scholarly journals Inverse Aerodynamic Design of Gas Turbine Blades using Probabilistic Machine Learning

2021 ◽  
pp. 1-16
Author(s):  
Sayan Ghosh ◽  
Govinda Anantha Padmanabha ◽  
Cheng Peng ◽  
Valeria Andreoli ◽  
Steven Atkinson ◽  
...  
Keyword(s):  
Machine Learning ◽  
Gas Turbines ◽  
Turbine Blades ◽  
Three Dimensional ◽  
Inverse Design ◽  
Aerodynamic Design ◽  
Industrial Gas Turbines ◽  
Turbine Blade Design ◽  
The Individual ◽  
Probabilistic Machine Learning

Abstract One of the critical components in Industrial Gas Turbines (IGT) is the turbine blade. Design of turbine blades needs to consider multiple aspects like aerodynamic efficiency, durability, safety and manufacturing, which make the design process sequential and iterative. The sequential nature of these iterations forces a long design cycle time, ranging from several months to years. Due to the reactionary nature of these iterations, little effort has been made to accumulate data in a manner that allows for deep exploration and understanding of the total design space. This is exemplified in the process of designing the individual components of the IGT resulting in a potential unrealized efficiency. To overcome the aforementioned challenges, we demonstrate a probabilistic inverse design machine learning framework, namely PMI (PMI), to carry out an explicit inverse design. PMI calculates the design explicitly without costly iteration and overcomes the challenges associated with ill-posed inverse problems. In this work the framework will be demonstrated on inverse aerodynamic design of three-dimensional turbine blades.

A Probabilistic Machine Learning Framework for Explicit Inverse Design of Industrial Gas Turbine Blades

2021 ◽  
Author(s):  
Sayan Ghosh ◽  
Valeria Andreoli ◽  
Govinda A. Padmanabha ◽  
Cheng Peng ◽  
Steven Atkinson ◽  
...  
Keyword(s):  
Machine Learning ◽  
Gas Turbines ◽  
Turbine Blades ◽  
Inverse Design ◽  
Learning Framework ◽  
Industrial Gas Turbines ◽  
Critical Components ◽  
Turbine Blade Design ◽  
The Individual ◽  
Probabilistic Machine Learning

Abstract One of the critical components in Industrial Gas Turbines (IGT) is the turbine blade. Design of turbine blades needs to consider multiple aspects like aerodynamic efficiency, durability, safety and manufacturing, which make the design process sequential and iterative. The sequential nature of these iterations forces a long design cycle time, ranging from a several months to years. Due to the reactionary nature of these iterations, little effort has been made to accumulate data in a manner that allows for deep exploration and understanding of the total design space. This is exemplified in the process of designing the individual components of the IGT resulting in a potential unrealized efficiency. To overcome the aforementioned challenges, we demonstrate a probabilistic inverse design machine learning framework, namely Pro-ML IDeAS, to carry out an explicit inverse design. Pro-ML IDeAS calculates the design explicitly without costly iteration and overcomes the challenges associated with ill-posed inverse problems. In this work the framework will be demonstrated on inverse aerodynamic design of 2D airfoil of turbine blades.

Improving Efficiency in Robot Assisted Belt Grinding of High Performance Materials

2014 ◽  
Vol 907 ◽  
pp. 139-149 ◽  
Author(s):  
Eckart Uhlmann ◽  
Florian Heitmüller
Keyword(s):  
Gas Turbines ◽  
High Performance ◽  
Scale Effects ◽  
Turbine Blades ◽  
Repair Process ◽  
Industrial Robots ◽  
Robot Assisted ◽  
Belt Grinding ◽  
Economy Of Scale ◽  
The Individual

In gas turbines and turbo jet engines, high performance materials such as nickel-based alloys are widely used for blades and vanes. In the case of repair, finishing of complex turbine blades made of high performance materials is carried out predominantly manually. The repair process is therefore quite time consuming. And the costs of presently available repair strategies, especially for integrated parts, are high, due to the individual process planning and great amount of manually performed work steps. Moreover, there are severe risks of partial damage during manually conducted repair. All that leads to the fact that economy of scale effects remain widely unused for repair tasks, although the piece number of components to be repaired is increasing significantly. In the future, a persistent automation of the repair process chain should be achieved by developing adaptive robot assisted finishing strategies. The goal of this research is to use the automation potential for repair tasks by developing a technology that enables industrial robots to re-contour turbine blades via force controlled belt grinding.

