This paper reported the mathematical modeling and numerical simulation of natural convection flow of Cu/water nanofluid in a square enclosure using the Lattice Boltzmann Method (LBM). The cavity is heated from below by heat source and cooled by the top wall. The vertical walls are adiabatic. After validating the numerical code against numerical and experimental data, simulations were performed for different Rayleigh numbers (104 - 0.5 × 107), nanoparticles volume fractions (0−8%) and cavity inclination angle (0° - 90°).The effects of the studied parameters on the streamlines, isotherms distributions within the enclosure and on the local and average Nusselt numbers are investigated. It was found that heat transfer and fluid flow structure depend closely to the nanoparticle concentration. Results show differences in stream separation between a base fluid and the nanofluid. Also, adding small nanoparticles fractions, less than 6%, to the base fluid enhances the heat transfer for higher Rayleigh numbers and cavity inclination angle less than 30°. It is concluded that the optimal dilute suspension of copper nanoparticles can be applied as a passive way to enhance heat transfer in natural convection engineering applications.