DEvelopment and applications of a VIrtual hybrid platform for multiscale analysis of advanced StructUres of aircraft
|Funded by:||MIUR, bando PRIN 2017|
|Start Date:||15/09/2019||Duration:||36 Months|
Damage tolerance in the design of metallic structures and aeroelastic control of aircraft are among the most significant accomplishments in the aerospace technology. These two fundamental achievements were due mostly to the work of eminent scientists and engineers whose work has led to the development of multi-disciplinary tools and knowledge to control failure of structures (i.e., fatigue behavior, progressive crack growth, non-destructive testing, residual stress evaluation, etc.), and to optimize flying conditions of aircraft (aeroelasticity, loading evaluation/alleviation, and active controls, among the others). Today, we are in a comparable paradigm for a few other and new problems, as demonstrated by current trends in research and resource drivers; Italian aircraft manufacturers are constantly working on the development of regional/executive aircraft and convertible tilt-rotor vehicles. These innovative airplanes are mostly made in composite materials. Safety, comfort and cost-effectiveness are the critical success factors of these airplanes.
The project “DEvelopment and applications of a VIrtual hybrid platform for multi-scale analysis of advanced StructUres of aircraft” (DEVISU) deals with the failure of composite structures and noise/vibration reduction, along with the investigation of new materials for aerospace applications. DEVISU will results into an integrated multi-disciplinary (structure mechanics, acoustics, control), multi-fidelity (classical and nonlocal mechanics, advanced theories of structures), multi-scale (from micro-mechanics to global/local analysis of complex structural assemblies) and hybrid approach together with its software tool implementation for the reliable, efficient and computationally effective simulation of aircraft structures. The focal point of DEVISU will be the development of a hybrid surrogate platform based on advanced theories. Surrogate models can substitute classical deterministic methodologies and alleviate the burden of complex analyses and simulations. These models will mimic the behavior of the simulation analysis as closely as possible while being computationally convenient. Metamodels are constructed using a data-driven, bottom-up approach and will be based on metadata extrapolated from DEVISU sub-tools and ad-hoc experiments.
MUL2 PhD students