摘要: |
为改善叶顶气膜冷却效果,基于全局优化算法,引入数据挖掘技术,建立了凹槽状叶顶气膜孔优化设计与数据挖掘框架。以叶顶的平均气膜有效度为优化目标和以冷气流量为约束条件,对GE_E3动叶叶顶的气膜孔进行优化设计。优化后叶顶的平均气膜有效度提高了3.7倍。流动结构与冷却分析表明,优化后气膜孔的分布得到了改善,孔径的改变使得冷气流量分布更为合理,从而增加了叶顶前缘的气膜覆盖面积,增强了主流对冷气的压制效应,喷射冷气更加贴近壁面,叶顶的平均气膜有效度显著提高。同时通过对设计空间进行知识挖掘,探究设计空间信息,结果表明叶顶前缘气膜孔对叶顶气膜冷却影响显著,增大叶顶前缘气膜孔孔径,将前缘气膜孔向前缘移动,减小中部气膜孔间距,可有效改善叶顶气膜冷却效果。 |
关键词: 凹槽状叶顶 气膜冷却 全局优化算法 知识挖掘 |
DOI: |
分类号: |
基金项目: |
|
Film-Cooling Holes Design Optimization andKnowledge Mining of a Squealer Tip |
LI Chen-xi,GUO Zhen-dong,SONG Li-ming,LI Jun,FENG Zhen-ping
|
(School of Energy and Power Engineering,Xi’an Jiaotong Universtiy,Xi’an 710049,China)
|
Abstract: |
In order to improve the film-cooling effectiveness of squealer tip,by integrating data mining techniques and effective global optimization (EGO) algorithm,an optimization method for film-cooling holes on squealer tip is proposed. Using this method,the design optimization is carried out for maximizing average film-cooling effectiveness of the squealer tip of GE_E3 rotor tip,while the constraint is coolant mass flow rate. After optimization,the film-cooling effectiveness of tip increased by 3.7 times. According to flow structure and cooling flow analysis,for optimal design,the distribution of film cooling holes is improved. And the optimal design changes the dimeter of film-cooling holes to adjust the mass flow of holes of coolant air and increase flow area of coolant air in the squealer cavity. In addition,the mainstream suppression effect enhances,so that the coolant air from film cooling holes is closer to squealer tip. Therefore,the cooling effectiveness of squealer tip increases significantly. Meanwhile,the knowledge discovery of design space is carried out,the effects of design variables near leading edge on film-cooling effectiveness of squealer tip are important. Increasing the dimeter of film cooling holes and reducing the distance between film cooling holes near leading edge should improve film cooling in squealer tip. |
Key words: Squealer tip Film cooling Effective global optimization Knowledge mining |