Computational Analysis of Mixing in a Gas Turbine Combustor

2012 ◽  
Author(s):  
Alka Gupta ◽  
Mohamed Saeed Ibrahim ◽  
R. S. Amano
Keyword(s):  
Gas Turbines ◽  
Computational Analysis ◽  
Blunt Body ◽  
Mixture Fraction ◽  
Turbine Blades ◽  
Three Dimensional ◽  
Equilibrium Mixture ◽  
Reacting Flow ◽  
Uniform Temperature ◽  
Combustion Gases

This paper presents the computational analysis of the dilution process involved in gas turbines order to cool the combustion gases to the desired temperature before it enters the turbine. Here, it should be noted that in order to focus only on the dilution process, non-reacting flow conditions were simulated and the complete system was reduced to mixing of a primary (hot) stream and dilution (cold) stream of air. Four different schemes were investigated based on the layout of the dilution holes and use of a blunt body. A complete three dimensional analysis was carried out for each case in order to investigate its effectiveness to produce a more uniform temperature conditions at the exit of the combustor, so as to reduce the detrimental effect these temperature non-uniformities have on the turbine blades. For comparison of the proposed schemes, a parameter is defined in terms of the temperatures of the dilution and primary flow streams at the inlet and the exit plane, called the mixture fraction. Based on this parameter, it was found that the staggered dilution holes with the blunt body has the mixture fraction closest to the equilibrium mixture fraction (0.4), which implies that this scheme with the mixture fraction of 0.36, resulted in best mixing and produced the most uniform temperature distribution at the exit amongst the four proposed schemes.

Some Aspects of Operated Blades HCF Analysis: Rejuvenation and Life Estimation

2018 ◽  
Author(s):  
Sergey A. Ivanov ◽  
Maxim G. Guralnik ◽  
Alexander I. Rybnikov
Keyword(s):  
Shot Peening ◽  
Gas Turbines ◽  
Turbine Blades ◽  
Life Extension ◽  
Ultrasonic Shot Peening ◽  
Industrial Gas Turbines ◽  
Hcf Strength ◽  
Blade Failure ◽  
Metal Properties ◽  
The Cost

The lifecycle of modern industrial gas turbines can reach hundred thousand hours and usually the turbine blades need to be replaced. The use of super alloys and application of advanced coatings makes the cost of turbine lifecycle rather high. The methods for blade rejuvenation and life extension are based on the analysis of the main defects which can considerably reduce blade strength. The effect of long operation and typical defects in turbine blades has been studied in correlation with HCF. The decrease of blades HCF under the effect of operation has been considered as the result of influence of mechanical and thermal factors. The influence of FOD on the blade HCF strength is studied. Some random defects in turbine blades which resulted in HCF decreasing and blade failure are considered. The rejuvenation heat treatment for the blades of ZhS6K and EI893 and its positive effect on metal properties is demonstrated. The ultrasonic shot peening for operated blades have been considered. It is demonstrated that HCF strength of blades after shot peening is about 25–30% higher. Relaxation of compressing stresses in operation is shown as not essential. The remaining life of operated blades can be estimated using the correlation of endurance limit and run time.

An Experimentally Derived Model to Predict the Water Film in a Compressor Cascade With Droplet Laden Flow

2019 ◽  
Vol 141 (9) ◽  
Author(s):  
Niklas Neupert ◽  
Janneck Christoph Harbeck ◽  
Franz Joos
Keyword(s):  
Film Thickness ◽  
Gas Turbines ◽  
Three Dimensional ◽  
Water Film ◽  
Compressor Cascade ◽  
Compressor Blades ◽  
Transonic Compressor ◽  
System A ◽  
Industrial Gas Turbines ◽  
Measured Flow

In recent years, overspray fogging has become a powerful means for power augmentation of industrial gas turbines. Despite the positive thermodynamic effect on the cycle, droplets entering the compressor increase the risk of water droplet erosion and deposition of water on the blades leading to an increase of required torque and profile loss. Due to this, detailed information about the structure and the amount of water on the surface is key for compressor performance. Experiments were conducted with a droplet laden flow in a transonic compressor cascade focusing on the film formed by the deposited water. Two approaches were taken. In the first approach, the film thickness on the blade was directly measured using white light interferometry. Due to significant distortion of the flow caused by the measurement system, a transfer of the measured film thickness to the undisturbed case is not possible. Therefore, a film model is adapted to describe the film flow in terms of height averaged film parameters. In the second approach, experiments were conducted in an undisturbed cascade setup and the water film pattern was measured using a nonintrusive quantitative image processing tool. Utilizing the measured flow pattern in combination with findings from the literature, the rivulet flow structure is resolved. From continuity of the water flow, a film thickness is derived showing good agreement with the previously calculated results. Using both approaches, a three-dimensional (3D) reconstruction of the water film pattern is created giving first experimental results of the film forming on stationary compressor blades under overspray fogging conditions.

