Feature extraction method of ship radiated noise based on BOA-VMD and slope entropy

Although the technical requirements for feature extraction of ship radiated noise (SRN) in fields national defense and economy increase with each passing day, complexity marine environment makes SRN difficult. The traditional method based on variational mode decomposition (VMD) is widely used SRN. Nevertheless, use VMD greatly affected by parameters. In this paper, butterfly optimization algorithm (BOA) introduced to optimize VMD, which called BOA-VMD algorithm, realizes optimal selection parameters K id="m2">? . To further improve efficiency method, combined slope entropy (SE), a SE proposed. experimental results simulated signal show that has smaller envelope value better effect than genetic (GA) particle swarm (PSO). highest recognition rate four values number extracted features, compared three dispersion (DE), fluctuation (FDE) permutation (PE), under different quantity reached 100% features.