Impact of Production Tolerances on Turbine Performance: A CFD Based Monte-Carlo Simulations Study

• Published in: Alexandria engineering journal 2025-04, Vol.117, p.289-300
• Main Authors: Elfarra, Munir Ali, Önel, Mehmet Nurullah, Yalin, Gorkem, Malik, Sheharyar
• Format: Article
• Language: English
• Subjects: Axial Turbine; CFD; Chord Length; Confidence Levels; Monte Carlo; Stagger Angle; Tolerances
• ISSN: 1110-0168
• DOI: 10.1016/j.aej.2024.12.099

Despite all design efforts targeting specific dimensions, every component exhibits some degree of tolerance due to manufacturing processes. Designers must therefore consider possible manufacturing variability besides the optimal shape when designing parts. The current research focuses on the problem of potential manufacturing discrepancies of the Aachen turbine geometry. The methodology is to use Monte-Carlo simulations based on Computational Fluid Dynamics (CFD) and Artificial Neural Networks (ANN). Deviations including the variation in the stagger angle and the chord length of the stators and the rotor are considered in this study. The performance of the turbine is characterized by the torque, mass flow rate and isentropic efficiency. This work introduces a novel methodology in the literature, focusing on identifying the key geometrical parameter that most significantly influences blade performance. Additionally, it aims to determine which performance output is most impacted by this parameter. A unique aspect of this study is the generation of confidence levels that indicate the likelihood of various performance outcomes. This approach is unprecedented and is expected to significantly aid manufacturers in predicting blade performance post-production. The results revealed that the stagger angle tolerances exert the most significant influence, with torque being the most impacted performance parameter. •Identifying the geometrical parameter that most significantly affects blade performance.•Determining which aspect of blade performance is most impacted by this parameter.•Developing a method to create confidence levels that predict the likelihood of different performance outcomes.•This innovative approach is a first in the field and is designed to assist manufacturers in forecasting blade performance post-production.