Development and Testing of Harsh Environment, Wireless Sensor Systems for Industrial Gas Turbines

2009 ◽  
Author(s):  
David Mitchell ◽  
Anand Kulkarni ◽  
Alex Lostetter ◽  
Marcelo Schupbach ◽  
John Fraley ◽  
...  
Keyword(s):  
High Temperature ◽  
Gas Turbine ◽  
Gas Turbines ◽  
Turbine Blades ◽  
Operating Conditions ◽  
Condition Based Maintenance ◽  
Harsh Environment ◽  
Sensor Systems ◽  
Wireless Telemetry ◽  
Industrial Gas Turbines

The potential for savings provided to worldwide operators of industrial gas turbines, by transitioning from the current standard of interval-based maintenance to condition-based maintenance may be in the hundreds of millions of dollars. In addition, the operational flexibility that may be obtained by knowing the historical and current condition of life-limiting components will enable more efficient use of industrial gas turbine resources, with less risk of unplanned outages as a result of off-parameter operations. To date, it has been impossible to apply true condition-based maintenance to industrial gas turbines because the extremely harsh operating conditions in the heart of a gas turbine preclude using the necessary advanced sensor systems to monitor the machine’s condition continuously. Siemens, Rove Technical Services, and Arkansas Power Electronics International are working together to develop a potentially industry-changing technology to build smart, self-aware engine components that incorporate embedded, harsh-environment-capable sensors and high temperature capable wireless telemetry systems for continuously monitoring component condition in the hot gas path turbine sections. The approach involves embedding sensors on complex shapes, such as turbine blades, embedding wireless telemetry systems in regions with temperatures that preclude the use of conventional silicon-based electronics, and successfully transmitting the sensor information from an environment very hostile to wireless signals. The results presented will include those from advanced, harsh environment sensor and wireless telemetry component development activities. In addition, results from laboratory and high temperature rig and spin testing will be discussed.

Ongoing Development and Recent Results of the Research in the Swedish Gas Turbine Centre (GTC)

2004 ◽  
Author(s):  
Sven Gunnar Sundkvist ◽  
Michael Andersson ◽  
Bogdan Gherman ◽  
Andreas Sveningsson ◽  
Damian Vogt
Keyword(s):  
Gas Turbine ◽  
Gas Turbines ◽  
Industrial Development ◽  
Turbine Blades ◽  
Three Dimensional ◽  
Life Time ◽  
Research Areas ◽  
Verification Methods ◽  
Acoustic Instabilities ◽  
Research Students

This paper describes a way of co-operation between industries, universities and government that has proven to be very fruitful. The Swedish Gas Turbine Centre (GTC) is constituted as a research consortium between technical universities and gas turbine industry. The overall goal of the centre, that was founded in 1996 on a governmental initiative, is to build up a basis of knowledge at Swedish universities to support the industrial development in Sweden of gas turbines of the future with expected requirements on low emissions, high efficiencies, high availability, and low costs. Since the start the research has had a focus on high temperature components of gas turbines (combustion chamber and turbine). This is also reflected in the on-going development phase where the research program consists of four project areas: cooling technology, combustion technology, aeroelasticity, and life time prediction of hot components. The projects are aiming at developing design tools and calculation and verification methods within these fields. A total of eleven research students (among them one industrial PhD student) are active in the centre at present. Numerical analysis as well as experimental verification in test rigs are included. The program has so far produced eleven Licentiate of Engineering and five PhD. On-going activities and recent results of the research in the four research areas are presented: • A new test rig for investigation of time-dependent pressures of three-dimensional features on a vibrating turbine blade at realistic Mach, Reynolds and Strouhal numbers and first experimental results. • Results of numerical simulations of heat loads on turbine blades and vanes, especially platform cooling. • First results of numerical investigations of combustion and thermo-acoustic instabilities in gas turbine chambers. • Experimental investigation of crack propagation in gas turbine materials using the scanning electron microscope (SEM).

