Implementation of Artificial Neural Networks for Forecasting the HVOF Spray Process and HVOF Sprayed Coatings

Abstract

In the high velocity oxygen fuel (HVOF) spray process, coating properties are sensitive to the characteristics of in-flight particles, which are mainly determined by process parameters. Obtaining a comprehensive multi-physical model or analysis of the HVOF process remains challenging because of the complex chemical and thermodynamic reactions that occur during the deposition procedure. This study proposes to develop a robust methodology via the artificial neural networks (ANN) to solve this problem for the HVOF sprayed coatings under different operating parameters. Two ANN models were developed and implemented to predict coating’s performances (microhardness, porosity and wear rate) and to analyze the influence of operating parameters (stand-off distance, oxygen flow rate, and fuel flow rate) while considering the intermediate variables (temperature and velocity of in-flight particles). A detailed procedure for creating and optimizing these two ANN models is presented in this work, which encodes the implicitly physical phenomena governing the HVOF process. Results show that the developed implicit models can satisfy the prediction requirements and clarify the interrelationships between the spraying conditions, behaviors of in-flight particles, and the final coating performances, resulting in providing better control of the HVOF sprayed coatings.