| 摘要: |
| 为了精确处理多级轴流压气机优化设计过程中的约束条件,提高优化设计可靠性,引入了随机排序微分进化算法。使用随机排序算法处理优化问题中普遍存在的约束条件,并将该算子与微分进化算法耦合。使用13个基准测试函数对算法进行了性能测试。测试结果表明该算法具有可观的精确性和稳定性。随后,对一弹用5级轴流压气机进行了优化设计,在满足约束的条件下,总压和等熵效率分别得到了1.979%和2.5%的效率提升。 |
| 关键词: 微分进化 随机排序 约束优化 流线曲率法 轴流压气机 |
| DOI： |
| 分类号: |
| 基金项目:国家自然科学基金资助项目 (51236006)。 |
|
| Optimization of Multi-Stage Axial-Flow Compressor Based on Stochastic Ranking Differential Evolution Algorithm |
|
YANG Xiao-dong, LIU Bo
|
|
(School of Power and Energy, Northwestern Polytechnical University, Xi’an 710072, China)
|
| Abstract: |
| Stochastic Ranking Based Differential Algorithm is presented with the purpose of disposing constraints precisely and enhancing the reliability of the optimization of multi-stage axial compressor. Stochastic ranking is introduced to handle the complex constraints which are common emergence and then it is coupled with differential evolutionary algorithm. 13bench-mark test functions are used to check the algorithm performance and satisfied results are obtained. The algorithm is validated by optimizing a five-stage compressor. 1.979% and 2.5% improvement is obtained for total pressure and isentropic efficiency, respectively, while all the constraints are obeyed. |
| Key words: Differential evolution Stochastic ranking Constrained optimization Streamline curvature Axial-flow compressor |
