Application of Genetic Algorithms to Design of an Internal Turbine Cooling System

Author(s):  
C. F. F. Favaretto ◽  
K. Funazaki

This paper deals with the development of an optimizing technique based on Genetic Algorithms (GA), which can be applied to the optimization of an internal cooling system of turbine nozzles. An impingement cooling system with pins and air-discharging holes is selected as target cooling system to be optimized using a single-objective GA code developed in this study. The optimization is performed for several design parameters such as the impingement and discharging hole diameters, pin diameter and pin height. The computational grid is automatically generated and boundary conditions prescribed. A commercial CFD code is used to evaluate the target function, which is in the present case simply defined as the ratio between the averaged heat transfer coefficient multiplied by the wetted area and the pressure loss. A hybrid FORTRAN/UNIX shell script program controls the whole process of the optimization, leading to the successful achievement for finding an optimum configuration of the cooling system concerned.

Author(s):  
Philippe T. Lott ◽  
Ingrid Lepot ◽  
Emmanuel Chérière ◽  
François Thirifay ◽  
Klaus Semmler ◽  
...  

The design of turbine cooling systems remains one of the most challenging processes in engine development. Modern turbine cooling systems indeed invariably combine internal convection cooling with external film cooling in complex flow systems. The heat transfer and cooling processes are at the limit of current understanding and engine designers heavily rely on empirical tools and engineering judgment to produce new designs. These designs are moreover developed in the context of continuously increasing Turbine Entry Temperature (TET) as the latter leads to improvement of Specific Fuel Consumption (SFC). The present contribution fits into the frame of the ongoing FP7 ER-ICKA project. It focuses on achieving a significant progress in understanding turbine blade passages internal cooling systems by gathering high quality experimental data and by developing cooling state-of-the-art design capabilities based upon computer codes calibrated through these experimental data. In this context, the paper will describe the design optimisation and analysis work performed for two different internal cooling passages configurations, namely a static leading edge LP configuration passage (baseline experimentally tested at Stuttgart University) and a rotating mid-chord HP configuration passage (baseline experimentally tested at ONERA). The aim of the work was to develop a design methodology to optimise turbulence promoting ribs shape and arraying to improve the thermal behaviour of the internal cooling passages while avoiding excessive head loss. The optimisation was driven using decoupled rib design parameters for each ribbed wall to enhance flow interactions and maximise disturbances, to maximise potential increase in Heat Transfer Coefficients (HTCs). Any improvement in the thermal behaviour of the cooling system may indeed allow to either reduce the coolant mass flow rate requirements or increase the TET. To drive these optimisations, the ultimate target was hence to reduce the maximum blade metal temperature. To this end, suitable cost functions (objectives and constraints) have been derived and implemented. They will first be presented and discussed along with the parameterisations, so as to define the complete optimisation specification. The computational chain setup, among which the challenging mesh regeneration choices set based on a mesh dependence study will then be detailed. Validation of the CFD evaluation against the experimental results will be described for the static baseline configuration at least (rotating test measurements are still ongoing) and the optimisation results, which have led to significant gains in HTCs, will finally be analysed, data mining techniques allow to identify key parameters, path taken in the conception space and major trends.


Author(s):  
D. Jackson ◽  
P. Ireland ◽  
B. Cheong

Progress in the computing power available for CFD predictions now means that full geometry, 3 dimensional predictions are now routinely used in internal cooling system design. This paper reports recent work at Rolls-Royce which has compared the flow and htc predictions in a modern HP turbine cooling system to experiments. The triple pass cooling system includes film cooling vents and inclined ribs. The high resolution heat transfer experiments show that different cooling performance features are predicted with different levels of fidelity by the CFD. The research also revealed the sensitivity of the prediction to accurate modelling of the film cooling hole discharge coefficients and a detailed comparison of the authors’ computer predictions to data available in the literature is reported. Mixed bulk temperature is frequently used in the determination of heat transfer coefficient from experimental data. The current CFD data is used to compare the mixed bulk temperature to the duct centreline temperature. The latter is measured experimentally and the effect of the difference between mixed bulk and centreline temperature is considered in detail.


