Loading...

Investigation of Geometric Uncertainties in Compressor Blades via Fluid-Structure Interaction Analysis and Artificial Intelligence

Mohammadi Joozdani, Arman | 2026

39 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 58814 (45)
  4. University: Sharif University of Technology
  5. Department: Aerospace Engineering
  6. Advisor(s): Farahani, Mohammad; Fooladi, Nematollah
  7. Abstract:
  8. This research analyzes the aeroelasticity of compressor blades and develops a surrogate model based on Artificial Neural Networks (ANN) and Polynomial Chaos Expansion (PCE) for the rapid prediction of structural deformation and aerodynamic behavior. Initially, a two-way Fluid-Structure Interaction (FSI) analysis is performed to investigate the blade's response under aerodynamic loading and centrifugal forces. The results demonstrate that two-way FSI provides a more high-fidelity prediction of real-world behavior compared to conventional one-way analyses Subsequently, to account for geometric uncertainties arising from manufacturing tolerances, a series of perturbed geometries are generated and evaluated through two-way FSI simulations. The resulting dataset serves as the foundation for training the surrogate model. Given the prohibitive computational cost of full FSI simulations, the developed model predicts blade deformation and performance solely based on geometric parameters with high precision. This integrated framework, combining parametric FSI with a data-driven approach, offers a significant reduction in computational costs while enhancing the advanced design process of compressor blades
  9. Keywords:
  10. Aeroelasticity ; Compressor Blade ; Fluid-Structure Interaction ; Surrogate Model ; Polynomial Chaos Expansion ; Artificial Intelligence ; Artificial Neural Network ; Two-Way Fluid-Structure Interaction (FSI)Simulation

 Digital Object List

 Bookmark

No TOC