Estimation of Forces in Longitudinal and Multi-Pass Milling of GFRP and ABACA Composites Utilizing Machine LearningTechniques

In this study, workpieces made of glass fibre reinforced polymer and ABACA composite materials were machined by longitudinal and multi-pass milling methods with three different feed and speed parameters. The effect of milling methods on the finish surface was discussed through microscope images. The effect of machining parameters on the force was analysed by measuring the force values changing during machining. For this purpose, advanced prediction modelling was performed with the Random forest machine learning method. The effect of machining methods and parameters on cutting force is predicted with an average success rate of 86%. The primary contribution of this research lies in providing a comparative assessment of natural and polymer composites under distinct milling conditions and introducing a data-driven approach for accurate cutting-force prediction. The findings provide new insights into the processing of natural fibre composites as an alternative to polymer composites in milling.