The aim of this paper is to investigate the use of Machine Learning to accelerate the processof gas turbine design optimization. Traditionally, the process simulations involved inoptimization are performed using commercial Computational Fluid Dynamics (CFD) softwaretools to find the best combinations of design parameters under specific conditionsand operational constraints. However, such simulations can become computationally toocomplex and slow for routine analysis of component designs. To address this challenge, ameta-modeling approach is developed in this study that uses "computer simulation data"collected from a small number of simulation runs. The trained metamodel is then used topredict the metal temperatures of the turbine blades to approximate the complex thermodynamicsimulations. The focus of this research is to accurately model the cooling processof a turbine by analyzing the effect of changing the diameters of the cooling holes drilledinto the gas turbine blades on the output metal temperatures.
| Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:liu-185739 |
| Date | January 2021 |
| Creators | Sharma, Namita |
| Publisher | Linköpings universitet, Institutionen för datavetenskap |
| Source Sets | DiVA Archive at Upsalla University |
| Language | English |
| Detected Language | English |
| Type | Student thesis, info:eu-repo/semantics/bachelorThesis, text |
| Format | application/pdf |
| Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.0021 seconds