CS-056: Using Flownex as a Tool to Develop New Engineering Methods and Correlations for Steam Turbine Performance Prediction

CHALLENGE

The challenge in developing new correlations is finding a tool to test them on. Often, one is left with the daunting task of developing additional code (such as solvers for conservation equations and fluid properties databases) to test the correlation in its entirety. Furthermore, multiple tests and parametric studies form part of the correlation development and this can be extremely time-consuming.

BENEFITS

Software that contains the fundamental governing equations, various material and fluid property databases, the most recent correlations for numerous engineering applications and third-party tool add-ins (e.g. Excel, Scripts, Engineering Equation Solver, etc.) provides the user with a full rounded tool to test and analyze newly developed correlations. Furthermore, having the ability to easily do parameter studies with various test cases empowers the user to quickly compare their models to realistic data sets.

SOLUTION

Flownex aided the development of new correlations and methods to model steam turbines more accurately using minimal geometric input data. Flownex could be further used to build a comprehensive set of guidelines and case studies for these methods, due to its ability to model an entire system of components, which further solidified the validity of these newly developed methods.

INTRODUCTION

Predicting the accurate performance of steam turbines can become a tedious task due to the intricate geometry of these types of machinery. Complex CFD software also requires one to obtain 3D geometrical models of these turbines – which is rarely provided by the original equipment manufacturers. One is often left with only a 2D-drawing of the cross-section of a turbine. A typically supplied steam turbine drawing is shown in Figure 1.

METHODOLOGY

With the aid of Flownex, Wim Fuls (an associate Professor at UCT and deputy director at ATProM Unit) was able to accurately predict the performance of steam turbines using very little geometrical inputs. Inputs which can be obtained from a 2D drawing, as shown in Figure 1.

The first method was developed using a simple Nozzle analogy and two, by Enhancing the traditional Ellipse Law for more accurate prediction of the behaviour of steam turbines.

For the Nozzle Analogy, Flownex proved to be a perfect tool in that it already had a nozzle component built-in. By applying a Nozzle for the stator and a nozzle for the rotor, with the appropriate inlet conditions and loss factors, a single-stage steam turbine could be modelled, specifically under abnormal conditions.

The second method, which is an enhancement of the traditional Ellipse law, was developed by determining methods to calculate the choking behaviour and critical pressure ratio.

CONCLUSION

Through the careful application of Flownex, correlations and methods could be developed to predict the performance of steam turbines using minimal design inputs.

Results indicated that these newly developed methods proved superior to traditional methods, such as Stodola’s Ellipse law, even under off-design conditions. Results for these newly developed methods are shown in Figure 3. 
 

Furthermore, a comprehensive set of guidelines and examples could be developed in Flownex to further solidify the validity of the correlations.

Flownex aided the development of new methods and correlations that can now be used by engineers to accurately model steam turbines with a finite number of stages under realistic conditions. Various papers have been published in journals such as Elsevier: Applied Thermal Engineering. These methods could also be presented at the ASME Turbo Expo in 2019.