A Fault Diagnosis Method for Rolling Bearings Based on Improved Snake Swarm Optimization and Support Vector Machine

Support vector machine (SVM) is widely used in rolling bearing fault diagnosis because of its excellent performance in minor sample problems. However, SVM performance is highly dependent on the choice of hyperparameters, such as the regularization parameter <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$C$</tex> and the kernel parameter <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\gamma$</tex>, and different combinations of hyperparameters can lead to significant differences in model performance. This paper proposes a fault diagnosis classification model that optimizes support vector machines with an improved snake swarm optimization algorithm (SOA). By introducing adaptive speed update, dynamic weight adjustment, and diversity maintenance strategies, the improved SOA significantly enhances the global search capability and convergence speed and effectively avoids the risk of falling into local optimal. Compared with the traditional hyperparameter optimization method, this method makes up for the inherent defects of SVM. It significantly improves the accuracy and efficiency of fault diagnosis, providing an effective and efficient solution for rolling bearing fault diagnosis.