Modeling adhesive wear resistance of Al-Si-Mg-/SiC_p PM compacts fabricated by hot pressing process, by means of ANN

Abstract

In this study, modeling adhesive wear resistance of Al-Si-Mg/SiC_p MMC compacts were performed by ANN, using a back-propagation neural network that uses gradient descent learning algorithm. Powder compacts were fabricated by PM hot pressing process with 5–10–20% SiC_p fractions and contents of specimens (N1, N2, N3 andN4) were given in Table 1. The wear tests were carried out under 10, 20 and 30 N variable loads, while disk rotation speed 90 rpm kept unchanged. Adhesive wear looses were measured and recorded for 250, 500, 1,000 and 1,500 m distances. Microstructure examination at wear surface was investigated by optical microscopy and EDS for metallographic evaluations. In neural networks training module, SiC_p reinforcement fractions (wt), loads and wear distances (m) were used as input, lost mass (g) of specimens were recorded 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 ANN was found successful in modeling of adhesive wear behavior and lost mass values of Al/SiC_p PM compacts.

Table 1 Mixture rations and density of specimens