A Feature Selection Approach Based on Memory Space Computation Genetic Algorithm Applied in Bearing Fault Diagnosis Model

The main objective of this study is to propose a motor fault diagnosis model based on machine learning. Compared with the traditional model, proposed can reduce computation time. This be divided into three steps: feature extraction, selection, and classification. In extraction step, original signal extracted by Hilbert-Huang transform (HHT), envelope analysis (EA), variational mode decomposition (VMD) methods. A selection method memory space genetic algorithm (MSCGA) applied in step. advantage MSCGA that it eliminates need compute data fitness values, saving unnecessary time repeatedly. classifiers use k-nearest neighbor (KNN) support vector machines (SVM). order verify stability efficiency university California Irvine (UCI) benchmark dataset, current datasets, case western reserve (CWRU) were used. UCI dataset used test method. Other datasets are compare models. simulation results have demonstrated effectiveness reducing without affecting compared model. Furthermore, performance proven better than other algorithm.