The Improved Bat Algorithm For Engineering Structure Optimization Problems

Optimization problem is a universal engineering mathematical problem.Its main goal is to find the best or ideal solution.The problems on the generalized optimization algorithms are broad and diverse,and the theoretical analysis has always been a research field worthy of attention.On the whole level,optimization problems can be divided into deterministic optimization and non-deterministic optimization.Although the deterministic optimization algorithm is relatively mature,its engineering application conditions are relatively harsh.It is usually difficult to apply to complex nonlinear optimization problems.This promotes the vigorous development of stochastic optimization algorithms,especially swarm intelligence optimization algorithms.Bat algorithm(BA)is a biologically heuristic swarm intelligence optimization algorithm.After analyzing the basic formula and operation principle of BA,this thesis proposes a new improvement strategy for the shortcomings of the original BA,and effectively improves the optimization performance of the algorithm.Then the improved algorithm is applied to a variety of engineering structure optimization problems.The main research contents and conclusions of this thesis can be summarized as follows:(1)As a famous algorithm for searching global optimal solution,BA still has the shortcomings of low convergence accuracy and be easy to fall into local optimum.In view of these shortcomings,this thesis proposes an adaptive bat algorithm BAS-BA based on hybrid beetle antennae search(BAS).The algorithm combines the improved BAS based on omni-directional antennae perception with the adaptive bat ultrasonic detection technology.The improved BAS search mechanism improves the diversity of the population,avoids the premature phenomenon,and enhances the global search ability of the algorithm BAS-BA.The adaptive bat ultrasonic detection mechanism more realistically simulates the echolocation behavior of bats in nature,which effectively improves the local optimization ability of BAS-BA.(2)Through a series of unimodal and multimodal benchmark functions,the BASBA,the original BA and other swarm intelligence algorithms are tested.The results show that the algorithm BAS-BA is significantly superior than BA and the other three types of swarm intelligence optimization algorithms in robustness and convergence accuracy.Numerical experiments show that the hybrid improved algorithm BAS-BA has better optimization performance.(3)The algorithm BAS-BA is applied to five classical engineering optimization problems and five truss structure size optimization problems.These engineering optimization problems include continuous variable,discrete variable and mixed variable optimization problems.The simulation results show that this algorithm can jump out of the local optimum trap faster and obtain more accurate solutions stably,compared with the original BA,the improved BAs and other types of swarm intelligence optimization algorithms.The above results show that the hybrid improved algorithm BAS-BA can be used to solve complex engineering structure optimization design problems.