一种基于遗传算法的装配序列规划新方法  

A new assembly sequence planning method based on the genetic algorithm

作　　者：韩晓东[1] 常伟杰[1] 

机构地区：[1]西南科技大学制造科学与工程学院,四川绵阳621010

出　　处：《机械》2008年第8期5-8,共4页Machinery

基　　金：国家自然科学基金资助项目(10576027)

摘　　要：装配序列规划的实质是NP组合优化问题,应用遗传算法来解决装配序列规划问题可以得到较好的结果。但是使用传统的遗传算法则收敛速度较慢,通过应用Boltzman变比技术来调整遗传算法在不同时期的选择压力,则可以较好的改善遗传算法的收敛速度较慢的问题。因此提出了一种应用Boltzman变比技术的遗传算法来解决装配序列规划问题的方法。通过使用精英选择技术,保证上一代种群中的最优个体能够顺利的进入到下一代中,以防止种群中优良个体的丢失。并提出了一种基于简化关联图来生成遗传算法初始种群染色体的新方法,该方法结合人工输入的方法,可以保证了初始种群染色体的质量。根据装配序列规划问题的特点,设计了具有针对性的杂交算子和变异算子。在实际应用中,该方法取得了良好的效果。The ASP＇s substance was combination and optimizing of NP, solving the problem of ASP with genetic algorithm could create a better result. But if traditional genetic algorithm was adopted the convergence rate would be slow, while if the technology of boltzman variational proportion was adopted to regulate genetic algorithm＇s pressure in different periods, the problem of slow convergence rate would be improved greatly. Therefore this paper presented the method to settle the problem of ASP with the technology of boltzman variational proportion. The technology of choosing celebrity ensured that the best individual of the last generation could come into the next generation so as to prevent the excellent individual in population from missing. And this article exhibited a new method to create the chromosome of Genetic algorithm＇s initializing generation with simplified association diagram. The method and manual input could guarantee the initializing generation＇s quality. According to ASP＇s specialty, cross and variant operator was presented. In real application, the method got a good effect.

关 键 词：装配序列规划 Boltzman变比技术 精英选择技术 遗传算法 简化关联图 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]