51st AIAA Aerospace Sciences Meeting including the New Horizons Forum and Aerospace Exposition, 2013
Industrial aerodynamic design optimization of helicopter rotor blades requires employment of mult... more Industrial aerodynamic design optimization of helicopter rotor blades requires employment of multi-objective optimization methods to account for the two distinct objectives in hover and forward flight. Genetic algorithms are preferred for finding the Pareto Optimal Front, as they allow the engineer to select out of optimal designs. An optimization loop is created, coupling the Dakota optimization library and two simulation methods for objective function evaluation: comprehensive rotor code HOST and CFD solver elsA. For low-cost rotor simulations by HOST, a genetic algorithm is employed to maximize hover and forward flight rotor performance in single-and two-point optimizations. Twist and chord laws of the 7A blade are optimized separately and simultaneously. As genetic algorithms require too many cost function evaluations for CFD-based optimizations, Surrogate Based Optimization (SBO) is employed. SBO is initialized by a preliminary Design of Experiment (DoE). The results are used to generate a metamodel for estimation of cost function evaluations in the optimization algorithm. The metamodel is updated using information from subsequent simulations. Validation of HOST-based SBO against full genetic algorithm optimizations shows that the Pareto Optimal Front is correctly represented by SBO, while requiring 88% less cost function simulations. Several sizes of initial DoE, number of update cycles and number of simulations added per cycle are tested. Then, a similar SBO optimization is carried out by replacing the HOST code by CFD for hover performance simulation. The results demonstrate the ability of both solvers and both optimization techniques to perform aerodynamic design optimization of helicopter rotor blades.
The design of an efficient organic rankine cycle (ORC) expander needs to take properly into accou... more The design of an efficient organic rankine cycle (ORC) expander needs to take properly into account strong real gas effects that may occur in given ranges of operating conditions, which can also be highly variable. In this work, we first design ORC turbine geometries by means of a fast 2-D design procedure based on the method of characteristics (MOC) for supersonic nozzles characterized by strong real gas effects. Thanks to a geometric post-processing procedure, the resulting nozzle shape is then adapted to generate an axial ORC blade vane geometry. Subsequently, the impact of uncertain operating conditions on turbine design is investigated by coupling the MOC algorithm with a Probabilistic Collocation Method (PCM) algorithm. Besides, the injector geometry generated at nominal operating conditions is simulated by means of an in-house CFD solver. The code is coupled to the PCM algorithm and a performance sensitivity analysis, in terms of adiabatic efficiency and power output, to variations of the operating conditions is carried out.
Notes on Numerical Fluid Mechanics and Multidisciplinary Design, 2015
Recent developments about the extension of high-order Residual-Based Compact schemes to unsteady ... more Recent developments about the extension of high-order Residual-Based Compact schemes to unsteady flows and complex configurations are discussed, with application to scale-resolving simulations and complex turbomachinery flows.
A systematic analysis of the effect of modeling uncertainties in the simulation of wet steam flow... more A systematic analysis of the effect of modeling uncertainties in the simulation of wet steam flows is performed through the coupling of a wet steam flow solver with an uncertainty quantification method. A Probabilistic Collocation Method (PCM) is selected due to its non-intrusive nature and exponential convergence. Uncertainties affecting calibration coefficients in the droplet nucleation and growth models are propagated through the wet steam flow solver using the PCM approach The separated and coupled effect of such uncertainties is studied for condensing flows through both high-pressure and low-pressure quasi-1D supersonic nozzles.
Astudy about convergence of Genetic Algorithms (GAs) inshape,optimization problems for inviscid f... more Astudy about convergence of Genetic Algorithms (GAs) inshape,optimization problems for inviscid flows of real gases is presented. Specifically, working fluids of the Bethe—Zel’dovich—Thompson ,(BZT) type ,are considered, which exhibit non classical dynamic behaviors in the transonic/supersonic regime, such as the disintegration of compression shocks. A reference, single- objective optimization problem, namely, wave drag minimization for a non-lifting inviscid transonic
Dense Gases (DG) are single-phase vapors of molecularly complex fluids operating close to saturat... more Dense Gases (DG) are single-phase vapors of molecularly complex fluids operating close to saturation conditions which can be used to improve the efficiency of Organic Rankine Cycles (ORC) turbines. This work is devoted to the design of efficient procedures for optimizing a 2D dense gas turbine with multiple sources of uncertainty (thermodynamic model, operating conditions, geometry). Uncertainty Quantification (UQ) stochastic tools, dense gas CFD solvers and a multi-objective optimizer are coupled to produce a set of robust optimal shapes using mean efficiency and its standard deviation as objectives.
