Reverse analysis via efficient artificial neural networks based on simulated Berkovich indentation considering effects of friction

Instrumented indentation test has become a popular method for characterization of materials of small volume such as those constitute the micro-electro-mechanical devices, micro-electronic packages and thin film. Berkovich indenter is one of the most popular indenter tips employed in the tests. The present study involves the finite element simulation of indentation by Berkovich-family of indenters to establish the load-displacement relations for elasto-plastic materials obeying power law. Effects of friction at the contact surfaces, which have been ignored by most of the researchers are considered in the analyses. Extensive 3-dimensional finite element analyses covering a wide practical range of materials have been carried out and the results adopted for material characterization via artificial neural network model based on an efficient reverse analysis algorithm. Direct mapping of the characteristics of the indentation curves to the material properties are performed and the characteristics of the network model deliberated. The tuned network can then be adopted to predict the mechanical properties of a new set of materials of small volume in micro-electro-mechanical components.