The autonomous swimming of fish in a complex flow environment is a nonlinear and intricate system, which is the focus and challenge in various fields. This study proposed a novel simulation framework for artificial intelligence fish. It employed a high-precision immersed boundary-lattice Boltzmann coupling scheme to simulate the interactions between fish and flow in real time, and utilized the soft actor-critic (SAC) deep reinforcement learning algorithm for fish brain decision-making module, which was further divided into a vision-based directional navigation and a lateral line-based flow perception modules, each matched with its corresponding macro-action space. The flow features were extracted using a deep neural network based on a multi-classification algorithm from the data perceived by the lateral line and were linked to the fish actions. The predation swimming and the various Kármán gait swimming were explored in terms of training, simulation, and generalization. Numerical results demonstrated significant advantages in the convergence speed and training efficiency of the SAC algorithm. Owing to the closed-loop “perceive-feedback-memory” mode, intelligent fish can respond in real-time to changes in flow fields based on reward-driven requirements and experience, and the accumulated experience can be directly utilized in other flow fields, and its adaptability, model training efficiency, and generalization were substantially improved.

Flood discharge and sediment transport are closely linked to channel resistance in steep mountain streams. Previous research has mainly focused on the resistance of fixed-bed channels with steep gradients and mobile-bed channels in alluvial rivers. The present study performs an experiment and establishes a calculation method for the fixed-bed resistance of mountain channels. The basic expression of the mobile-bed resistance of steep mountain channels is derived by determining the controlling factors of the bed load movement on the riverbed resistance. The proposed formula can accurately predict the variation of the bed load resistance. The results of the present research improve the understanding of fluid dynamics and sediment transport in steep mountain channels.

The study of fish swimming behavior and locomotion mechanisms holds substantial scientific and engineering significance. With the rapid progression of artificial intelligence, the integration of artificial intelligence with high-precision numerical simulation methods presents a novel and highly efficient tool for investigating fish behavior. In this paper, we proposed a fish perception model that more closely reflects natural logic. By introducing the individual vision and partially visibility model, a physics-based visual system that mirrored the sensory capabilities of live fish was developed. Furthermore, through the construction of a flow vision using conventional neural networks, we enhanced the intelligent fish's ability to detect unsteady hydrodynamic parameters via numerical lateral line. The validity of the new model was demonstrated through experiments, which the intelligent fish hunts complex moving targets in unsteady flow. Finally, we applied the model to study the refuge/predation behaviors of coral reef fish under varying unsteady flow pressures. The results indicated that swimming capabilities significantly impact fish survival strategies in high flow velocity, highly unsteady hydrodynamic environments, shaping distinct evolutionary decision-making traits. These insights may help to understand the niche competition mechanisms of fish in different flow conditions.

Fish schooling, an adaptive strategy shaped by long-term evolution, is closely associated with core biological needs such as survival and reproduction. This study, having adopted a numerical simulation framework built by combining the deep reinforcement learning algorithm with the immersed boundary-lattice Boltzmann fluid–solid interaction simulation method, focused on investigating the effects of modal transitions on energy efficiency and array dynamics during large-scale turning of fish schools under low hydrodynamic pressure environments. Initially, strategic control was applied to two fish utilizing body–caudal fin propulsion, aiming to explore their schooling modes and the intrinsic energy-saving mechanisms inherent to each mode. Subsequently, the school size was expanded non-proportionally, and the influence of modal transitions on energy consumption was analyzed through a leader-guided turning strategy. Results indicated that under low hydrodynamic pressure, carangiform fish exhibited three modes in their modal composition, namely, the vortex action mode, the lateral force action mode, and the combined action mode. Additionally, with an increased number of schooling individuals, the modal transitions of carangiform fish demonstrated greater arbitrariness and flexibility. Notably, exclusive reliance on hydrodynamic information perception via the lateral line could enhance efficiency in small-scale schools. However, as the school expanded, the complex flow field limited the ability of this approach to significantly improve overall schooling performance; meanwhile, in low hydrodynamic pressure environments, the collective movement of fish schools does not depend solely on hydrodynamic information. These findings provide theoretical guidance and strategic foundations for the design of clustered bionic underwater intelligent robots with different undulatory propulsion modes.

The design and optimization of fishways and other fish-passage facilities are one of the critical issues in hydraulic engineering. Traditional methods using physical experiments for fishway design and optimization face challenges such as the uncontrollability of fish behavior, limited non-intrusive measurement techniques, and nonlinear scale effects. Numerical simulation methods can provide the hydraulic performance information of fishways by analyzing flow velocity, turbulence energy, and flow patterns, but fail to account for the active responses of fish to the hydrodynamic characteristics of the environment. In this study, a research paradigm of using embodied intelligent fish to optimize eco-hydraulic facilities was attempted. The core of this research paradigm is a simulation platform for intelligent fish based on deep reinforcement learning (DRL) and the immersed boundary–lattice Boltzmann (IB-LB) coupling algorithm. Based on this platform, by endowing fish with biological tendencies and biometric features related to perception/feedback/decision-making at the individual scale, embodied intelligent fish can achieve multimodal perception and autonomous decision-making in complex fishway digital flow fields based on the potential habits of live fish. Using the proposed framework, conduct end-to-end deep reinforcement learning training on intelligent fish and then deploy the trained embodied intelligent fish in a virtual vertical-slot fishway for ecological hydrodynamic experiments. Comparative analyses were conducted on the fish-passage performance of three different fishway geometric shapes. The results demonstrated that the new research paradigm evaluates the structural characteristics of fishways through the adaptive response behavior of fish to the hydrodynamic environment, which can provide quantitative guidance for fishway design from the terms of passage path, passage time, energy consumption, etc. This study belonged to an individual-scale digital twin and was an innovative attempt to utilize virtual intelligent lifeforms at the individual scale to conduct simulation-based experimental research.

In biomimetic underwater systems, high-efficiency and low-power propulsion remains a core challenge. Mimicking the characteristics of fish caudal fins and exploring highly biomimetic muscle-driven approaches is regarded as one of the key strategies to address this issue.This study combines the immersed boundary-lattice Boltzmann (IB-LB) method with deep reinforcement learning (DRL) to investigate the interactive effects of caudal fin structural stiffness and active muscle control on propulsive performance and energy consumption.By constructing a virtual fish model with a closed-loop ``perception--decision--action'' feature, the agent can autonomously learn to output tail torque based on environmental feedback, thereby regulating the deflection behavior of the caudal fin. The research evaluates the differences in dynamic responses between rigid and flexible caudal fin configurations under both passive states and active control intervention. The results indicate that rigid caudal fins exhibit significant phase lag and increased energy consumption without control; however, driven by DRL strategies, they can achieve phase compensation and a substantial improvement in propulsive performance. In contrast, flexible caudal fins, relying on stronger passive adaptability, can achieve superior propulsive efficiency in the uncontrolled state, while their speed and energy consumption can be further optimized with the introduction of active regulation.To realize dynamic trade-offs between speed and energy consumption, this study develops a task-sensitive multi-objective dynamic reward function, enabling the agent to switch between ``high-speed propulsion'' and ``high-efficiency energy-saving'' strategies according to requirements. This research not only reveals the synergistic relationship between structural compliance and active control but also demonstrates the potential of deep reinforcement learning in exploring optimal control strategies without prior knowledge. It provides a new research path and theoretical support for the intelligent regulation of bionic fish caudal fins and the design of flexible underwater robots.