Abstract:

The aero?engine clamp has an assembly constraint relationship with the pipeline, and its layout position has an important influence on the direction of the pipeline and the vibration performance. An optimization method of engine clamp position adjustment based on improved Grey Wolf Optimization (GWO) was proposed. Based on the existing pipe?clamp geometric layout scheme, this method takes the adjustment of the natural frequency of the pipe?clamp to avoid resonance as the main optimization goal, takes the pipeline obstacle avoidance and the pipeline?clamp meeting assembly constraints as constraints, and applies the improved GWO algorithm to carry out the secondary optimization and adjustment of the clamp position. In order not to affect the geometric layout of the pipeline, an optimized space for the position of the clamp based on the pipeline projection line was established. In order to improve the efficiency of optimization calculation, the Kriging surrogate model reflecting the position of clamp and the natural frequency of pipeline was established to replace the time?consuming finite element calculation program in the optimization process. In order to improve the search performance of GWO, a GWO algorithm with improved convergence factor was proposed. This method can optimize the position of the clamp under the premise of ensuring the fine adjustment of the pipe geometric layout, and can ensure that the pipe and clamp meet the assembly constraints. Finally, the effectiveness of the proposed method was verified by the test function example and the clamp arrangement example.

Key words: Natural frequency, Kriging model, Clamp layout, Gray wolf optimization

摘要：

航空发动机卡箍与管路具有装配约束关系，其布置位置对管路走向和振动性能具有重要影响。提出一种基于改进灰狼算法（Gray Wolf Optimization, GWO）的发动机卡箍位置调整优化方法。该方法在已有管路?卡箍几何布局方案基础上，以调节管路?卡箍固有频率避过共振为主要优化目标，以管路避障、管路?卡箍满足装配约束等为约束条件，应用改进GWO算法对卡箍位置进行二次优化调整。为了不影响管路几何布局方案，建立基于管路投影线的卡箍位置优化空间。为提高计算效率，建立反映卡箍位置和管路固有频率的Kriging模型，代替耗时的有限元计算程序。为提高GWO的搜索性能，提出改进收敛因子的GWO算法。所提方法可在保证管路几何布局微调的前提下，对卡箍位置进行优化，且可保证管路与卡箍满足装配约束关系。最后通过测试函数算例和卡箍布置算例验证所提方法的有效性。

关键词: 固有频率, Kriging模型, 卡箍布置, GWO优化