汽轮机叶片枞树型叶根轮缘优化研究  

Design Optimization for Fir-Tree Root of Steam Turbine Blade

作　　者：谢永慧[1] 马丹丹[1] 张荻[1] 丰镇平[1] 

机构地区：[1]西安交通大学能源与动力工程学院,西安710049

出　　处：《热力透平》2012年第2期116-121,159,共7页Thermal Turbine

基　　金：国家863计划项目(2009AA04Z102)

摘　　要：叶片是汽轮机中重要的零部件之一,其可靠性直接关系到整个机组的安全运行。因此,通过改进叶片结构来弥补材料强度方面的不足,减少叶片事故,进而提高整个机组的安全性是非常必要的。对汽轮机叶片枞树型叶根轮缘完成了多变量优化分析工作,基于APDL编程语言对枞树型叶根轮缘进行了多参数建模,通过求解7个特征变量的优化值,以达到叶根轮缘最大、等效应力最小的目的;并对具体的叶片分别采用零阶算法结合一阶算法、智能优化算法(遗传算法、粒子群算法、模拟退火算法)及模式搜索等方法进行了优化。分析结果表明,在综合考虑精度及优化时间的情况下,模式搜索算法是解决本问题的最佳方法。研究结果可以为汽轮机叶片叶根轮缘部分的设计提供理论支持,并在一定程度上提高透平机组的运行可靠性。As turbine blade is the one of the most important components in steam turbine, its reliability is directly relevant to the safety performances of the unit. Improving the structure of turbine blade can countervail the shortcoming of the fatigue performance of material, reduce the accidents caused by blade failure and further improve the safety performances of the whole unit. The structure optimization of multi-variable for turbine blade was conducted. Based on the APDL, a multi-variable model of the fir-tree root-rim was established. The maximum equivalent stress of root-rim can be reduced to minimum by changing seven critical geometrical variables. A turbine blade fir-tree root was chosen as the optimization object, several optimal algorithms, including the combination of Zero-order Algorithm and First-order Algorithm, the Intelligence Optimization Algorithm （Genetic Algorithm, Particle Swarm Algorithm and Simulated Annealing Algorithm） , and Pattern Search Algorithm were adopted respectively. In consideration of the precision and design time, Pattern Search Algorithm was assessed as the best algorithm to solve this optimization problem. Relevant results of this study are expected to support the design of turbine blade and then improve the operation reliability of turbomachinery to some extent.

关 键 词：优化设计 枞树型叶根 透平叶片 最大等效应力 

分 类 号：TK263.3[动力工程及工程热物理—动力机械及工程]