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.