Residual strength prediction of artificially damaged composite laminates based on neural networks

This paper deals with the evaluation of residual tensile strength of composite laminates containing artificial defects, consisting of impact damages of different severity and implanted holes of various diameters. Sensor fusion of acoustic emission and load data was carried out through artificial neural networks, to obtain a reliable prediction of residual tensile strength as early as possible in the loading history. The results show that neural network processing offers an effective method for the monitoring of composite specimens based on acoustic emission detection and analysi