基于混沌序列的起重机箱形主梁粒子群优化

为了解决起重机箱形主梁优化设计中的多变量、多约束问题,提出了一种新的基于混沌序列的粒子群优化算法。利用混沌变量的遍历性和随机性,通过Logistic映射生成初始粒子群位置。根据平均粒距描述种群的多样性,通过混沌扰动对早熟个体进行局部搜索,从而跳出局部极小点。起重机箱形主梁优化试验结果表明,新混沌粒子群算法具有收敛速度快,优化精度高的优点。

Particle swarm optimization for box girder of crane based on chaos series

To solve the problems of the multi - variables and multi - constraints in crane main girder optimization design, the paper proposes a new particle swarm optimization algorithm based on chaos series. It also describes how the initial position of the particle swarm is produced by means of the ergodicity and randomness characteristics of chaostic variables and the Logistic mapping, how the mature individuals are locally searched by means of the population diversity evaluated by distance - to - average - point and chaos disturbance, and how to jump out of the local minimum. The optimization result shows that it can greatly improve the convergence speed and optimization precision.