船舶齿轮箱故障诊断及失效预报方法研究

Research on fault diagnosis and failure prediction method for marine gearbox

针对船舶齿轮箱故障诊断正确率低的难题,提出蚁群优化神经网络的船舶齿轮箱故障诊断方法。首先采集船舶齿轮箱故障诊断的数据,并采用小波分析提取船舶齿轮箱故障诊断特征,然后采用神经网络建立船舶齿轮箱故障诊断模型,并采用蚁群算法克服神经网络存在的缺陷,最后构建了船舶齿轮箱失效预测方法,实验表明,本文方法提高了船舶齿轮箱故障诊断效果,并获得了高精度的船舶齿轮箱失效预测结果。

In view of the problem of the accuracy of the fault diagnosis of the ship gear box, an ant colony optimization neural network has been proposed for the fault diagnosis of the ship gear box. First, the data of the fault diagnosis of the ship gear box is collected, and the wavelet analysis is used to take the fault diagnosis feature of the ship gear box. Then the neural network is used to establish the fault diagnosis model of the ship gearbox, and the ant colony algorithm is used to overcome the defects of the neural network. Finally, the ship's tooth is constructed. The failure prediction method of the wheel box is proved by the experiment. This method improves the fault diagnosis effect of the ship gear box, and obtains the high precision failure prediction result of the ship gear box.