In recent years，simulation systems based on computer graphics have developed rapidly，drawing particular attention to water surface wave effects in ship-handling simulators. As these simulators are increasingly employed in inland pilot training and in assessing the navigational safety of water-related structures，it has been discovered that inland river flow features （i.e.，water flow tendencies） serve as crucial visual cues for safe vessel operation. Nevertheless，current simulators fail to offer realistic visual effects that capture inland river flow characteristics. To address this gap，the key technologies，fundamental principles，developmental progress，and resulting simulation effects of physics-based fluid simulation methods were reviewed，including Eulerian grid approaches，Lagrangian particle methods，hybrid Euler-Lagrange schemes，and the lattice Boltzmann method with its extensions. Particular emphasis was placed on analyzing their computational efficiency under graphics processing unit（GPU） parallel acceleration，equation simplification，and multi-technique coupling，along with a comparison of real-time simulation performance and a summary of their respective features and applicable scenarios，aiming to identify a suitable approach for river modeling in inland ship-handling simulator visualization systems. Finally，future research directions were discussed，leading to three main conclusions： for applications demanding high accuracy，offline simulation is preferable. The smoothed particle hydrodynamics-shallow water equations（SPH-SWE） method enables real-time and highly realistic large-scale fluid-solid coupling. Integrating machine learning techniques holds promising potential for achieving high-accuracy real-time simulations.