In modern industry, resort to numerical simulation represents the standard at the base of every new design. Virtual prototyping allows to save costs and time associated to the production of new optimized components, and the accuracy of the numerical solution plays a crucial role in the reduction of number of iterations required to arrive at the final design. Research in computational mechanics is increasingly aiming at the formulation of complex interaction modelling for multi-physics. The final goal is to pave the way toward simulating as close as possible the anticipated behaviour of the prototype in its different operational states.
The present project follows this approach, by capitalizing on the expertise of its members in the different fields of mechanics. The main aim is to achieve a complete description in terms of models, numerical formulations and interface algorithms to simulate a rotor machine, when a strong interaction takes place between the dynamics of fluid and solid (e.g., fluid structure interaction-FSI and/or particle laden flows).
As far as rotor design for open (propellers, wind turbines, helicopter rotors) and ducted (turbomachinery and fans) applications is concerned, the current state-of-the-art is quite satisfactory. However, the introduction of new materials and concepts markedly increased the flexibility of the blades, so that aero-elastic effects become crucial in evaluating blade performance and loads.
Similar considerations apply to material wearing. This process may considerably alter the shape and roughness of the profile during its operating life, thus affecting the flow field and, in turn, the motion of transported particles and their erosive effect.
The present proposal is divided in two macro-areas: a) fluid structure interaction; b) particle¿structure and aerodynamic interaction. Each of them will be analysed in depth by means of dedicated models.
The innovative points of the present proposal reside in the development of an integrated multi-physics modeling platform based on the adoption of detailed sub-models in the fluid-solid interaction problems. Under the standpoint of virtual prototyping, this is supposed to lead to the quick generation of printed prototypes, adopting for instance new plastic materials and additive manufacturing technology. Experimental analysis will be carried out to validate the outcome of the developed platform.
In open and ducted rotor blade the efficiency depends to a crucial extent on blades aerodynamics. In this respect, flows exhibiting separation as well as non-linearity and turbulent features, require complex but high-fidelity models.
The rotor performance can be severely deteriorated by a number of multi-physics issues.
a) Fluid-structure interaction. From a structural viewpoint, additive manufacturing and composite technology open the way to the use of solid (non-shell) light and high flexible materials. In this case, structure dynamic response must be evaluated by non-linear models, accounting for large displacement and non- (or neo-) Hookean anisotropic material behaviour. In this respect, the state-of-the-art solutions are typically routed on BEM (Blade Element Momentum) or potential theory for the fluid phase coupled with one-dimensional beam model for the structure. The innovation against such a background, is associated to: i) the strong coupling method, and ii) the three-dimensional flow field modelling (with appropriate turbulence scales resolution). With respect to the standard and very fast FSI approaches, the proposed research strategy promises to solve for separation induced phenomena such as unsteady wake shedding or dynamic stall.
b) Particle-blade interaction. The efficiency of a rotor can be downgraded by up to 1% by the deposition of soot particles (particulate formed by hydrocarbon combustion) on blades due the sticking nature of such particles. Moreover, when in case of dispersed phase made of hard particles, the performance of the blade can be deteriorated by the progressive material erosion. Predictions of deposit/erosion phenomena are particularly challenging. In this respect, the innovation of the present project advocates the development of particle transport models able to account for: i) fluid-particle coupling with non-spherical particle shape, and ii) particle-blade interaction with changing geometries.
The new features developed by the present research implies the simulation of the mutual interaction among the several simulation techniques for the different physical sub-systems, in order to improve the reliability and the consistency of the proposed integrated approach, ultimately leading to the capability to generate a full CAE (Computer Aided Engineering) tool, allowing the direct simulation of the aero-structural behaviour of rotating machines. This envisages promoting a new paradigm of virtual prototyping, where simulations and experiments (on which performance and operational tests are based) can reproduce in a more reliable and complete way the dynamic behaviour of the structure and of the fluid operated by the rotor.
Obviously, the proposed technology can also serve as an inspiration for the development of standardized design methods. This work is accordingly intended to set up a theoretical and implementation basis for the development of advanced, even more efficient, tools of such a kind. The final goal would be to produce automatic optimization algorithms, allowing to drastically shorten the cut-and-try phase for both structural resistance and aerodynamic performance.
A further goal of the proposed research is life prediction capability of the virtual prototyping tools, in order to reduce the costs associated to the maintenance of these mechanical assemblies. The development of the technology and methods here proposed is expected to give a more correct prediction of fatigue cycles and aging, as well as the anticipated performance reduction. Consequently, the designer will be in a better position to guarantee given performance and lifetime, with obvious repercussions on the optimization of the maintenance scheduling.