Tool Wear Prediction Using GRNN in Sustainable Machining of AISI H13 Steel

Cutting fluids are often used to reduce friction and temperature at the cutting zone. However, excessive use of cutting fluids leads to environmental impacts, increased machining costs, health risks for operators, and challenges in fluid disposal due to stringent environmental regulations. This has driven a shift towards sustainable machining practices to achieve both environmental and economic benefits. Consequently, researchers are focused on dry machining and textured tool machining, which are integral to sustainable machining practices. Integration of artificial intelligence practice in the sustainable manufacturing is essential and it requires proper decision making in the process and heuristic approach. Heuristic approaches are further employed to optimize machining parameters, improving productivity and reducing manufacturing costs. These approaches are used to prioritize efficiency and practicality over strict optimality. In the final phase, Generalized Regression Neural Networks (GRNN) are used to develop predictive models for machining performance responses. The GRNN model is trained using 70% of the experimental data and is tested and validated with the remaining 30%. The developed GRNN model can predict all machining performance responses with a ±3<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">%</sup> error, effectively establishing the relationship between machining parameters and performance responses.