Intelligent bearing fault diagnosis using swarm decomposition method and new hybrid particle swarm optimization algorithm

The quality of information extracted from the vibration signals, and accuracy bearing status detection depend on methods used to process signal select informative features. In this paper, a new hybrid approach is introduced in which relatively swarm decomposition (SWD) method optimized compensation distance evaluation technique (OCDET) are enhance processing stage improve optimal features selection process, respectively. Firstly, signals decomposed into their Oscillatory Components (OCs) using SWD. feature matrix constructed by computing time-domain for OCs. CDET consequently utilized most sensitive corresponding status. On other hand, contains parameter called threshold affects number selected way, optimization algorithm, combination Particle Swarm Optimization (PSO) algorithm with Sine–Cosine Algorithm (SCA) Levy flight distribution, has been support vector machine (SVM) classifier. proposed ability evaluated different defects various speeds. results indicate capability fault diagnosis identifying very small-size under conditions. Finally, presented shows better performance comparison well-known case studies.