Application of Optimization Techniques in the Design of Engine Components 2008-01-0219
Due to recent advancements of computational resources, engineers have been focusing not only in the solution of single or repetitive complex CAE analyses, but also in the development of a CAE optimization environment, which is capable to drive design parameters towards regions where selected characteristics of the project can be further improved. In the present work two cases are presented in order to illustrate, respectively, the application of a Multi-Objective optimization algorithm and a Robust Design Optimization technique in the design of real engine components. In the first example a Multi-Objective Genetic Algorithm is used in the optimization of a Conrod-Bearing, aiming to minimize its mass without endangering its performance when peak torque conditions are applied. In this case, the commercial code Ansys was used to compute the stiffness matrix of the Conrod-Bearing, while AVL's Excite was used to compute the Peak Oil Film Pressure (POFP) and Minimum Oil Film Thickness (MOFT) at the bearing. Design variables were restricted to geometry parameters, while optimization objectives comprised reduction of mass and POFP, as well as maximization of MOFT. The second case consists in the optimization of piston and rings for a specific diesel engine, with the goal to reduce the blow-by effect at low speed and full load conditions. The MIT code was used to evaluate the ring package performance based on geometric parameters provided by the user. Geometry parameters were also input variables for the optimization study. Geometric tolerances were also considered, giving raise to a Robust Design Optimization Problem. Objectives are restricted to minimization of blow-by flow, so the single objective algorithm SIMPLEX was used to obtain fast convergence. In both cases the optimized solution is compared with the original design, illustrating the advantages of optimization algorithms in real engineering applications. The commercial code modeFRONTIER was used in both cases as the process integrator and optimization tool.