Modelling of abrasive wear resistance by means of artificial neural networks of Al-Sicp composıtes produced by cold pressıng method

Abstract

In this study, modelling of abrasive wear resistance by means of artificial neural networks of Al-SiCp composites produced by cold pressing method were obtained using a back-propagation neural network that uses gradient descent learning algorithm. SiC particles with a 32 ?m mean diameter were added to the matrix at 5, 10, 15 (wt) % fractions and powders were mixed with 99 % Al. MMC's were fabricated by powder mixing and cold pressing under 400 MPa load and sintering at 400°C. The wear tests were performed in loads of 2 and 8 N, the abrasive paper 120 and 400 mesh, the wear distance of 20, 40 and 60 m by abrasive test apparatus and the wear losses were calculated. Microstructure examination at wear surface were investigated by optical microscopy, SEM and EDS. Specimens were tested for optical microscopy, SEM, EDS and metallographic evaluations. After the completion of experimental process and relevant test, to prepare the training and test (checking) set of the network, results were recorded in a file on a computer. In neural networks training module, different SiC reinforcement fractions (wt), different wear distances, different feasible loads and different abrasive paper were used as input, mass loss of abrasive wear specimens at surface were used as outputs. Then, the neural network was trained using the prepared training set (also known as learning set). At the end of the training process, the test data were used to check the system accuracy. As a result the neural network was found successful in the prediction of modelling of mass loss values of Al/SiCp metal matrix composite materials processed with abrasive wear method and behavior.