Scholarlink

<p>The performance prediction accuracy of quasi-3D design is highly dependent on the empirical correlation model. The state of art model from experimental data is hard to meet the request of new blade development and higher aerodynamic design demand. In order to solve the problem by ignoring the influence of blade thickness position on the Reynolds number correction model, numerical research based on the middle section of the low-pressure compressor rotor is conducted. The sample database is constructed based on&#xa0;a uniform sampling method considering the incidence, thickness,&#xa0;and Reynolds number at a certain inlet Mach number, and the new model to design and off-design condition is both established by multiple nonlinear regression and adaptive simulated annealing algorithm. The results show that the root-mean-square error of the total pressure loss is reduced by 78.92% and 60.22%, error of the deviation angle is reduced by 78.44% and 78.56%, under design and off-design incidence angles respectively. The new model can provide more reliable predicted results for modern compressor design and optimization.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Model correction for loss and deviation angle of compressor cascade based on multiple nonlinear regression

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年4月3日

查看全文
  • Ziqi Zhang,
  • Bin Jiang,
  • Chunmei Zhang,
  • Zhidong Chi,
  • Huiqing Tao

摘要

The performance prediction accuracy of quasi-3D design is highly dependent on the empirical correlation model. The state of art model from experimental data is hard to meet the request of new blade development and higher aerodynamic design demand. In order to solve the problem by ignoring the influence of blade thickness position on the Reynolds number correction model, numerical research based on the middle section of the low-pressure compressor rotor is conducted. The sample database is constructed based on a uniform sampling method considering the incidence, thickness, and Reynolds number at a certain inlet Mach number, and the new model to design and off-design condition is both established by multiple nonlinear regression and adaptive simulated annealing algorithm. The results show that the root-mean-square error of the total pressure loss is reduced by 78.92% and 60.22%, error of the deviation angle is reduced by 78.44% and 78.56%, under design and off-design incidence angles respectively. The new model can provide more reliable predicted results for modern compressor design and optimization.