Integration Of Machining Constraints In Design Optimization Of A Guide Vane Cascade

Introduction In hydraulic turbomachinery, design optimization of blades to improve performances is an important issue. The conventional blade design and production process could be described in three steps. First, a Computer-Aided Design (CAD) modeling is done based on Computational Fluid Dynamics (CFD) simulations. Then, the tool path is computed, using Computer-Aided Manufacturing (CAM) software. Finally, blades are machined on 3 or 5-axis machine tool. The expected hydraulic performances are computed using zero defect CAD models and software [2]. On another hand, blade functional surfaces are made of freeform surfaces, obtained by high added value machining operations, like 5-axis milling. As machining constraints are not taken into account in the design step, difficulties come in the tool path computation and tool path interpolation [5] and, in the end, hydraulic performances are not those expected. In this paper, a new approach is proposed to design hydraulic blades based on machining considerations and applied to guide vane design. Main Idea In order to avoid geometrical deviations between machined blades and the CAD model, the proposed approach consists in taking into account machining constraints at a very early stage in the CAD modeling. In the proposed paradigm, the machining tool path is placed on the heart of the design process [1]. The polynomial machining tool path is computed so that the machined blade, not the CAD model, is optimum with respect to the hydraulic performances. Then, the optimization validates the surface geometry resulting from a machining simulation (Z-buffer simulation) and CAD model is the 3D representation of this simulation. Manufacturing is easier, faster, geometric quality and perceived quality are improved and hydraulic performance should meet expectations. The proposed work focuses on the comparison between the standard optimization process, called "Design", and the proposed approach called "Toolpath" based on the polynomial description of the tool trajectory using flank milling. Those two methods will be developed on a 2D study case: the geometric optimization of a guide vane cascade, one of the turbine parts which convert hydraulic power to electric power. The purpose of this article is to find out if performances obtained with a classic optimization (Design) and with machining constraints based optimization (Toolpath) are comparable. The optimization process used here can be decomposed in four stages. The first stage consists in defining a parametric model of the guide vane (Design model) or of the tool trajectory during milling (Toolpath model), e.g. Fig. 1..