ISSN 2360-7955
Abstract
Machining parameter optimization is a major issue in modern manufacturing, owing to its direct impact on efficiency, stability, and product quality. Predictive modeling and optimization of shearing power in turning operations under constraints was the aim of this study, where a comparative evaluation of the usefulness of Response Surface Methodology (RSM) and Artificial Neural Networks (ANN) methods is provided. Three major factors affecting shearing power and machining behaviour are the concern of this study: depth of cut, cutting speed, and feed rate. Using well-planned and executed experiments, the data necessary to construct predictive models was gathered. Using Response Surface Methodology (RSM), equations that describe the intricate interaction of these variables were obtained. Artificial Neural Networks (ANN) was a very effective means of modeling the intricate patterns in the data, giving a more detailed image of the machining process. When the two techniques were compared against each other, Artificial Neural Networks (ANN) emerged as the better predictive model, delivering spot-on predictions of shearing power with impressively low error margins and robust regression performance. While both models showed statistical effectiveness, ANN's capability for capturing the intricate relationships between machining variables was unparalleled. This implies that ANN could revolutionize process optimization, unlocking new levels of machining efficiency. By tapping into the potential of data-driven decision-making, this work fosters smart manufacturing, enabling machine processes that are more agile, responsive, and self-tuning than ever before.
Keywords: Shearing Power, Turning, Machining, RSM, ANN