*Start of Readme.txt*
Parametric design of scaled-down pressurised thermal shock test specimens using inelastic analysis - Version 1.0
Copyright Harry Coules 2016
This file is best viewed with word wrap on.
This file last modified on 01/08/2016.
%INTRODUCTION
These data are the results from a series of models of steel test specimens undergoing Pressurised Thermal Shock (PTS). This package of results is designed to accompany the following article:
H. E. Coules et al. "Parametric design of scaled-down pressurised thermal shock test specimens using inelastic analysis" in Engeineering Fracture Mechanics.
Finite Element Analysis was used to calculate the (elastic-plastic) strain energy release rate J which occurs at a semi-elliptical defect in each simulated test specimen, as a function of time during a thermal shock event. Mode I stress intensity factor and elastic T-stress were also calculated for an equivalent linear-elastic specimen. Different models with different specimen geometries, angular velocities, and initial temperature parameters were performed. The results of this parametric study were compared with results from a simulation of a previous large-scale thermal shock test known as NESC-1.
%FILES INCLUDED IN THIS DATASET
NESC-1_basic_ElPl.mat
NESC-1_basic_LinEl.mat
NESC-1_MPV_ElPl.mat
NESC-1_MPV_RS_ElPl.mat
parametric_ElPl.mat
parametric_LinEl.mat
Readme.txt
%SOFTWARE
These data files were generated using MATLAB R2015a (The Mathworks Inc., Natick, USA). They should be compatible with any subsequent version of MATLAB (and some previous versions).
%DATA STRUCTURE
1. Files
Six MATLAB .mat files contain the data:
1. NESC-1_basic_ElPl.mat - NESC-1 cylinder, residual stress and mechanical property variation not considered, elastic-plastic material.
2. NESC-1_basic_LinEl.mat - NESC-1 cylinder, residual stress and mechanical property variation not considered, linear elastic material.
3. NESC-1_MPV_ElPl.mat - NESC-1 cylinder, mechanical property variation considered (but not residual stress), elastic-plastic material.
4. NESC-1_MPV_RS_ElPl.mat - NESC-1 cylinder, residual stress and mechanical property variation considered, elastic-plastic material.
5. parametric_ElPl.mat - Parametric series of models, residual stress and mechanical property variation not considered, elastic-plastic material.
6. parametric_LinEl.mat - Parametric series of models, residual stress and mechanical property variation not considered, linear elastic material.
2. Structures found within the .mat files
Each .mat file contains two data structures and a cell array:
paramRangeStruct - This structure was used as input to the routines used to run the models. It indicates the range of
parameters for which models whould be attempted.
paramLogStruct - Contains a log of the parameters used for each model and information about how each model ran.
outputArray - Cell array of structures containing the model results. The number of cells in the array is the same as the number of models run
Further information on paramLogStruct and outputArray follows.
4. paramLogStruct
This structure contains a log of the parameter combinations used for each model, along with information relating to how it was executed. The most important sub-structures are:
paramLogStruct(i).naturalParams - Contains the geometry of the specimen in non-dimensionalised ('natural') form.
paramLogStruct(i).modelParams - Contains the geometry of the crack in the form used to specify it in the finite element model, i.e. in actual spatial coordinates.
paramLogStruct(i).jobExitStatus - Logical to indicate if the (mechanical) finite element model ran successfully. In this study, no numerical problems were encountered in the finite element models.
5. outputArray
Cell array containing the finite element results. The fields include:
outputArray{i,j}.k - Stress intensity factors
outputArray{i,j}.j - J-integral
outputArray{i,j}.t - Elastic T-stress
outputArray{i,j}.stepIncTimes - n-by-6 array with columns indicating: step #, increment #, increment duration (sec.), increment duration (fraction of step), step time, total time.
outputArray{i,j}.k, outputArray{i,j}.j and outputArray{i,j}.t are cell arrays and contain results for different points in time and different locations on the crack tip line. See example below.
outputArray{i,j}.k and outputArray{i,j}.t will be populated for linear elastic models only. outputArray{i,j}.j will be populated for elastic-plastic models only.
%EXAMPLE
Q:
What is the J-integral at the deepest point on a crack in a cylinder with:
Ri/Ro = 0.3, Ro = 200 mm, a/t = 0.4229, a/c = 0.74
Initial angular velocity of 659.73 rad/s (i.e. 6300 rpm)
Initial temperature of 400 deg. C
at a time of 345 s after the start of quenching?
A:
The model with these parameters is #258, which can be verified by inspecting the contents of paramLogStruct(258).naturalParams.
outputArray{1,258}.stepIncTimes shows us that for the final step of the model where quenching occurs, which is Step #3 in this case, 345 sec. occurs at the 32nd increment in the step (see outputArray{1,258}.stepIncTimes(40,:)).
The result (in N/mm) is given by:
>> outputArray{1,258}.j{1,3}{2,32}(end,end)
ans =
88.2900
That is: J for the 258th model, 3rd step, 2nd output type for the 32nd increment, last evaluated contour and last point on the crack tip line (i.e. the deepest point).
nB. In all of the models in parmetric_ElPl.mat, parametric_LinEl.mat and NESC-1_basic_LinEl.mat, nodes are evenly-spaced along the elliptical crack front. In NESC-1_basic_ElPl.mat, NESC-1_MPV_ElPl.mat and NESC-1_MPV_RS_ElPl.mat, which needed to be capable of including the effect of cladding, there are more nodes closer to the innner diameter of the specimen. This can be seen in Figure 4, 8 and 9 of the paper.
%CONTACT
harry.coules@bristol.ac.uk
*End of Readme.txt*