Abstract: This article is focused on the component layout design optimization under random excitations to fulfill the requirements of engineering applications. In order to solve problems related to the overlapping of components during the design process, finite circle approximations are proposed to describe the exterior contour of each component. Furthermore, a finite element model is established to predict the responses under random excitations. The model and the numerical results are validated experimentally. Based on this, a general optimization model is established by defining the objective function as the minimization of the sum of the root mean square (RMS) response measured on the degrees of freedom (DOFs) of specific points of the components. Layout optimization problems with one component and four components are studied. The gradient optimization algorithm and genetic algorithm are used to solve both problems. The results show that the proposed method can deal efficiently with the layout design optimization of component structures with random vibration responses. Both optimization algorithms can reduce significantly dynamic responses at key locations of the component structure.

Key words: components, random vibration, layout optimization, dynamic response, power spectral density