The Structure Topology Optimization Based On Evolutionary Algorithm

This paper is devoted to the mathematical formulation of topology optimization as well as the numerical realization through the application of the state-of-the-art optimization techniques. The objective of general topology optimization is to obtain the optimal structure through the redistribution of material in design domain, but the geometry has great influence on optimal topology. The main research of this chapter focuses on the concurrent design of topology and geometry of structureIn recent years, the evolutionary structural optimization (ESO) method has been developed into an effective tool for engineering designs. However, it limits ESO's applications that it cannot recover the finite elements removed by mistakes. Here, under the condition of that a big physical domain, whose finite element mesh has been set, is given, and based on element strain energy and an idea that man2made material is added around optimal structural cavities and boundaries, combining ESO method, a set of criteria for adding and removing finite elements, is set, and a kind of topology optimization procedure based on strain energy is given. Several examples show that the proposed method is very valid and effective for structural topology optimization, and is of good engineering application value. Evolutionary structural optimization algorithm (ESO) is a newly advanced structural optimal algorithm used in many areas. In civil engineering, it acts as guides for designers to acquire the optimal topology configurations.In the present paper, a new method of topology optimization of linearly elastic continuum structures subject to design-dependent loads is proposed. This paper gives an improved Level Set Method for topology optimization. Considering the shortcomings of LSM (Level Set Methods) that the resultant topology depends highly on that of initialization, the initial topology of complex structures, in which proper number and position of holes included in, cannot be determined in advance. Nature-Inspired computation of topology optimization shows a wonderful ideal solution to solving the problem. Combined the merits of both evolutionary optimization algorithm and LSM, the newly proposed criteria in this paper consists in inserting new holes at candidate region with low deformation energy during proper iteration. The proposed algorithm improves considerably the ability of LSM to find the optimal topology. In addition to achieving more accurate topology, the proposed algorithm is also more efficient. Numerical examples are presented to show the effectiveness and the high efficiency of the improved algorithm.