Bi-directional Evolutionary Structural Optimization with Adaptive Step Size

• 1、Automation School, Beijing University of Posts and Telecommunications, Beijing 100876

Abstract：For the problems of slow convergence and unit deletion caused by fixed step size in evolutionary structural optimization (ESO) and bi-directional evolutionary structural optimization (BESO) methods. The iterative evolution step size is updated by introducing discrete performance indicators and control factors, which improves the convergence speed and stability of the progressive structure optimization algorithm, and obtains the topology structure that satisfies the target volume. Furthermore, the minimum harmonic response of the continuum structure element node displacement is taken as the objective function, and a bidirectional progressive structure optimization method with adaptive step size is proposed based on the controllable discrete coefficient. The method comprehensively considers the change of structure volume and the relative level of unit sensitivity, avoiding the empirical calculation of evolutionary step size and the oscillation phenomenon of optimization process. The numerical results show that the simple harmonic response of the structural volume constraint and the displacement of the target node can be asymptotically converged. The optimized structure has a clear topological configuration and effectively improves the computational efficiency.?

Keywords： Continuum structure；Topology Optimization；Bi-directional evolutionary structural；Adaptive step size