Search

This paper presents the formulation of an axi-symmetric element based on an efficient method called 'Inverse Approach' (I.A.) for the numerical modeling of cold forging process. In contrast to the classical incremental methods, the Inverse Approach exploits the known shape of the final part and executes the calculation from the final part to the initial billet. The assumptions of the proportional loading and the simplified tool actions make the I.A. calculation very fast. The metal's incompressibility is ensured by the penalty method. The comparison with ABAQUS® and FORGE® shows the efficiency and limitations of the I.A. This simplified method will be a good tool for the preliminary preform design. [ABSTRACT FROM AUTHOR]

This paper presents an approach of combined direct compensation and surface approximation implemented in a forging optimization system in order to achieve net-shape forging for aeroengine blade components. This approach is based on modification of individual nodal positions on the die surface and approximation of the modified die surface using a B-spline surface for improved surface quality. The results show that, with this method, forging optimization computations not only produce fast converged solution in forging simulation but also provide significant reduction of the total forging error. © 2007 American Institute of Physics [ABSTRACT FROM AUTHOR]

This paper presents several approaches for preform shape optimization in 3D forging. The process simulation is carried out using the FORGE3® finite element software, and the optimization problem regards the shape of initial axisymmetrical preforms. Several objective functions are considered, like the forging energy, the forging force or a surface defect criterion. Both deterministic and stochastic optimization algorithms are tested for 3D applications. The deterministic approach uses the sensitivity analysis that provides the gradient of the objective function. It is obtained by the adjoint-state method and semi-analytical differentiation. The study of stochastic approaches aims at comparing genetic algorithms and evolution strategies. Numerical results show the feasibility of such approaches, i.e. the achieving of satisfactory solutions within a limited number of 3D simulations, less than fifty. For a more industrial problem, the forging of a gear, encouraging optimization results are obtained. © 2004 American Institute of Physics [ABSTRACT FROM AUTHOR]