At European XFEL, computational techniques such as finite element analysis (FEA) and computational fluid dynamics (CFD) are widely applied in various scientific and engineering fields, such as damage simulation due to heat load, bleaching effect study of gas attenuator, optimization of fluid cooling system for detectors and characterization of liquid sheet jets for sample delivery system. Without being constrained by experimental conditions, the multiphysics and multiscale models in simulation could virtually replicate the interaction process of XFEL beam with different materials, taking into consideration heat transfer, structural deformation and phase transition. In this contribution, to gain comprehensive insights into the fluid behaviors of the detector cooling system, as well as the performance of reduced order modelling solvers, parametric studies are conducted using CFD simulation code. Furthermore, a realistic simulation requires a secured process of Verification and Validation (V&V) of the computational model. Specific guides and standards need to be followed to ensure the credibility and accuracy of the simulation results. Besides following the FAIR principle (Findable, Accessible, Interoperable, and Reusable), a smart simulation data management system using machine learning algorithm is under construction. Moreover, the large amount of data from the simulations in the past can be utilized to train the machine learning model, which can be used for simulation results prediction without running further simulations. Further AI and machine learning tools are going to be employed to set up generative design workflow and digital twin scheme for the beamline components, serving as a new safety constraint for monitoring and optimizing of the facility operation.