摘要
为了解决航空发动机管路与卡箍智能布局优化问题,该文提出一种基于改进的鲸鱼算法进行计算求解。在该方法中考虑到鲸鱼算法易于陷入局部最优,对鲸鱼算法引入自适应权重调整策略、自适应搜索策略和随机差分变异进行改进。在卡箍布置方面,通过调整卡箍的位置来改善固有频率实现避过共振,为提高管路固有频率的分析效率以支持优化迭代,建立反应管路路径与振动性能关系的EEMD_IWOA_LSSVM代理模型。在优化计算过程中运用所建立的代理模型代替传统反复使用耗时的有限元分析程序,提高了整体优化效率。同时在管路布局优化设计过程中同时考虑优化固有频率,实现几何布局和避过共振同时优化,通过仿真试验验证了所提方法的可行性。
In order to solve the intelligent routing optimization problem of aero-engine pipe and clamp,this paper proposes an improved whale algorithm to solve it.Considering that the whale algorithm is easy to fall into local optimum,adaptive weight adjustment strategy,adaptive search strategy and random difference mutation are introduced to improve the Whale Algorithm.In order to improve the analysis efficiency of the natural frequency of the pipe and support the optimization iteration,the EEMD_IWOA_LSSVM surrogate model reflecting the relationship between the pipe path and the vibration performance is established by adjusting the position of the clamp to improve the natural frequency of the pipe to avoid the resonance.In the optimization calculation process,the surrogate model is used instead of the traditional time-consuming finite element analysis program,so as to improve the overall optimization efficiency.At the same time,in the process of pipe routing optimization design,the natural frequency is optimized at the same time,so as to realize the simultaneous optimization of geometric routing and avoiding resonance.the feasibility of the proposed method is verified by simulation experiments.
作者
沈海成
熊烨雯
SHEN Haicheng;XIONG Yewen(School of Information and Control Engineering,Liaoning Petrochemical University,Fushun 113001,China;School of Economics and Management,Fuzhou University,Fuzhou 350108,China)
出处
《电子设计工程》
2024年第9期31-36,共6页
Electronic Design Engineering
关键词
卡箍位置
固有频率
鲸鱼算法
最小二乘支持向量机
clamp position
natural frequency
Whale Algorithm
least squares support vector machine