Static and dynamic bi-objective Evolutionary Structural Optimization based on sensitivity of both strain energy and frequency

Abstract

At present, when evolutionary structural optimization (ESO) is applied, only a single static or dynamic goal is set, which is difficult to meet engineering needs for stuctural design. In this situation, based on dimensionless processing of the strain energy sensitivity and frequency sensitivity commonly used in single objective optimization, and combined with the multi-objective optimization theory, the static and dynamic bi-objective ESO was developed. Several deep flexural members with different boundary conditions are taken as numerical examples to prove the stability and universal applicability of the new method, and meanwhile obtain the value suggestion of the weight coefficient ratio between the static optimization and the dynamic optimization. The results of finite element comparative analysis show that the new method can take the static and dynamic performance of the structure into account. Compared with the traditional single objective optimization, it can reduce material consumption but basically maintain the static stiffness of the structure, and the material utilization rate and the first-order natural frequency are continuously improved.