中文摘要：

针对船舶推进轴系的振动问题,基于小波包、Shannon熵、遗传算法（GA）和支持向量机（SVM）理论,提出了一种船舶轴系故障诊断的新方法,简称WPS-GS方法.该方法依托船舶螺旋桨状态监测模拟实验平台,利用小波包分解技术分析船舶轴系发生故障时的振动信号,将其Shannon熵作为SVM的输入特征向量.在训练SVM时,采用遗传算法对SVM的参数进行全局寻优,使SVM具有更高的识别准确率.实验结果表明,WPS-GS方法对故障诊断的准确度和识别率较传统SVM和交叉验证SVM方法高,适用于船舶轴系故障诊断.

英文摘要：

Aiming at the vibration problems of the ship which is based on the theory of wavelet packet （wavelet propulsion shafting, a new method of fault diagnosis, packet）, Shannon entropy, genetic algorithm （GA） and support vector machine （SVM）, is proposed, and referred to as WPS-GS method in this paper. For the simulation of platform ship propeller shafting, the wavelet packet decomposition and strong fault-tolerant Shannon entropy are jointly used to compute the feature vectors of vibration signals which are served as the input vectors of SVM; GA is adopted to optimize the parameters of SVM for achieving the higher veracity. The simulation results show that the WPS-GS method can attain higher reliability and veracity than the conventional SVM and K-CV SVM, which suggests that the proposed method is more suitable for the condition monitoring and fault diagnosis of rotating shaft system.