基于杂交算法的机翼结构布局优化设计

Layout Optimization of Aircraft Wing with Topology/Shaping/Sizing Variables All Considered

为了解决机翼结构布局优化问题,提出一种同时进行拓扑优化、形状优化及几何尺寸优化的杂交优化算法。对拓扑设计变量和形状设计变量采用混合编码方式构造染色体结构,利用MSC/NASTRAN实现尺寸优化,并将其结果作为布局遗传操作的依据,利用混合编码遗传算法进行布局优化。为了加快收敛进程,利用专家知识的启发性功能对布局设计区域进行了有效缩减,以产生符合工程实际要求的布局形式。通过对大展弦比机翼结构的布局优化设计计算,表明文中所提结构布局优化方法减重效果明显,是可行和有效的。

Aim. To our knowledge, there is as yet no paper in the open literature synchronously optimizing topology/shaping/sizing variables for aircraft wing. We now present a new hybrid algorithm that can implement such layout optimization and can, in our opinion, get better weigh-cutting results. In the full paper, we explain our new hybrid algorithm in detail; in this abstract, we just add some pertinent remarks to listing the two topics of explanation： （1） description of layout optimization of aircraft wing and （2） hybrid algorithm; topic 2 begins with Fig. 1 in the full paper dealing with the flow chart of the hybrid algorithm; according to this flow chart, the genetic algorithm（GA） is used to optimize the topology and shaping design variables and MSC/NASTRAN used to optimize the sizing design variables, then the results of sizing optimization are transferred to GA as the basis of genetic operation in layout optimization; the two subtopics of topic 2 are the layout decision based on expert knowledge （subtopic 2.1） and the GA mixed code （subtopic 2. 2）; under subtopic 2. 1, with the heuristic function of expert knowledge, the layout solution space to be optimized by GA can be reduced greatly so as to expedite search process; under subtopic 2. 2, the topology design variables are coded with binary numbers and the shaping design variables are coded with decimal numbers. Finally, we give a numerical example. The quick convergence of our GA（less than 100 iterations）, is shown in Fig. 4 in the full paper. For this particular wing the optimized weight is 203. 38 kg. We believe that, compared with the conventional method, our hybrid algorithm is effective and feasible for solving complex optimization and has good efficiency in the cutting of weight.