Preparation of Ni-SiC Nanocoatings and Prediction of Their Characteristics by Artificial Neural Networks

The Ni-SiC nanocoatings have been produced on a substrate comprising 45 steels following the jet pulse electrodeposition approach. X-ray photoelectron spectroscopy, transmission electron microscopy, scanning electron microscopy, Vickers test, and electrochemical techniques were used to assess the morphology, microstructure, corrosion characteristics, and microhardness of these prepared Ni-SiC nanocoatings. A backward propagation (BP) artificial neural network was utilized for predicting the corrosion mass loss and microhardness parameters of the deposited Ni-SiC nanocoatings and subsequently compared with experimental data. When the plating parameters were set at a jet rate of 5.5 m/s, a SiC concentration of 5 g/L, and a pulse density of current of 4 A/dm2, the Ni-SiC nanocoating processed the highest microhardness value (~863.4 HV). The mean grain diameters of the nanoparticles of SiC and Ni grains were 29.6 nm and 54.3 nm, respectively. In addition, when a 5.5 m/s jet rate, 5 g/L SiC concentration, and a current density of 4 A/dm2 were used, the Ni-SiC nanocoating exhibited the least corrosion density of current equivalent to 5.3×10−5 A/cm2. These results showed that the Ni-SiC nanocoating had the highest value of impedance and best anti-corrosion potential. Moreover, the highest mean square errors (MEs) of microhardness and corrosion mass loss of the nanocoatings predicted by the BP model were 3.1 and 3.3%, respectively.