In-situ mechanical property identification and delamination growth prediction of laminates

Abstract

Accurately determining the mechanical properties of laminates and predicting their mechanical behavior are vital for structural design and reliability assessment. In this study, an identification method is proposed for simultaneously identifying in-situ elastic properties and interface properties of laminates. To alleviate the ill-posedness and improve the identification accuracy, the staged response characteristics of laminates were considered, and sensitivity analysis was used to decrease the quantity of properties being identified concurrently. Meanwhile, a multi-source information fusion strategy was adopted to improve the accuracy of property identification, which was validated through the compression experiment of open-hole laminates with delamination. The results show that the maximum deviation in property identification compared to the standard test is 5.34 %. Compared with using the single observation information, using multi-source information fusion for property identification has better accuracy in predicting structural response. Furthermore, the identified properties were employed to predict the delamination growth of the laminate. The difference between the predicted and actual delamination growth length is only 3.23 %.