A GWO-AFSA-SVM Model-Based Fault Pattern Recognition for the Power Equipment of Autonomous vessels

Abstract

The power equipment of autonomous vessels usually works in a particular environment, which provides few samples of fault monitoring signal. The method of the support vector machine (SVM) is adopted to deal with the problem of fault pattern identification under small sample conditions. An improved GWO-AFSA algorithm is proposed to avoid the poor convergence accuracy caused by the random selection of kernel parameters and penalty factors for SVM. The Grey Wolf algorithm (GWO) is adopted to optimize the erratic behavior of fish swarm, which improves the problem that the traditional Artificial Fish Swarm Algorithm (AFSA) is easy to fall into local extremum. A benchmark example demonstrates that the GWO-AFSA-SVM model can accurately and effectively identify the fault pattern types of ship power equipment for autonomous vessels.