LAUSR

The present study aims to implement a non-destructive approach to determine the tensile properties of magnesium-based nanocomposites reinforced with ZrO2 and TiO2 nanoparticles. Micron-sized magnesium particles were blended with 0, 1.5, 3, and 5 volume percentage of ZrO2 and TiO2 nanoparticles and hot-pressed at 450 °C under the pressure of 600 MPa. Next, the spherical indentation test was performed on the produced composites to obtain the load–penetration curves. A finite element model of the indentation test was then developed using the Hollomon material model with randomly chosen materials constants. At the next stage, load–penetration curves were obtained for each composite using simulations. A Levenberg–Marquardt neural network was then trained and utilized to find the correct material constants by minimizing the differences between the experimental and simulated load–penetration curves. The results indicated that there is a linear relationship between the tensile strength and content of the reinforcement phase, while it is inversely proportional to the size of the reinforcing particles. Magnesium composites reinforced with 5 volume percentage of ZrO2 and TiO2 nanoparticles showed tensile strengths 2.5 and 2.1 times greater than that of unreinforced magnesium, respectively. It was shown that the proposed method is able to calculate the tensile properties of magnesium-based composites in an accurate and inexpensive manner.