← Back to Use Cases

Thermo-mechanical Analysis of a Turbine Disc

Authors:
Name Affiliation Phone Number Email Address
Ron Bates Rolls-Royce PLC. matt.butchers@ktn-uk.org
Industrial Sector:
1. DESCRIPTION OF USE CASE

In modern turbo-machinery, robustness to uncertain operating conditions, as well as geometrical and material variability is vital for performance and operational lifetimes. 
This Use Case concentrates on an adapted Rolls-Royce training example designed to assess the effect of component temperature distributions, stresses, tip clearance etc. on changes to model geometry, material and boundary conditions.

2. KEY UQ&M CONSIDERATIONS
2.1 Process Inputs

The model has been devised to represent a simple gas turbine spool. A single stage drilled rim high-pressure turbine, linked to a 3-stage axial compressor drum is assumed, with an internal cooling air system. As well as boundary conditions representing heat transfer coefficients, mass flows, and heat pick-up terms - a dozen or so (below) - the Use Case aims to investigate the effect of component geometry with hundreds of user defined parameters.
 

Boundary Condition

Field

Factor

Default Value

Streams 22 to 25

HTC

Fhdiaf

1

Stream 26

HTC

Fhpresw

1

Void 5

HTC

Fhcob1

1

Duct 8

HTC

Fhcob2

1

Stream 37 & Void 6

HTC

Fhcob3

1

Streams 60 to 64

HTC

Fhdiar

1

Stream 50, 53 & 57

HTC

Fhdrill

1

Stream 1

Massf

Fm1

1

Stream 6

Massf

Fm6

1

Streams 22 to 25

HPU

Fwdiaf

1

Stream 26

HPU

Fwpresw

1

Stream 60 to 64

HPU

Fwdiar

1

 
Distribution Mean Std Low Mode High Lower Upper
HTCs Triangular 1 0 1 2 0.5 1.5
Mass Flows Normal 1 0.2 0.5 1.5
HPU Triangular 1 0 1 2 0.8 1.2
  HTC = Heat Transfer Coefficients Massf = Mass Flow HPU = Heat Pick-up (per unit area)   Correlations can be inferred by separate studies on air system models - to be confirmed

2.2 Propagation

The simulation is either a combined transient thermo-mechanical analysis or a thermal analysis followed by a series of single time point stress analyses using interpolated temperature distributions.

2.3 Interpretation and Communication of Results

The UQ&M objective is to propagate the uncertainty on various input parameters (conceivably many hundred) through the model to assess their impact on performance. There are benchmark results for some performance measures based on Monte Carlo analysis. The idea would be to contrast these with other methods. Key questions:

  • How does variation in the boundary conditions affect the temperature at the rim (blue star on the below image)? This value (and variation) affects the life of the disc.
  • What methods exist to examine the (potentially) very high dimensional problem the geometry / material variables introduce.
  • What effect do 3D features have on the simplified model
  [caption id="attachment_1278" align="aligncenter" width="527"]Figure 1: Simplified 2D model and rim assessment location Figure 1: Simplified 2D model and rim assessment location[/caption]

3. CURRENT STATE OF MATURITY

This Use Case is an adaption of a 2D training example. The adapted problem will be posed at the Second Uncertainty Quantification and Management Study Group with Industry on the 8th - 10th March 2017.
 
Resources Available for this Problem:
 
  • A 3D parametric model representing a simple gas turbine spool. A single drilled rim HP turbine, linked to a 3-stage axial compressor drum is assumed, with an internal cooling air system similar to the RB211-535E4 engine.
  • Engineering experts from Rolls-Royce able to run the workflow (described above)
  • Data on input uncertainties based on performance decks

Reference: