Carlo Poloni

Hybridization of a multi-objective genetic algorithm, a neural network and a classical optimizer for a complex design problem in fluid dynamics

Abstract This paper describes the combination of several optimization technologies that can be used to tackle challenging design problems. The approach, that uses a multi-objective genetic algorithm, a neural network, and a gradient-based optimizer, is first outlined with the help of a computationally inexpensive mathematical test function. Then the methodology is applied to the design of a sailing yacht fin keel, coupling the optimization codes to 3D Navier–Stokes simulations. To perform the multi-objective optimization task a parallel computer is employed.

Strength of Ferritic Steels: Neural Networks and Genetic Programming

An analysis is presented of a complex set of data on the strength of steels as a function of chemical composition, heat treatment, and test temperature. The steels represent a special class designed to resist deformation at elevated temperatures (750–950 K) over time periods in excess of 30 years, whilst serving in hostile environments. The aim was to compare two methods, a neural network based on a Bayesian formulation, and genetic programming in which the data are formulated in an evolutionary procedure. It is found that in the present context, the neural network is able more readily to capture greater complexity in the data whereas a genetic program seems to require greater intervention to achieve an accurate representation.

Multi-objective optimization of turbomachinery cascades for minimum loss, maximum loading, and maximum gap-to-chord ratio

This paper illustrates an automatic multi-objective design optimization of a two-dimensional airfoil cascade row having a finite number of airfoils. The objectives were to simultaneously minimize the total pressure loss, maximize total aerodynamic loading (force tangent to the cascade), and minimize the number of airfoils in the finite cascade row. The constraints were: fixed mass flow rate, fixed axial chord, fixed inlet and exit flow angles, fixed blade cross-section area, minimum allowable thickness distribution, minimum allowable lift force, and a minimum allowable trailing edge radius. This means that the entire airfoil cascade shape was optimized including its stagger angle, thickness, curvature, and solidity. The analysis of the performance of intermediate airfoil cascade shapes were performed using an unstructured grid based compressible Navier-Stokes flow-field analysis code with k-e turbulence model. A robust stochastic algorithm was used in the automatic multi-objective constrained shape design process that had 18 design variables, 5 nonlinear constraints, and 3 objectives. Simultaneous reductions of the total pressure loss, increases of the total loading, and decreases of the number of airfoils were achieved using this method on a VKI high subsonic exit flow axial turbine cascade. 1Graduate Research Assistant. Student member AIAA. 2 Professor. Member of Russian Academy of Sciences. 3 Visiting Research Associate. 4 Professor. Director of MAIDO Laboratory. Associate Fellow AIAA. 5 Associate Professor.

Fluid-dynamic design optimization of hydraulic proportional directional valves

This article proposes an effective methodology for the fluid-dynamic design optimization of the sliding spool of a hydraulic proportional directional valve: the goal is the minimization of the flow force at a prescribed flow rate, so as to reduce the required opening force while keeping the operation features unchanged. A full three-dimensional model of the flow field within the valve is employed to accurately predict the flow force acting on the spool. A theoretical analysis, based on both the axial momentum equation and flow simulations, is conducted to define the design parameters, which need to be properly selected in order to reduce the flow force without significantly affecting the flow rate. A genetic algorithm, coupled with a computational fluid dynamics flow solver, is employed to minimize the flow force acting on the valve spool at the maximum opening. A comparison with a typical single-objective optimization algorithm is performed to evaluate performance and effectiveness of the employed genetic algorithm. The optimized spool develops a maximum flow force which is smaller than that produced by the commercially available valve, mainly due to some major modifications occurring in the discharge section. Reducing the flow force and thus the electromagnetic force exerted by the solenoid actuators allows the operational range of direct (single-stage) driven valves to be enlarged.

A Competitive Game Approach for Multi Objective Robust Design Optimization

This paper describes an application of Robust Design methodology in the transonic airfoil design. It has been observed that, minimizing the drag at a single design point (Mach number and angle of attack fixed), it is possible to find solutions characterized by poor offdesign performances (over-optimizing problem). For this reasons, the stability of the performances inside the range of operative conditions is an important objective in the design. Once the operative conditions are defined (range of Mach number and angle of attack), a Multi Objective approach is ne eded; in particular, two are the objectives to be optimized: the mean performances inside the range of operative conditions (optimise mean value of the aerodynamic coefficients) and the stability of the solution (minimize variance of the coefficients). In this Multi Objective optimization problem, we have applied a competitive Game Strategy, based on Nash equilibrium, combined with a particular mono-objective algorithm, the Simplex. The players are in charge of different objectives, corresponding to the two objectives, that have to be optimized by the Simplex algorithm. Since the variables space is split between the two players, each player influences the choices of the other one in the course of the optimisation, until an equilibrium point, corresponding to the best compromise between the objectives, is found. About the optimization test case, the range of operative conditions is Mach=0.73±0.0 5 and angle of attack 2°±0.5, and the original RAE2822 airfoil is parameterized. To reduce the high number of CFD analysis based on Navier -Stokes equations, a statistic extrapolation method, based on an adaptation of DACE, is used to define the required response surfaces. According to our results, the methodology seems to be a promising approach which offers a new possibility to the designer, in particular when a good compromise of performance and stability is required, with cheap computational resources.

Genetic Algorithm with Redundancies for the Vehicle Scheduling Problem

A real vehicle scheduling problem concerning the urban public transportation system of the city of Mestre (Venice) has been approached by Genetic Algorithm enhanced using redundancies. Redundant alleles fix the string at cross-over positions in order to improve solution feasibility. The scheduling problem has been studied both as a single and as a multiple objective optimisation problem. A significant reduction of resources as compared to the currently used solution has been achieved.

Multi-Objective Robust Design Optimization of an Engine Crankshaft

When designing a commercial product, engineers have to meet several requirements which boil down to finding the better performances and the higher reliability as possible. Another significant factor that determines product quality is its sensitivity to external or uncontrollable variations. This methodology of design is generally called Robust Design [1]. This paper shows an application of Robust Design methodology to a multi-disciplinary optimization of an engine crankshaft by considering uncertainties in terms of manufacturing errors over the shaft dimensions as well as dynamic loads variability. The application is run using ANSYS Workbench solver and modeFRONTIER [2], through a direct interface between the two codes that has been recently developed. A full Robust Design analysis is applied in order to check the stability of the best candidate solutions according to uncertainties in terms of both manufacturing errors and forcing loads The results obtained are very encouraging, and the procedure described can be applied, in principle, to even more complex problems.

An Integrated Design Approach for Micro Gas Turbine Combustors: Preliminary 0-D and Simplified CFD Based Optimization

The present work presents a novel approach for the optimised design of small gas turbine combustors, that integrates a 0-D code. CFD analyses and an advanced game theory multi-objective optimization algorithm. The output of the 0-D code is a baseline design of the combustor, given the required fuel characteristics, the basic geometry (tubular or annular) and the combustion concept (i.e. lean premixed primary zone or diffusive processes). For the optimization of the baseline design a parametric CAD/mesher model is then defined and submitted to a CFD code. Free parameters of the optimization process are position and size of the liner holes arrays, their total area and the shape of the exit duct, while different objectives are the minimisation of NOx emissions, pressure losses and combustor exit Pattern Factor. As a first demonstrative example, the integrated design process was applied to a tubular combustion chamber with a lean premixed primary zone for a recuperative methane-fuelled small gas turbine of the 100 kW class.Copyright

Optimization of a Boomerang shape using modeFRONTIER

The paper describes the optimization of a boomerang, simulating its trajectory by a dynamic model coupled to CFD analysis to compute aerodynamic coefficients. The optimization process flow and formulation is built within the commercial process integration and optimization software modeFRONTIER. The design variables involved are primarily the geometric parameters to change the shape of the boomerang. To steer the complete process of optimization, from variables variation to performance evaluation, modeFRONTIER is coupled to Catia v5 software for geometry modification and mass properties evaluation, to MATLAB for dynamic simulation, and to the commercial Computational Fluid Dynamics (CFD) tool StarCCM+ for aerodynamic analysis. In addition, dedicated RSM (Response Surfaces Methods), available in modeFRONTIER, are used to extrapolate the aerodynamic coefficients as a function of the angle of incidence and velocity, as required by the dynamic model, through a reduced number of CFD analyses (database) for each geometric configuration. Different design simulations are therefore executed automatically by modeFRONTIER following a dedicated optimization strategy, until the optimal configuration of the boomerang is found, accordingly to the specified requirements, such as minimum energy required for the launch, maximum accuracy in returning, and a guaranteed minimum range. The physical complexity of this, apparently simple, problem, and the not standard application case, has been selected as an interesting benchmark that can be disclosed in full to test the multi-objective and multidisciplinary capabilities of the optimization environment modeFRONTIER.

Application of Evolutionary Algorithms and Statistical Analysis in the Numerical Optimization of an Axial Compressor

The purpose of this work is to optimize the stator shape of an axial compressor, in order to maximize the global efficiency of the machine, fixing the rotor shape. We have used a 3D parametric mesh and the CFX-Tascflow code for the flow simulation. To find out the most important variables in this problem, we have run a preliminary series of designs, whose results have been analyzed by a statistic tool. This analysis has helped us to choose the most appropriate variables and their ranges in order to implement the optimization algorithm more efficiently and rapidly. For the simulation of the fluid flow through the machine, we have used a cluster of 12 processors.