Unsteady turbulent transonic flows over oscil-lating airfoils are investigated by using a novel t... more Unsteady turbulent transonic flows over oscil-lating airfoils are investigated by using a novel third-order accurate scheme on block-structured meshes. Two turbu-lence models (algebraic and two-transport equation mod-els) are studied in order to generate flow fields in agree-ment with experimental data. Numerical results are pre-sented for 2-D problems related to helicopter rotors in for-ward flight, including shock-stall phenomena. The advan-tages of using a high-order accurate method and a suitable turbulence model for this kind of applications are demon-strated.
Journal of Engineering for Gas Turbines and Power, 2014
Transonic flows through axial, multi-stage, transcritical ORC turbines, are investigated in detai... more Transonic flows through axial, multi-stage, transcritical ORC turbines, are investigated in detail by using a numerical solver including advanced multiparameter equations of state and a high-order discretization scheme. The working fluids under investigation are the refrigerants R134a and R245fa, classified as dense gases due to their complex molecules and relatively high molecular weight. Both inviscid and viscous numerical simulations are carried out in order to quantify the impact of dense gas effects and viscous effects on turbine performance. Both supercritical and subcritical inlet conditions are studied for the considered working fluids. In the former case, flow across the turbine is transcritical, since turbine output pressure is subcritical. Numerical results show that, thanks to dense gas effects characterizing the flow at supercritical inlet conditions, supercritical ORC turbines enable, for a given pressure ratio, a higher isentropic efficiency than subcritical turbines using the same working fluid. Moreover, for the selected operating conditions, R134a provides a better performance than R245fa.
International Journal of Computational Fluid Dynamics, 2014
Residual-based-compact schemes (RBC) of 2nd and 3rd-order accuracy are applied to a challenging 3... more Residual-based-compact schemes (RBC) of 2nd and 3rd-order accuracy are applied to a challenging 3D flow through a transonic compressor. The proposed schemes provide almost mesh-converged solutions in good agreement with experimental data on relatively coarse grids, which allows achieving a given accuracy level with computational cost reductions by a factor between 2 and 4 with respect to standard solvers.
51st AIAA Aerospace Sciences Meeting including the New Horizons Forum and Aerospace Exposition, 2013
Industrial aerodynamic design optimization of helicopter rotor blades requires employment of mult... more Industrial aerodynamic design optimization of helicopter rotor blades requires employment of multi-objective optimization methods to account for the two distinct objectives in hover and forward flight. Genetic algorithms are preferred for finding the Pareto Optimal Front, as they allow the engineer to select out of optimal designs. An optimization loop is created, coupling the Dakota optimization library and two simulation methods for objective function evaluation: comprehensive rotor code HOST and CFD solver elsA. For low-cost rotor simulations by HOST, a genetic algorithm is employed to maximize hover and forward flight rotor performance in single-and two-point optimizations. Twist and chord laws of the 7A blade are optimized separately and simultaneously. As genetic algorithms require too many cost function evaluations for CFD-based optimizations, Surrogate Based Optimization (SBO) is employed. SBO is initialized by a preliminary Design of Experiment (DoE). The results are used to generate a metamodel for estimation of cost function evaluations in the optimization algorithm. The metamodel is updated using information from subsequent simulations. Validation of HOST-based SBO against full genetic algorithm optimizations shows that the Pareto Optimal Front is correctly represented by SBO, while requiring 88% less cost function simulations. Several sizes of initial DoE, number of update cycles and number of simulations added per cycle are tested. Then, a similar SBO optimization is carried out by replacing the HOST code by CFD for hover performance simulation. The results demonstrate the ability of both solvers and both optimization techniques to perform aerodynamic design optimization of helicopter rotor blades.
The design of an efficient organic rankine cycle (ORC) expander needs to take properly into accou... more The design of an efficient organic rankine cycle (ORC) expander needs to take properly into account strong real gas effects that may occur in given ranges of operating conditions, which can also be highly variable. In this work, we first design ORC turbine geometries by means of a fast 2-D design procedure based on the method of characteristics (MOC) for supersonic nozzles characterized by strong real gas effects. Thanks to a geometric post-processing procedure, the resulting nozzle shape is then adapted to generate an axial ORC blade vane geometry. Subsequently, the impact of uncertain operating conditions on turbine design is investigated by coupling the MOC algorithm with a Probabilistic Collocation Method (PCM) algorithm. Besides, the injector geometry generated at nominal operating conditions is simulated by means of an in-house CFD solver. The code is coupled to the PCM algorithm and a performance sensitivity analysis, in terms of adiabatic efficiency and power output, to variations of the operating conditions is carried out.
