大型复杂结构动态性能自动优化设计的实现

白车身等大型结构通常是由几百个不同形状的冲压薄板焊接而成的复杂系统.由于影响参数多,应用商业结构分析软件自动实现其动态优化十分困难.为优选设计变量,分别推导了固有频率和振型矢量对其几何参数的灵敏度公式.将解析法与数值计算相结合,对商业有限元软件进行了二次开发.通过修改少量零件的厚度参数,使固有频率和振型均得到明显改善.3个有关汽车车身优化的应用实例,表明采用的方法对实现大型复杂结构的固有频率和振型优化是有效的.

Implementation of Automatic Optimization on Dynamic Characteristics of Large Complex Structures

Large structures such as automobile body-in-white, are generally complex systems welded together by hundreds of stamping sheets with different geometry shapes. Being influenced and controlled by lots of parameters, it is difficult to achieve the dynamic optimization automatically by commercial structural analysis software. To reduce the number of design variables, the sensitivity of both natural frequency and mode shape vector to component geometry parameters were derived respectively. Then combining analytical and numerical approach, the commercial finite element software was re-developed. By modifying the thickness parameters of a few components, both frequencies and mode shapes were optimized markedly. Three application examples on automobile body optimization show that the presented method is an effective way attaining automatic optimization of frequencies and mode shapes for large complex structures.