Parameters Optimization of SVM Based on Improved FOA and Its Application in Fault Diagnosis

摘要

In most cases, fault diagnosis is essentially a pattern recognition problem and support vector machine (SVM) provides a new solution for the diagnosis problem of systems in which the fault samples are few. However, the parameters selection in SVM has significant influence on the diagnosis performance. In this paper, improved fruit fly optimization algorithm (IFOA), which is basically the standard fruit fly optimization algorithm (FOA) combined with Levy flight search strategy, is proposed to determine the SVM parameters. Some benchmark datasets are used to evaluate the proposed algorithm. Furthermore, the proposed method is used to diagnose the faults of hydraulic pump. Experiments and engineering application show that the proposed method outperforms standard FOA, genetic algorithm (GA) and particle swarm optimization (PSO) methods.