摘要
管道支撑在安装过程中存在安装误差,使得管道实际运行状态偏离理论设计状态,导致管口受力增大、设计失效。为此提出了一种两步优化方法以降低管口受力对安装误差的敏感性。第一步,确定性优化方法,采用多岛遗传算法进行优化求解。第二步,以支撑安装误差为噪声因素,采用蒙特卡洛模拟方法进行方差分析,以求得稳健设计解。结果表明,两步优化方法不仅获得了支撑位置的稳健解,而且减少了设计中数值模拟的次数,提高了设计效率。
Due to the installation error during the supports＇ installation process, the pipeline＇s real operation state always deviates from the state of theoretical design, thus resulting in the increase of pipe stress and design failure. For this, proposed a two-step optimization method to reduce the sensitivity of the pipe stress to the installation error. The first step, certainty optimization method, uses Multi-lsland Genetic Algorithm to optimize supports＇ location. The second step, taking the installation error as noise factor, conducting variance analysis based on Monte Carlo Simulation. The results prove that, not only did the method gain the robust design solution of supports＇ location, but it reduced the simulation times and improved the design efficiency as well.
出处
《机械设计与研究》
CSCD
北大核心
2016年第1期114-117,共4页
Machine Design And Research
基金
上海市自然基金资助项目(15ZR1424100)
关键词
管道支撑位置
安装误差
稳健设计
多岛遗传算法
蒙特卡洛模拟方法
piping supports＇
location
installation error
robust optimization design
multi-Island genetic algorithm
Monte Carlo simulation
