The interaction of waves with structural barriers, such as dam breaking, plays a critical role in flood defense and tsunami disasters. In this work, we explore the dynamic changes in wave surfaces impacting various structural shapes—circle, triangle, and square—using deep learning techniques. We introduce the “DamFormer,” a novel transformer-based model designed to learn and simulate these complex interactions. Additionally, we conducted zero-shot experiments to evaluate the model's ability to generalize across different domains. This approach enhances our understanding of fluid dynamics in marine engineering and opens new avenues for advancing computational methods in the field. Our findings demonstrate the potential of deep learning models like the DamFormer to provide significant insights and predictive capabilities in ocean engineering and fluid mechanics.

The rational placement of pollutant monitoring sensors has long been a prominent research focus in ocean environment science. Our method integrates an incremental Proper Orthogonal Decomposition with a mobility-constrained sensor selection strategy, enabling efficient monitoring and dynamic adaptation to spatio-temporal field changes. At each time step, the position of the sensors is updated based on the incoming measurements to minimize the reconstruction error while adhering to movement constraints. This online approach considers the need for mobility distance, making it suitable for long-term deployments in resource-limited scenarios. The proposed framework is validated in three scenarios: a linear advection–diffusion system with multiple moving pollution sources, the distribution of particulate matter with an aerodynamic diameter smaller than 2.5 μm (PM2.5) across the United States, and scalar transport in flows past side-by-side cylinder arrays in the ocean. The results demonstrate that the method achieves high reconstruction accuracy with significantly fewer sensors. This study conducts a comparative analysis of three typical mobility constraints and their respective effects on reconstruction accuracy. In addition, the proposed localized sensor mobility strategy effectively tracks evolving plume structures and maintains a low approximation error, providing a generalizable solution for sparse monitoring of the marine environment.

Secure and user‐friendly anti‐counterfeiting technologies are critical for safeguarding products in modern society. However, achieving a synergy between high‐level encryption and efficient readability remains a significant challenge. Here, we present an all‐in‐one multimodal optical anti‐counterfeiting strategy based on fluorescent hierarchical structures through a sequential imprinting process. The resulting structures exhibit an intact macroscopic contour while retaining randomly distributed microscopic defects. Macroscopically, hidden photoluminescent (PL) patterns revealed by UV irradiation, together with angle‐dependent nano‐ and micro‐structural color patterns observed under daylight, enable direct visual decoding. Microscopically, random fluorescent inks and structural defects generate two independent physical unclonable functions under UV and daylight, whose encryption capacity is equivalent to the product of two independent keys, providing ultrahigh security. Moreover, the hierarchical structure in multimodal optical anti‐counterfeiting enhances PL emission, provides robust friction resistance, and improves hydrophobicity, collectively enabling reliable performance in practical applications. Using a painting as a platform, we demonstrate a feasible authentication strategy for next‐generation anti‐counterfeiting.

The intelligent decision-making systems of Autonomous Underwater Vehicle (AUV) are undergoing a significant transformation, shifting from traditional control theories to data-driven paradigms. Deep learning (DL) serves as the primary driving force behind this evolution; however, its application in complex and unstructured underwater en-vironments continues to present unique challenges. To systematically analyze the de-velopment, current obstacles, and future directions of DL-enhanced AUV deci-sion-making systems, this paper proposes an innovative "four-module" decomposition framework, which are information processing module, information understanding module, information judgment module and output module. This framework enables a structured review of the progression of deep learning technologies across each stage of the AUV decision-making information flow. To further bridge the gap between theo-retical advancements and practical implementation, we introduce a task complexity–environment uncertainty four-quadrant analytical matrix, offering strategic guidance for selecting appropriate DL architectures across diverse operational scenarios. Addi-tionally, this work identifies key challenges in the field as well as anticipates future developments to solve these challenges.This paper aims to provide researchers and engineers with a comprehensive and strategic overview to support the design and op-timization of next-generation AUV decision-making architectures.

1. Flow past a cluster of cylinders (Happel, 1959; Heideman & Sarpkaya, 1985; Walker & Taylor, 2005) is a ubiquitous phenomenon in ocean and environment engineering, for example jacket platforms (T...