Advanced

Validation of one-dimensional loss models for axial gas turbines

Hirsch, Kim LU and Moghadam, Ramin LU (2016) MVK920 20161
Department of Energy Sciences
Abstract (Swedish)
Denna masteruppsats är ett sammarbete mellan Siemens Industrial Turbomachi-
nery (SIT) och Lunds Tekniska Högskola (LTH). Uppsatsen berör validering av
ett nytt in-house verktyg på SIT som heter ”Mean Line Tool” (MLT), framför allt
sekundärförlustmodellerna som verktyget använder. Valideringen gjordes genom
att undersöka och jämföra predikteringar från MLT med experiment utförda på en
testturbin vid Kungliga Tekniska Högskolan (KTH). Målet med examensarbetet
var att identifiera de bakomliggande orsakerna till de dåliga predikteringarna. För
att bredda vidden av denna undersökning så har förlustmodellerna som används i
MLT jämförts med modeller i öppen litteratur.
Experimenten utfördes på en enstegsturbin på KTH av Ph.D. student... (More)
Denna masteruppsats är ett sammarbete mellan Siemens Industrial Turbomachi-
nery (SIT) och Lunds Tekniska Högskola (LTH). Uppsatsen berör validering av
ett nytt in-house verktyg på SIT som heter ”Mean Line Tool” (MLT), framför allt
sekundärförlustmodellerna som verktyget använder. Valideringen gjordes genom
att undersöka och jämföra predikteringar från MLT med experiment utförda på en
testturbin vid Kungliga Tekniska Högskolan (KTH). Målet med examensarbetet
var att identifiera de bakomliggande orsakerna till de dåliga predikteringarna. För
att bredda vidden av denna undersökning så har förlustmodellerna som används i
MLT jämförts med modeller i öppen litteratur.
Experimenten utfördes på en enstegsturbin på KTH av Ph.D. student Johan Dahlqvist.
I denna masteruppsats har två olika steg analyserats, steg 4b och steg 5. Steg 4b
är ett lågreaktionssteg designat för att efterlikna typiska SIT-ångturbiner. Steg 5 är
modellerat efter en typisk SIT-gasturbin har därför en högre reaktionsgrad.
Den övergripande slutsatsen är att den predikterade sekundärförlusten från MLT är
kraftigt underpredikterad. Detta tros vara en effekt av en för svag inverkan av höjd-
breddförhållande. En modell med ett modernare tillvägagångssätt som tar hänsyn
till gränsskikt ser ut att vara bättre på att prediktera förlusterna, men vilken testad
modell som är bäst har inte kunnat fastställas. Ytterligare en slutsats är att förlust-
modellerna i MLT är överkänsliga för fel anströmningsvinkel. (Less)
Abstract
This thesis for the degree of Master of Science is a collaboration between Siemens
Industrial Turbomachinery (SIT) and Lund Institute of Technology (LTH). This
thesis concerns validation of a new SIT in-house code Mean Line Tool (MLT) and
especially the secondary loss models used by this tool. The validation was made
by investigating and comparing predictions from MLT with experiments carried
out on a test turbine rig at the Royal Institute of Technology (KTH). The main goal
of this investigation was to identify sources of bad prediction in order to find ar-
eas in need of improvement. To broaden the scope of this investigation, secondary
loss model used in MLT were also compared to models used in other SIT in-house
codes as well... (More)
This thesis for the degree of Master of Science is a collaboration between Siemens
Industrial Turbomachinery (SIT) and Lund Institute of Technology (LTH). This
thesis concerns validation of a new SIT in-house code Mean Line Tool (MLT) and
especially the secondary loss models used by this tool. The validation was made
by investigating and comparing predictions from MLT with experiments carried
out on a test turbine rig at the Royal Institute of Technology (KTH). The main goal
of this investigation was to identify sources of bad prediction in order to find ar-
eas in need of improvement. To broaden the scope of this investigation, secondary
loss model used in MLT were also compared to models used in other SIT in-house
codes as well as models available in the open literature.
The Experiments were carried out on a single stage test rig at KTH by Johan
Dahlqvist. In this thesis two different stages have been analyzed, stage 4b and
stage 5. Stage 4b is a low reaction stage modelled after typical SIT steam turbine.
Stage 5 modelled after a typical SIT gas turbine and therefore has a significantly
higher degree of reaction.
The overall conclusion is that the secondary loss model used by MLT is greatly
under predicting the secondary losses. It is believed that this is caused by a weak
dependence on aspect ratio and that a more modern approach with models utilizing
boundary layers should be more accurate, but no best alternative model could be
determined. In addition, it has also been concluded that there is an over-sensitivity
to incidence in the codes used in MLT. (Less)
Please use this url to cite or link to this publication:
author
Hirsch, Kim LU and Moghadam, Ramin LU
supervisor
organization
course
MVK920 20161
year
type
H2 - Master's Degree (Two Years)
subject
keywords
gas turbine, loss model, turbine
report number
ISRN LUTMDN/TMHP-16/5362-SE
ISSN
0282-1990
language
English
id
8885275
date added to LUP
2016-06-27 14:02:48
date last changed
2016-06-27 14:02:48
@misc{8885275,
  abstract     = {This thesis for the degree of Master of Science is a collaboration between Siemens
Industrial Turbomachinery (SIT) and Lund Institute of Technology (LTH). This
thesis concerns validation of a new SIT in-house code Mean Line Tool (MLT) and
especially the secondary loss models used by this tool. The validation was made
by investigating and comparing predictions from MLT with experiments carried
out on a test turbine rig at the Royal Institute of Technology (KTH). The main goal
of this investigation was to identify sources of bad prediction in order to find ar-
eas in need of improvement. To broaden the scope of this investigation, secondary
loss model used in MLT were also compared to models used in other SIT in-house
codes as well as models available in the open literature.
The Experiments were carried out on a single stage test rig at KTH by Johan
Dahlqvist. In this thesis two different stages have been analyzed, stage 4b and
stage 5. Stage 4b is a low reaction stage modelled after typical SIT steam turbine.
Stage 5 modelled after a typical SIT gas turbine and therefore has a significantly
higher degree of reaction.
The overall conclusion is that the secondary loss model used by MLT is greatly
under predicting the secondary losses. It is believed that this is caused by a weak
dependence on aspect ratio and that a more modern approach with models utilizing
boundary layers should be more accurate, but no best alternative model could be
determined. In addition, it has also been concluded that there is an over-sensitivity
to incidence in the codes used in MLT.},
  author       = {Hirsch, Kim and Moghadam, Ramin},
  issn         = {0282-1990},
  keyword      = {gas turbine,loss model,turbine},
  language     = {eng},
  note         = {Student Paper},
  title        = {Validation of one-dimensional loss models for axial gas turbines},
  year         = {2016},
}