Estimation of Damage and Fracture Process by a Markov Process Inverse Analysis Model for Micro-sized CFRP

Abstract

The application of carbon fiber reinforced plastics (CFRP) has been expanding as large structural members, whereas there is a growing trend to expand its use to structural members of small devices such as “drones”. From such background, this study sampled strength data from tensile tests of micro-sized unidirectionally reinforced CFRP in diameter or thickness and discussed the damage and fracture process, as well as the number n_c fiber-break points up to fracture from the data. In our recent paper, the authors estimated the number of fiber-break points from previously obtained sampling data using an inverse analysis model of Markov processes. In this study, furthermore, another inverse analysis model, in which analytical solution can be obtained, was newly proposed. This model was in a good agreement with absorbing state probabilities and n_c obtained from the previous model. However, it was also found that there was a data population for which the fracture probability was limited to less than one. This is because the estimation limit occurs when some of the data are somewhat far from the data population. This problem can be solved to some extent by using the maximum likelihood estimation method for complete or censored data.