摘要
遗传算法在引入动态自适应策略和"免疫算子"之后,其收敛效率有了较大的提高。对于以可靠性作为约束的随机结构系统优化问题,以罚函数法为基础,构造了有效的约束与目标函数向适应值函数的映射公式。建立合理的安全余量,采用随机有限元法处理结构分析中所涉及的有关参量的随机性,计算结构响应和敏度分析,并进行系统的可靠性计算。通过两种算法的结合,利用自适应免疫遗传算法、随机有限元法及可靠性的基本理论,建立了一套完整的基于遗传算法与随机有限元的结构系统可靠性优化设计方法。对舰船舱段结构梁系截面的优化计算结果表明,算法对多随机因素结构系统的可靠性优化具有良好的收敛效果。
After dynamic adaptive strategy and immunity operator are introduced into Genetic Algorithm （GA）, the algorithm can be more manageable in structural optimization. In dealing with reliabilitybased structural optimization, particularly for random structure object, based on exterior penalty function method, mapping formula of objective function and constraint transformed to fitness function is made. In this process, rational safety margin formula is established, and using Stochastic Finite El- ement Method （SFEM） the random variables are dealt with effectively, the structure response and sensitivity analysis are performed as well as the system reliability index is calculated. Combining the theory of Adaptive Immunity Genetic Algorithm （ALGA）, SFEM and basal reliability theory, an inte- grated AlGA and SFEM-based method for reliability-based optimization is presented, and by means of the method the calculation of the ship cabin beams is carried out. The calculation result shows that the algorithm recommended by author is effective in reliability based optimization for structural system with many random factors.
出处
《中国造船》
EI
CSCD
北大核心
2007年第4期26-32,共7页
Shipbuilding of China
关键词
船舶
舰船工程
舰船舱段
可靠性优化
遗传算法
适应值函数
随机有限元
ship engineering
ship cabin
reliability-based structural optimization（RBSO）
genetic algorithm
fitness function
stochastic finite element method （SFEM）
