Font Size: 

One of the most restricting aspects of the biaxial fatigue test is the design of the cruciform specimen. Nowadays, with the introduction of the new in-plane biaxial fatigue testing machines, based on linear electrical motors, with much lower force capacity than conventional hydraulic machines, make the optimization of the cruciform geometry crucial for successful biaxial tests. The aim of this paper is to investigate the influence of the geometric design on the stress distribution in cruciform specimens and optimize the shape of the cruciform specimens under biaxial loading conditions. The optimization is performed with two main objectives: to concentrate and initiate fatigue damage in the center of the specimen within a uniform stress region, and to minimize undesirable phenomena, such as premature failure outside the area of interest due to stress concentrations. For treating with the multi-objective optimization problem (MOO) presented, a new derivative-free methodology, called direct multi-search (DMS) method is used in order to determine the non-dominated points of the Pareto front. The evaluation is made with finite element analysis (FEA) software, ABAQUS, which uses PYTHON language for scripting. The optimization processes give much improved results and are validated by experiments.