CWRU) Machinery Failure Prevention Technology (MFPT) dataset. compared existing methods certify robustness experiment results.