论文部分内容阅读
针对传统FOA算法全局收敛能力差、易陷入局部极值的缺陷,提出了具有混沌映射及协同进化功能的改进果蝇算法。首先利用Logistic混沌映射功能在整个收敛域范围内搜索并初始化果蝇种群,保证算法的全局计算能力,然后根据当前果蝇个体的位置赋予搜索的方向与距离,以期全面提高算法的计算速度。采用两个优化函数测试改进后算法优化的特性,优化计算的结果显示了该算法具有良好的全局优化能力,在通用桥式起重机金属结构轻量化设计中的成功应用,体现了该算法在结构设计轻量化方面的优越性。
Aiming at the defect that the traditional FOA algorithm has poor global convergence ability and easy to fall into local extremum, an improved Drosophila algorithm with chaos mapping and co-evolution is proposed. Firstly, the Logistic chaos mapping function is used to search and initialize the fruit fly population in the entire convergence domain to ensure the global computing ability of the algorithm. Then, the search direction and distance are given according to the current Drosophila individuals’ position so as to improve the calculation speed of the algorithm. Two optimization functions are used to test the characteristics of the improved algorithm. The results of the optimization show that the algorithm has a good global optimization ability. The successful application of the algorithm in lightweight design of the universal overhead crane metal structure shows that the proposed algorithm has the advantages in structural design Lightweight advantages.