In this study, we propose a real-time ship anomaly detection method driven by Automatic Identification System (AIS) data. The method uses ship trajectory clustering classes as a normal model and a deep learning algorithm as an anomaly detection tool. The method is divided into three main steps: (1) quality maintenance of the original AIS data, (2) extraction of normal ship trajectory clusters using Density-Based Spatial Clustering of Applications with Noise (DBSCAN), in which a segmented improved Dynamic Time Warping (DTW) algorithm is used to measure the degree of trajectory similarity, (3) the clustering results are used as a normative model to train a Bi-directional Gated Recurrent Unit (BiGRU) recurrent neural network, which is used as a trajectory predictor to achieve real-time ship anomaly detection. Experiments were conducted using real AIS data from the port of Tianjin, China. The experimental results are manifold. Firstly, the data pre-processing process effectively improves the quality of raw AIS data. Secondly, the ship trajectory clustering model can accurately classify the traffic flow of different modes in the sea area. Moreover, the trajectory prediction result of the BiGRU model has the smallest error with the actual ship trajectory and has a better trajectory prediction performance compared with the Long Short-Term Memory Network model (LSTM) and Gated Recurrent Unit (GRU). In the final anomaly detection experiment, the detection accuracy and timeliness of the BiGRU model are also higher than LSTM and GRU. Therefore, the proposed method can achieve effective and timely detection of ship anomalous behaviors in terms of position, heading and speed during ship navigation, which provides insight to enhance the intelligence of marine traffic supervision and improve the safety of marine navigation.

Timely and effective maintenance is imperative to minimize operational disruptions and ensure the reliability of marine vessels. However, given the low early warning rates and poor adaptability under complex conditions of previous data-driven fault prediction methods, this paper presents a hybrid deep learning model based on multi-layer attention mechanisms for predicting faults in a marine diesel engine. Specifically, this hybrid model first introduces a Convolutional Neural Network (CNN) and self-attention to extract local features from multi-feature input sequences. Then, we utilize Long Short-Term Memory (LSTM) and multi-head attention to capture global correlations across time steps. Finally, the hybrid deep learning model is integrated with the Exponential Weighted Moving Average (EWMA) to monitor the operational status and predict potential faults in the marine diesel engine. We conducted extensive evaluations using real datasets under three operating conditions. The experimental results indicate that the proposed method outperforms the current state-of-the-art methods. Moreover, ablation studies and visualizations highlight the importance of fusing multi-layer attention, and the results under various operating conditions and application scenarios demonstrate that this method possesses predictive accuracy and broad applicability. Hence, this approach can provide decision support for condition monitoring and predictive maintenance of marine mechanical systems.

We propose a theoretical method for constructing an initial two-component wave function that can be transformed into a knotted velocity field with finite kinetic energy and enstrophy. The wave function is constructed using two complex-valued polynomials, with one determining the desired shape of the knotted central axis and the other encoding the twisting nature of vortex lines, which facilitates the study of helicity conversions. We construct six knotted vortex fields with various centerline and twist helicity as initial conditions for direct numerical simulation of incompressible Schrödinger flow (ISF) in a periodic box. Although the evolution of morphological structure is similar for ISF and classical viscous flow, with all the knots becoming untied after a short time to form one or more separate vortex rings, their statistics are quite different. During the critical period of vortex reconnection, the increase in enstrophy is much more moderate in ISF than in viscous flow, indicating that the Landau–Lifshitz term in ISF inhibits the energy cascade from large to small scales. We also find that the centerline helicity changes dramatically during reconnection, which is consistent with the evolution of the geometrical shape of vortex lines.

Under the current background of navigation education, students generally lack a comprehensive grasp of ship firefighting equipment’s operation. Therefore, we develop a novel ship firefighting training simulator with a multi-sensory human–computer interaction function for teaching and training marine students. In the simulator, we consider a container ship of 1.8w containers as the prototype ship, and the entire ship models are built using three-dimensional modeling technology. We design various interactive modes and realize a full-process operation simulation of several standard ship firefighting equipment. Furthermore, we propose a purely Lagrangian vortex dynamics framework to simulate smoke and flame in fire scenarios. In this framework, we model fluids using velocity and vorticity fields discretized on discrete vortex segments. The main components of the framework include a stable geometric stretching solution and particle strength exchange method for solving the diffusion term. The simulation results show that the simulator has good behavioral realism and scene immersion and can be applied to ship firefighting training. To the best of our knowledge, this is the first study on real-time smoke simulation using a physics-based method in a firefighting training simulator.

For the design of ship equipment and crew training, it would be useful to develop software for the three-dimensional simulation of a totally enclosed lifeboat. To improve the simulation accuracy and immersion of the software, we present a multibody dynamics model for a lifeboat lowered from a ship, accounting for the coupled motion among the ship, lifeboat slings, a cable-pulley system, and the lifeboat. The equations of the whole system are formulated using Kane's method. The model of ship manoeuvring mathematical group is used to calculate the forces and moments acting on the hull of the ship. The hydrodynamic and wind loads on the boat are modeled using the strip theory. The impact force between the ship and the boat colliding is estimated using the contact theory of Hertz and an elastoplastic model. The method of lumped mass is used to model the lifeboat slings. For the cable-pulley system, we present an efficient model for the dynamics of the pay-out/reel-in process based on the framework of Kane's method. The local load of each cable segment between two pulleys is calculated by the model of a linear spring on the basis of the amount of cable passages over pulleys and the variation of the pulley positions, conversely, the cable segment exerts force and moment on the pulleys. The motion equations of the whole system are solved using fourth-order Runge-Kutta. The model can simulate the lowering of the lifeboat, and obtain the three-dimensional motion parameters of the ship, the lifeboat, slings and pulleys, and the local tension load of the cable. The results show that the simulation curves are near the ones of the model experiment, and their trends are coincident. Thus, it can be concluded that our model is feasible. According to our model, the motion of the ship has a significant effect on the magnitude of the lifeboat's oscillation when the sea state is above level 4; it is safe when the initial clear distance is greater than 1.5 times the width of the boat and the sea conditions are below level 5. Finally, the model is applied to the software for three-dimensional simulation.