Numerical Simulation For ANN Training and validation For Impact Detection

Composite structures, today, have a relevant role in our life. The most of the object we use daily are done of composite materials. The recent increase in use of composite materials can be explained if we consider: * They have great strength; * They have low weight; The combination of these two characteristics is the main reason the composite materials are so appreciate in engineering. Every time we need to produce a light structure with high strength composite materials are the good approach to design it. Instead of benefits, composite materials have also some issues. From a structural point of view the most critical is the difficult to monitoring them. Metal materials integrity is easy to check, the most of the time a visual inspection is enough to detect failures also in early stage. In composite materials, due the way they are maiden, often is impossible detect failures until it is too late to repair them. The most common problem composite materials have is the delamination. It consists in a detachment of plys inside the material due to an impact on the material itself. From an exterior point of view it is impossible to detect by eye. It suddenly propagates inside the structure inducing failures. The task of this work is to develop an artificial neural network (ANN) able to detect impacts and restrict the part of structure to monitor looking for damages. II