Purpose: In plasma spraying, the coating properties such as porosity, hardness, strength, etc. are directly determined by particle behaviors, i.e. the temperature and velocity. Therefore, it is necessary and meaningful to predict the particle behaviors under a certain combination of process parameters before the spraying process is executed. Design/methodology/approach: In this study, SVM (Support Vector Machines) is applied to the prediction of in-flight particle temperature and velocity in plasma spraying by argon flow rate, hydrogen flow rate and electric current. The influences of the three parameters on particle temperature and velocity are also investigated. Findings: In the leave-one-out cross validation on an orthogonal experiment with 9 sets of parameters, the maximum relative errors of prediction for particle temperature and velocity are 0.68% and 1.42% respectively. The prediction results reveal that the most influential parameter for particle temperature is hydrogen flow rate, and argon flow rate exerts the greatest influences on particle velocity. Research limitations/implications: Future work should focus on the modeling of the whole spraying proces with all the spraying parameters. Practical implications: It will be helpful to the prediction and controll of particle behaviors in plasma spraying. Originality/value: First application of SVM to modeling the in-flight particle behaviors in plasma spraying.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.