Abstract: The reliability of hydraulic pipelines in waterjet propulsion devices is crucial to the navigation safety of waterjet-propelled ships. To overcome the limitations of traditional methods in real-time detection, positioning accuracy, and environmental adaptability, this paper proposes a method for leak detection and localization of waterjet propulsion pipelines based on a backpropagation (BP) neural network and Amesim simulation. A hydraulic system simulation model of a certain type of waterjet propulsion device is built using Amesim software and system simulation signals under different pipeline leak locations are collected to form the dataset. A BP neural network model is then constructed, trained, and validated to achieve fault localization. The prediction results of the BP neural network model are in good agreement with experimental measurements, showing a strong linear relationship, which indicates that the BP neural network model has satisfactory predictive performance. This study can provide a technical reference for precise leak localization and condition assessment of hydraulic pipelines in complex ship cabin environments.

Key words: ship power unit, pipeline inspection, Amesim simulation, waterjet propulsion, backpropagation (BP) neural network

摘要： 喷水推进装置液压管路可靠性对喷水推进船舶的航行安全至关重要,为弥补传统方法在实时性检测、定位精度及环境适应性上的不足,该文提出了一种基于BP神经网络与Amesim仿真的喷水推进管道泄漏定位检测方法。文章通过Amesim软件搭建某型喷水推进装置液压系统仿真模型,采集不同位置泄漏管路的系统仿真信号,完成数据收集;再构建BP神经网络模型完成网络训练与验证,实现系统的故障定位检测。BP神经网络模型预测结果和实验测量结果高度吻合,两者呈高度线性关系,表明BP 神经网络模型具有良好的预测效果,可为复杂船舱环境下的液压管路泄漏精准定位及状态评估提供技术参考。

关键词: 船舶动力装置, 管路检测, Amesim仿真, 喷水推进装置, 反向传播神经网络