Thermoplastic materials, especially polyethereketone (PEEK) and its composites are widely used in industrial area. PEEK material is extensively used to produce automobiles, machines, medical applications, robots, aeronautical and biomechanical parts due to excellent properties such as low density, high strength and durability, good toughness, better environmental resistance, high temperature service performances. The short fibers are added to the matrix, in order to improve tribological and mechanical properties of unreinforced PEEK. In this study, an Adaptive Neural Fuzzy Inference System (ANFIS) is applied to predict the cutting force during the turning operation of unreinforced and reinforced PEEK with 30 % v/v of carbon fibers (PEEK CF30) and 30 % v/v of glass fibers (PEEK GF30) machining. The cutting speed, feed rate, type of material, and cutting tools are defined as input parameters and cutting force is defined as output of the system. The experimental results and test results predicted by using ANFIS model are compared in terms of the coefficient of determination (R2), root mean square (RMS) and the mean absolute percentage error (MAPE). The test results show that ANFIS model provides good prediction accuracy and it is convenient to predict cutting force in the turning operation of PEEK composites reinforced with carbon and glass fibers.
Anahtar Kelimeler: Polyetheretherketone (PEEK), Adaptive Neural Fuzzy Inference System (ANFIS), Turning, Composite.
|
