Predictive Modeling of Surface Roughness and Cutting Temperature Using Response Surface Methodology and Artificial Neural Network in Hard Turning of AISI 52100 Steel with Minimal Cutting Fluid Application

Hard turning is a precision machining process used in the manufacturing industry for the finishing of hardened alloy steel, which is known for its high hardness and wear resistance. In this work, an experimental investigation was conducted to predict surface roughness and cutting temperature during the hard turning of AISI 52100 steel using the minimal cutting fluid application (MCFA). The MCFA is a sustainable high-velocity pulsed jet technique that has emerged as an eco-friendly approach for reducing the environmental impact and improving surface integrity in machining processes. The influence of key machining parameters, such as cutting speed, feed rate, and depth of cut, on the performance indicators was modeled using the response surface methodology (RSM) and the artificial neural network (ANN). The RSM was employed for a structured, statistical analysis, while an ANN provided a data-driven approach for capturing complex non-linear relationships. Various network architectures were established and evaluated with a fixed number of cycles. Results showed that the ANN exhibited superior accuracy in predicting both responses. In comparison to the QR model, the ANN exhibited the lowest average error rate in accurately predicting the response. This was further validated through experimental trials, demonstrating that the ANN consistently outperformed the RSM across different parameter settings. Additionally, the use of the MCFA contributed to sustainable manufacturing by minimizing the use of cutting fluids while maintaining machining quality.