This paper evaluates the ability of a CFD-QMOM modeling approach to predict the dynamics of the particle size distribution (PSD) of airborne non-spherical nanoparticles. The experimental case of a small chamber ventilated by a steady turbulent airflow at moderate Reynolds number is considered. In this configuration, the aerosol dynamics is essentially driven by convection and turbulent-Brownian aggregation. Numerical results are compared to available experimental data: space-resolved PSD measurements for the particulate phase and mean airflow velocity, turbulence kinetic energy and mean age of air profiles for the gas phase. Given the experimental uncertainties, a good agreement between experimental and simulation results is found. The use of two types of aerosols, with identical initial concentration but different compositions, sizes and morphologies highlights the influence of these parameters on aggregation kinetics. The highest particle number reduction due to aggregation is achieved for the smallest and least compact nanoparticles.