Prediction and Optimization of Cutting Forces required to Minimize Tool Breakage during Dry Drilling Operations

Abstract

Drilling is one of the most common and rudimental machining methods in the manufacturing industries for removal of unwanted material from the workpiece. How the cutting instrument and workpiece interact results in a mechanical force that causes the formation of chips during penetration, and these chips are evacuated through the flute created on the body of the drill tool. The interplay of forces at the drill point generates high temperatures, needed for the physical and chemical processes that weaken tools and lead to breakage. Optimal experimental designs are very important to obtain accurate optimization of engineering processes, an expert method was used to design the experimental layout and utilizing the Design Expert software, an experimental matrix which developed the parameters design of twenty experimental runs. The present study uses Artificial Neural Networks (ANN) and Response Surface Methodology (RSM) to forecast and optimize cutting forces during dry drilling operations. In developing the model, a dataset that included several factors such as depth of cut, feed rate, and cutting speed was used. The RSM model demonstrated a significant correlation between input parameters and cutting forces, as evidenced by its high coefficient of determination (R²) of 0.9493. On the other hand, the ANN model, which was trained using 70% of the data and validated using 15% of the data, showed a little lower R² value of 0.81434, but it was still able to make accurate predictions. Cutting forces were well predicted by both models, with RSM exhibiting a somewhat better performance in terms of accuracy. The results indicate that both RSM and ANN can be useful instruments for dry drilling cutting force optimization, offering insights for increased productivity and efficiency in machining operations.