Investigative study on machinability aspects of unreinforced and reinforced PEEK composite machining using ANN model

Polyetheretherketone (PEEK) composite material is widely used in structural components due to high specific strength and temperature properties. In order to enhance mechanical and tribological properties, short fibers are added to unreinforced thermoplastics. Unreinforced and reinforced PEEK composites are generally manufactured by extrusion and hence need additional machining operations. This paper presents application of artificial neural network (ANN) model to study the machinability aspects of unreinforced polyetheretherketone (PEEK), reinforced polyetheretherketone with 30% of carbon fibers (PEEK CF 30) and 30% of glass fibers (PEEK GF 30) machining. A multilayer feed forward ANN has been employed to study the effect of parameters such as tool material, work material, cutting speed and feed rate on two aspects of machinability, namely, power and specific cutting pressure. The input-output patterns required for training are obtained from the experiments planned through full factorial design. The analysis reveals that minimum power results from a combination of lower values of cutting speed and feed rate for all work-tool combinations. However, higher values of feed rate are required to achieve minimum specific cutting pressure. The investigation results also show that, K10 tool provides better machinability for PEEK and PEEK CF 30 materials, while PCD tool is preferred for PEEK GF 30 material.