Analysis and Prediction of Transonic Turbine Blade Losses

1991 ◽  
Author(s):  
C. L. S. Farn ◽  
D. K. Whirlow ◽  
S. Chen
Keyword(s):  
Boundary Layer ◽  
Turbine Blade ◽  
Turbine Blades ◽  
Three Dimensional ◽  
Variable Approach ◽  
Transonic Turbine ◽  
Steam Turbine Blades ◽  
Turbine Blade Design ◽  
Empirical Approaches ◽  
Mixing Losses

The use of simple computational means to determine the performance of cascades of turbine blades is attractive because it can quickly and economically yield results that can be used for optimization of classes of blades. Fully viscous flow computations are not at the point where they are economical for use in a routine way, and most computational methods lack the resolution to determine shock losses in the transonic flow regime. There is still a need for approaches that combine computation and empiricism. We describe approaches that combine quasi-three dimensional inviscid codes and boundary layer methods for blade passage flow with empirical approaches for shock losses and base pressure deficits to predict the losses in cascades of blades. Downstream mixing losses are handled by a distributed variable approach that uses the inviscid and boundary layer results to determine the distribution of variables at the trailing edge plane. The method gives accurate predictions for the set of distinctly different steam turbine blades for which it was run, and forms the basis for the development of rule-based turbine blade design.

scholarly journals Influence of Rim Seal Geometry on Hot Gas Ingestion Into the Upstream Cavity of an Axial Turbine Stage

1999 ◽  
Author(s):  
Dieter Bohn ◽  
Bernd Rudzinski ◽  
Norbert Sürken ◽  
Wolfgang Gärtner
Keyword(s):  
Mach Number ◽  
Gas Turbines ◽  
Three Dimensional ◽  
Main Flow ◽  
Rotor Blades ◽  
Flow Rates ◽  
Hot Gas ◽  
Carbon Dioxide Gas ◽  
Industrial Gas Turbines ◽  
Rotor Disk

The ingestion of hot gas at the rim seal of a turbine has been investigated for a complete stage with nozzle guide vanes and rotor blades for two types of geometry: 1. the simple axial gap between a flat rotor disk and a flat stator disk, commonly used for industrial gas turbines and 2. an axial lip of the rim seal on the stator combined with a flat rotor disk, often found in aero engine applications. The clearance of the axial gap has been varied for the second type. The efficiency of the rim seal has been examined for different seal flow rates, rotational Reynolds numbers and Mach numbers in the main flow. For the determination of the sealing effectiveness carbon dioxide gas concentration measurements have been carried out in the wheelspace. The distribution of the static pressure in the vicinity of the seal and inside the wheelspace has been measured by means of pressure taps at the stator disk. It is shown that the external flow Mach number in the main flow has a significant effect on the sealing efficiency. As Mach number increases sealing efficiency goes down. The rotational Reynolds number has a distinct effect on the rim seal efficiency depending on the examined configuration. Even for high seal flow rates the ingestion of hot gas can not be fully avoided. The experimental results were the motivation for a three-dimensional CFD approach neglecting the influence of the rotor blades. The results give further insight into aerodynamic features of the ingestion phenomenon.

Turbine Tip Clearance Measurement System Evaluation on an Industrial Gas Turbine

1997 ◽  
Author(s):  
S. J. Gill ◽  
M. D. Ingallinera ◽  
A. G. Sheard
Keyword(s):  
Gas Turbine ◽  
Measurement System ◽  
Gas Turbines ◽  
Turbine Blades ◽  
Tip Clearance ◽  
Real Time Measurement ◽  
Gap Measurement ◽  
Industrial Gas Turbines ◽  
Blade Tip ◽  
Industrial Gas Turbine

The continuing development of industrial gas turbines is resulting in machines of increasing power and efficiency. The need to continue this trend is focusing attention on minimizing all loss mechanisms within the machine, including those associated with turbine blade tip clearance. In order to study tip clearance in the turbine, real time measurement is required of clearance between turbine blades and the casing in which they run. This measurement is not routinely performed, due to the harsh nature of the turbine environment. On those occasions when turbine tip clearance is measured, it is typically in development vehicles, often using cooled probes that are somewhat unsuitable for use in production gas turbines. In this paper a program of work is reported that was undertaken with the purpose of identifying a promising turbine tip clearance measurement system that used the capacitive gap measurement technique. Issues surrounding the application of three systems to the turbine section of a GE MS6001FA gas turbine are identified and reported. Performance of the three evaluated systems is analyzed.

Sign in / Sign up

Export Citation Format

Share Document