Author(s):  
Zhongran Chi ◽  
Haiqing Liu ◽  
Shusheng Zang ◽  
Guangyun Jiao

This paper discusses the methodology of impingement cooling optimization of a gas turbine 2nd stage vane with 3D conjugate heat transfer (CHT) CFD analysis applied. The vane is installed with a novel impingement cooling structure in the leading cavity and a pin-fin array in the trailing edge. This study involves the optimization of the impingement cooling structure, including the location of the jet holes and the diameter of each hole. The generation of 3D model and CHT mesh was realized using an in-house code developed specifically for turbine cooling optimization. A constant pressure drop was assumed within the cooling system during optimization. To make the optimization computationally faster, a metamodel which can predict the detailed distribution of metal temperature on the vane surface was used in the second-level search together with a genetic algorithm. An optimal nonuniform impingement cooling structure in the leading cavity was automatically designed by the optimization process costing only dozens of CFD runs, which provided a more uniform temperature distribution on the vane surface and required no more coolant amount compared with the initial impingement cooling structure.


Author(s):  
Donghwa Kim ◽  
Hyungyu Lee ◽  
Jungsoo Lee ◽  
Jinsoo Cho

To supply cooling air which delivered from turbine stationary parts to the rotational parts, pre-swirl system have been widely used in gas turbines with various shapes and types. The key functions of pre-swirl system is to provide turbine cooling temperature lower than pre-swirl system inlet within required pressure drop target. In modern gas turbine system, pre-swirl system is a key component in the turbine cooling system because first rotor entering flow qualities were defined by pre-swirl system. Especially most of modern blade cooling system were required adequate coolant pressure level to meet film cooling conditions and also need to keep lower level of temperature condition to improve cooling effectiveness in their cooling circuit. To meet the design target, cooling system designer should built effective pre-swirl system by increase swirl ratio within required pressure margin. When the design results of pre-swirler system was satisfied initial design requirements, designer should select best pre-swirler system based on its discharge coefficient characteristics. To verify analytic methodologies, to compare main design parameters of this research and former studies which reference data already validated analytically and experimentally. To find best pre-swirl nozzles shape, total 15 kinds of airfoil was selected for detail investigation. Best available was selected by pre-swirl performance comparing. Also shape optimization was applied to improve pre-swirl performance within minor geometrical changes within mechanical integration limit. To verify inlet duct system effects, the effect of the inlet duct shape was added on to verify desired pre-swirl system. To find the best model, case study was conducted by CFD approaches to increase the discharge coefficient and the adiabatic effectiveness of the pre-swirl system. New pre-swirl test rig was construct to verify pre-swirl characteristics and to find best pre-swirl nozzles shape. Discharge coefficients of the tested data was relatively higher than 3-D CFD results due to the pre-swirl inlet conditions was relatively fair in test rig compared with real gas turbine conditions.


Author(s):  
Ashutosh Kumar Yadav ◽  
Parantak Sharma ◽  
Avadhesh Kumar Sharma ◽  
Mayank Modak ◽  
Vishal Nirgude ◽  
...  

Impinging jet cooling technique has been widely used extensively in various industrial processes, namely, cooling and drying of films and papers, processing of metals and glasses, cooling of gas turbine blades and most recently cooling of various components of electronic devices. Due to high heat removal rate the jet impingement cooling of the hot surfaces is being used in nuclear industries. During the loss of coolant accidents (LOCA) in nuclear power plant, an emergency core cooling system (ECCS) cool the cluster of clad tubes using consisting of fuel rods. Controlled cooling, as an important procedure of thermal-mechanical control processing technology, is helpful to improve the microstructure and mechanical properties of steel. In industries for heat transfer efficiency and homogeneous cooling performance which usually requires a jet impingement with improved heat transfer capacity and controllability. It provides better cooling in comparison to air. Rapid quenching by water jet, sometimes, may lead to formation of cracks and poor ductility to the quenched surface. Spray and mist jet impingement offers an alternative method to uncontrolled rapid cooling, particularly in steel and electronics industries. Mist jet impingement cooling of downward facing hot surface has not been extensively studied in the literature. The present experimental study analyzes the heat transfer characteristics a 0.15mm thick hot horizontal stainless steel (SS-304) foil using Internal mixing full cone (spray angle 20 deg) mist nozzle from the bottom side. Experiments have been performed for the varied range of water pressure (0.7–4.0 bar) and air pressure (0.4–5.8 bar). The effect of water and air inlet pressures, on the surface heat flux has been examined in this study. The maximum surface heat flux is achieved at stagnation point and is not affected by the change in nozzle to plate distance, Air and Water flow rates.


