The computational fluid dynamics coupled with the population balance model (CFD-PBM) based on the monodisperse model is widely used for the design of aerosol synthesis nanoparticles. Due to the monodisperse assumption, it is insufficient to accurately describe the particle polydispersity and morphology during particle evolution in the reactor. To solve these limitations, we employed the mesoscale Langevin dynamics (LD) simulation for particle coagulation and combined it with CFD-PBM to investigate the particle evolution in a tubular aerosol reactor. Particle coagulation is simulated by LD simulations, which is split into two separate stages including the pure coalescence stage at high temperatures and the subsequent pure agglomeration stage at low temperatures. The evolution function of morphology (Df) and the in-real-time coagulation rate (β) are derived from LD and then applied for the PBM calculation. CFD-PBM model estimates the primary and agglomerate nanoparticle diameters in a tubular aerosol reactor within 90% and 75% accuracy, respectively, compared to the experimental measurements in the literature. By this LD-combined CFD-PBM model, the evolution of particle morphology along the axis can also be evaluated.