<p>The intense fluid flow within the molten pool significantly influences the solidification process, resulting in notable effects on solute segregation, microstructural features, solidification rate, and grain boundary morphology. In this study, a validated cross-scale finite element-lattice Boltzmann-cellular automaton model was employed to examine the development of dendrites in the flowing molten pool of the In718 alloy. The numerical calculation of the thermal cycle of the molten pool was performed using the macroscopic finite element model based on the thermal conduction mechanism. The flow characteristics of molten metal particles at high temperatures and their convection-diffusion transport behavior were determined using the lattice Boltzmann model and the finite difference model, respectively. The microscopic cellular automata models were used to reproduce the whole process of dendrite evolution. The results demonstrated that the existence of flow reduces the concentration peak value and has a positive impact on the achievement of a homogeneous distribution of the Nb solute. In addition, the mechanism of adjusting the space between primary dendrite arms in the presence of fluid flow involves speeding up the elimination of dendrites and reducing the solute concentration gradient at the tips of dendrites. The heat source model and boundary conditions were validated through the analysis of the weld seam profile. The simulated results of the competitive growth within the molten pool demonstrate significant agreement with the experimental observations.</p>