Composite Fault Diagnosis of Rolling Bearing Based on Chaotic Honey Badger Algorithm Optimizing VMD and ELM

In order to effectively extract the characteristic information of bearing vibration signals and improve classification accuracy, a composite fault diagnosis method rolling based on chaotic honey badger algorithm (CHBA), which optimizes variational mode decomposition (VMD) extreme learning machine (ELM), is proposed in this paper. Firstly, aiming solve problem that HBA optimization process can easily fall into local slow convergence speed, sinusoidal mapping introduced HBA, advantages CHBA are verified by 23 benchmark functions. Then, taking Gini index square envelope (GISE) as fitness function, VMD optimized with obtain optimal number modes K quadratic penalty factor. Secondly, first four IMF components largest GISE values selected, grouped “Systematic Sampling Method (SSM)” calculate signal energy form feature vector. Finally, error rate vector input ELM model classify identify different types faults. Through experimental analysis, compared BP, ELM, GWO-ELM, HBA-ELM, has better results for faults, accuracy reach 100%, provides new way diagnosis.