Notes on Numerical Fluid Mechanics and Multidisciplinary Design, 2015
Recent developments about the extension of high-order Residual-Based Compact schemes to unsteady ... more Recent developments about the extension of high-order Residual-Based Compact schemes to unsteady flows and complex configurations are discussed, with application to scale-resolving simulations and complex turbomachinery flows.
A systematic analysis of the effect of modeling uncertainties in the simulation of wet steam flow... more A systematic analysis of the effect of modeling uncertainties in the simulation of wet steam flows is performed through the coupling of a wet steam flow solver with an uncertainty quantification method. A Probabilistic Collocation Method (PCM) is selected due to its non-intrusive nature and exponential convergence. Uncertainties affecting calibration coefficients in the droplet nucleation and growth models are propagated through the wet steam flow solver using the PCM approach The separated and coupled effect of such uncertainties is studied for condensing flows through both high-pressure and low-pressure quasi-1D supersonic nozzles.
Astudy about convergence of Genetic Algorithms (GAs) inshape,optimization problems for inviscid f... more Astudy about convergence of Genetic Algorithms (GAs) inshape,optimization problems for inviscid flows of real gases is presented. Specifically, working fluids of the Bethe—Zel’dovich—Thompson ,(BZT) type ,are considered, which exhibit non classical dynamic behaviors in the transonic/supersonic regime, such as the disintegration of compression shocks. A reference, single- objective optimization problem, namely, wave drag minimization for a non-lifting inviscid transonic
Dense Gases (DG) are single-phase vapors of molecularly complex fluids operating close to saturat... more Dense Gases (DG) are single-phase vapors of molecularly complex fluids operating close to saturation conditions which can be used to improve the efficiency of Organic Rankine Cycles (ORC) turbines. This work is devoted to the design of efficient procedures for optimizing a 2D dense gas turbine with multiple sources of uncertainty (thermodynamic model, operating conditions, geometry). Uncertainty Quantification (UQ) stochastic tools, dense gas CFD solvers and a multi-objective optimizer are coupled to produce a set of robust optimal shapes using mean efficiency and its standard deviation as objectives.
Unsteady turbulent transonic flows over oscil-lating airfoils are investigated by using a novel t... more Unsteady turbulent transonic flows over oscil-lating airfoils are investigated by using a novel third-order accurate scheme on block-structured meshes. Two turbu-lence models (algebraic and two-transport equation mod-els) are studied in order to generate flow fields in agree-ment with experimental data. Numerical results are pre-sented for 2-D problems related to helicopter rotors in for-ward flight, including shock-stall phenomena. The advan-tages of using a high-order accurate method and a suitable turbulence model for this kind of applications are demon-strated.
Journal of Engineering for Gas Turbines and Power, 2014
Transonic flows through axial, multi-stage, transcritical ORC turbines, are investigated in detai... more Transonic flows through axial, multi-stage, transcritical ORC turbines, are investigated in detail by using a numerical solver including advanced multiparameter equations of state and a high-order discretization scheme. The working fluids under investigation are the refrigerants R134a and R245fa, classified as dense gases due to their complex molecules and relatively high molecular weight. Both inviscid and viscous numerical simulations are carried out in order to quantify the impact of dense gas effects and viscous effects on turbine performance. Both supercritical and subcritical inlet conditions are studied for the considered working fluids. In the former case, flow across the turbine is transcritical, since turbine output pressure is subcritical. Numerical results show that, thanks to dense gas effects characterizing the flow at supercritical inlet conditions, supercritical ORC turbines enable, for a given pressure ratio, a higher isentropic efficiency than subcritical turbines using the same working fluid. Moreover, for the selected operating conditions, R134a provides a better performance than R245fa.
International Journal of Computational Fluid Dynamics, 2014
Residual-based-compact schemes (RBC) of 2nd and 3rd-order accuracy are applied to a challenging 3... more Residual-based-compact schemes (RBC) of 2nd and 3rd-order accuracy are applied to a challenging 3D flow through a transonic compressor. The proposed schemes provide almost mesh-converged solutions in good agreement with experimental data on relatively coarse grids, which allows achieving a given accuracy level with computational cost reductions by a factor between 2 and 4 with respect to standard solvers.
Uploads
Papers by Paola Cinnella