Theoretical and experimental investigation on modeling of surface roughness for in-situ laser assisted diamond cutting of fused silica

• A surface roughness prediction model for in-situ LADC was developed. • Grey Wolf Optimization algorithm was used to train the fitted coefficients. • The influence of processing parameters on surface quality was evaluated. • The maximum relative error of the model is below 8%. In-situ laser assisted diamond cutting has achieved high-quality machining of fused silica. However, the surface quality is influenced by multiple factors, making it challenging to predict the surface roughness accurately. In this research, a novel surface roughness prediction model for in-situ laser assisted diamond cutting of fused silica was established by integrating theoretical models with experimental data. The model considered pivotal elements such as brittle fracture and variations in temperature-dependent mechanical properties. Due to the limitation of theoretical calculation, the impacts of feed rate on plastic side flow and cracks on surface roughness were determined by establishing fitting equations. The fitted coefficients in the model were trained by the Grey Wolf Optimization algorithm based on the experimental data. Results demonstrate that the established model provides a comprehensive analysis of the underlying factors contributing to roughness formation. It can effectively predict the surface roughness of fused silica during in-situ laser assisted diamond cutting under different machining parameters accurately through a small quantity of experimental data, with a maximum relative error below 8%.