Fault Diagnosis Method Based on CND-SMOTE and BA-SVM Algorithm

Abstract The problem of unbalanced data classification has gotten extensive attention in the past few years. Unbalanced sample makes fault diagnosis and accuracy rate low, capability to classify minority-class samples is restricted. To address that algorithm machine learning insufficient identify minority class for problems. Therefore, this paper proposes an improved support vector (SVM) method based on synthetic over-sampling technique (SMOTE). For sampler, characteristics neighborhood distribution (CND-SMOTE) used equilibrate majority samples. classifier, parameter optimization machines bat (BA-SVM) solve multi-classification faulty Finally, experimental results prove CND-SMOTE+BA-SVM can synthesize high-quality samples, increase decrease time spent classification.