Development of a Model for Prediction and Optimization of Hardness of Electrodeposited Cu/SiC Composite Using RSM and ANN-PSO

Abstract

Abstract Nanocomposite coating has a substantial impact on thermomechanical characteristics. There are several methods for developing nanocomposite coatings; however, electrodeposition is one of the additive manufacturing approaches that has been shown to be suitable due to its adjustable parameters, cost-effective and ease of process. Copper composite coatings have been extensively utilized in the aerospace and automotive industries due to their superior mechanical, electrical, and thermal properties. In the proposed study, a “In-house pulse electrodeposition setup” was developed and used to effectively deposit Cu/SiC nanocomposite coatings from an aqueous sulphate solution. Scanning electron microscopy (SEM) and x-ray diffractometer (XRD) are used to examine the surface morphology, which confirmed the successful co-deposition. The microhardness of the Cu/SiC composite coatings and effect of process parameters is predicted using response surface methodology (RSM). Artificial neural network-particle swarm optimization (ANN-PSO) is used to train, test, and validate experimental data and to investigate the correlation between electrodeposition process parameters and their effects on microhardness. The process was operated at: pulse frequency (10 Hz-100 Hz), duty cycle (20–80%), bath agitation (200–500 rpm), and SiC concentration (1–5 g/L). The determination coefficient (R2), root mean square error (RMSE), mean bias error (MBE), and mean absolute percentage error (MAPE) is used to validate several ANN-PSO models. The optimal morphology with the highest microhardness is produced under the following conditions: 10 Hz pulse frequency, 50% duty cycle, 350 rpm bath agitation, and 5 g/L SiC concentration. The findings showed that the suggested model is an appropriate, flexible, and reliable way for estimating the microhardness of Cu/SiC nanocomposite coatings.