Empirical models of mechanical and electrical drilling power of mild steel

Abstract Two empirical models for electrical and mechanical power prediction were developed for drilling mild steel. Multiple linear regression analysis was implemented using a three-level three-variable (33) factorial design experiment for mild steel. The drill diameter, feed rate and cutting speed are the independent variables. Both models were in close agreement with the experimental results. The correlation coefficient for both models was 91 per cent. These models can be effectively utilized in conjunction with a proposed on-line monitoring system to predict the tool condition. Tool replacement decision making can be implemented using either model. However, the model for electrical power was a better indicator of tool wear.