As port operations rapidly evolve toward intelligent and heavy-duty applications, fault diagnosis for core equipment demands higher levels of real-time performance and robustness. Deep learning, with its powerful autonomous feature learning capabilities, demonstrates significant potential in mechanical fault prediction and health management. This paper first provides a systematic review of deep learning research advances in rotating machinery fault diagnosis over the past eight years, focusing on the technical approaches and application cases of four representative models: Deep Belief Networks (DBNs), Convolutional Neural Networks (CNNs), Auto-encoders (AEs), and Recurrent Neural Networks (RNNs). These models, respectively, embody four core paradigms, unsupervised feature generation, spatial pattern extraction, data reconstruction learning, and temporal dependency modeling, forming the technological foundation of contemporary intelligent diagnostics. Building upon this foundation, this paper delves into the unique challenges encountered when transferring these methods from generic laboratory components to specialized port equipment such as shore cranes and yard cranes—including complex operating conditions, harsh environments, and system coupling. It further explores future research directions, including cross-condition transfer, multi-source information fusion, and lightweight deployment, aiming to provide theoretical references and implementation pathways for the technological advancement of intelligent operation and maintenance in port equipment.

The interfacial interaction between the main oxidant filler ammonium perchlorate (AP) and hydroxyl-terminated polybutadiene (HTPB) matrix in AP/HTPB propellants were studied via an all-atom molecular dynamics simulation. The results of the simulation showed the effects of the microscopic cross-linked structure of the matrix, stretching rate during uniaxial stretching, and contact area between the filler and matrix on the mechanical properties, such as the stress and strain of the composite solid propellant. Among the aforementioned factors, the stretching rate considerably affects the mechanical properties of the solid propellant, and the maximum stress of the solid propellant proportionally increases with the stretching rate. When defects were introduced on the surface of the AP filler, the contact area between the filler and matrix affected the strain type of the matrix molecules. Owing to the interaction between the molecules and atoms, the strain behaviour of the matrix molecule changed with the change in its microscopic cross-linked structure during uniaxial stretching. Molecular dynamics simulations were used to explore the characteristics at the AP–HTPB interface in AP/HTPB propellants. The aforementioned simulation results further revealed the interfacial interaction mechanism of the AP–HTPB matrix and provided a theoretical basis for the design of high-performance propellants.

The self-organization of five model side-chain decorated polyaromatic asphaltene molecules with or without toluene solvent was investigated by means of molecular dynamic (MD) simulations.

Understanding the noncovalent (weak) interactions between asphaltene molecules is crucial to further comprehending the viscosity and aggregation behavior of asphaltenes. In the past, intermolecular interactions were characterized indirectly by calculating the radial distribution function and the numerical distribution of distances/angles between atoms, which are far less intuitive than the average reduced density gradient (aRDG) method. This study selected three representative asphaltene molecules (AsphalteneO, AsphalteneT, and AsphalteneY) to investigate the relationship between viscosity and weak intermolecular interactions. Firstly, a non-equilibrium molecular dynamics (NEMD) simulation was employed to calculate the shear viscosities of these molecules and analyze their aggregation behaviors. In addition, the types of weak intermolecular interactions of asphaltene were visualized by the aRDG method. Finally, the stability of the weak intermolecular interactions was analyzed by the thermal fluctuation index (TFI). The results indicate that AsphalteneY has the highest viscosity. The aggregation behavior of AsphalteneO is mainly face–face stacking, while AsphalteneT and AsphalteneY associate mainly via offset stacking and T-shaped stacking. According to the aRDG analysis, the weak interactions between AshalteneT molecules are similar to those between AshalteneO molecules, mainly due to van der Waals interactions and steric hindrance effects. At the same time, there is a strong attraction between AsphalteneY molecules. Additionally, the results of the TFI analysis show that the weak intermolecular interactions of the three types of asphaltene molecules are relatively stable and not significantly affected by thermal motion. Our results provide a new method for better understanding asphaltene molecules’ viscosity and aggregation behavior.