Author(s):  
Grzegorz Nowak

This paper discusses the problem of cooling system optimization within a gas turbine airfoil regarding to thermo-mechanical behavior of the component, as well as some economical aspects of turbine operation. The main goal of this paper is to show the possibilities of evolutionary approach application to the cooling system optimization. This method, despite its relatively high computational cost, seems to be a valuable tool to such technical problems. The analysis involves the optimization of location and size of internal cooling passages within an airfoil. Initially cooling is provided with circular passages and heat is transported by convection. During the optimization the number of channels can vary. The task is approached in 3D configuration. Each passage is fed with cooling air of constant parameters at the inlet. Also a constant pressure drop is assumed along the passage length. The thermal boundary conditions in passages vary with diameter and local vane temperature (passage wall temperature). The analysis is performed by means of the genetic algorithm for the optimization task and FEM for the heat transfer predictions within the component. In the present study the airfoil profile is taken as aerodynamically optimal and the objective of the search procedure is to find cooling structure variant that at given external conditions provides lower stresses, material temperature and indirectly coolant usage.


Author(s):  
Daniel Lo¨rstad

The main parts of the annular combustor liner walls of the Siemens gas turbine SGT-800 are convectively cooled using rib turbulated cooling. Due to the serial system of cooling and combustion air there is a potential of further reduction of total combustor pressure drop by improvements of the cooling system. Apart from the rib cooling, also the cooling channel bypass entrance is related to a significant part of the total cooling system pressure drop. In this study, an investigation is performed for a rib cooled channel which is related to the considered combustor liner and where empirical correlations are available in order to evaluate the methodology used. The study includes an assessment of the Reynolds Averaged Navier-Stokes (RANS) and Large Eddy Simulation (LES) models available within commercial Computational Fluid Dynamics (CFD) codes and includes also an investigation of model size when using periodic boundaries for LES simulations. It is well known that a small geometrical distance in the direction of the periodic boundaries may have a strong effect on the flow field but is often neglected in practice in order to speed up LES calculations. Here the effect is assessed in order to show what size is required for accurate results, both for time averaged and transient results. In addition too small domains may be affected by spurious low frequencies originating from the periodic boundaries requiring additional simulation time for time converged statistics, but also the averages may be significantly affected. In addition the simulation period for time converged statistics is evaluated in order to show that larger model size in the periodic direction does not necessarily require longer practical simulation time, due to the fact that larger volumes may be used for the combined time and space averaging. The aim is to obtain practical guidelines for LES calculations for internal cooling flows. Then the study is extended step by step to investigate the importance due to high Reynolds number, variable fluid properties and large temperature gradients in order to cover the ranges and specifics required for SGT-800 engine conditions.


2015 ◽  
Vol 51 (12) ◽  
pp. 1779-1790 ◽  
Author(s):  
Hokyu Moon ◽  
Kyung Min Kim ◽  
Jun Su Park ◽  
Beom Seok Kim ◽  
Hyung Hee Cho

2021 ◽  
Vol 75 (4) ◽  
pp. 3-12
Author(s):  
Zayats Yuriy Aleksandrovich ◽  
◽  
Zayats Tatiana Mikhailovna ◽  
Savelyev Maksim Anatolevich ◽  
◽  
...  

Logistics support of products at all stages of the life cycle is gaining increasing influence. This is facilitated by the increasing complexity of structures, a large number of elements, the intro-duction of mechatronic systems. Under these conditions, the relevance of developing methods for analyzing the design of samples increases. The developed model for analyzing the diesel cooling system is based on the principles of cognitive modeling. The practical significance of cognitive models is shown, which consists in the possibility of predicting changes in the influence of system elements on the target function in various operating conditions.


Sign in / Sign up

Export Citation Format

Share Document