Abstract

Data-driven neural network methods have been widely applied for the prediction of stress–strain behavior, but have proven ill-suited for the extrapolation of time-dependent creep behavior. To overcome this problem, we embedded a physics-based model into feedforward neural networks (FFNNs) to construct a model-guided neural network (MGNN). We proposed a new initialization method for the weights in the model, based on selecting the appropriate physics-based model and activation function, and the resulting MGNN was used for predicting the creep behavior of blade-grooves in a steam turbine rotor under in-service conditions. We compared the performance of the MGNN with baseline methods, namely MGNN0, a FFNN, and a nonlinear autoregressive network with exogenous inputs (network). The results showed that the physics-based model and the neural network in the MGNN complemented each other: the model provided physical relationships to guide the neural network, and the neural network provided stress-fluctuation-tracking for the model. This functionality enabled primary creep behavior to be used as training data for the MGNN, enabling it to predict secondary creep behavior.

References

References
1.
Kassner
,
M. E.
, and
Pérez-Prado
,
M. T.
,
2000
, “
Five-Power-Law Creep in Single Phase Metals and Alloys
,”
Prog. Mater. Sci.
,
45
(
1
), pp.
1
102
.10.1016/S0079-6425(99)00006-7
2.
Yazdani Nezhad
,
H.
, and
O'Dowd
,
N. P.
,
2015
, “
Creep Relaxation in the Presence of Residual Stress
,”
Eng. Fract. Mech.
,
138
, pp.
250
264
.10.1016/j.engfracmech.2015.03.037
3.
Wang
,
Y. Q.
,
Spindler
,
M. W.
,
Truman
,
C. E.
, and
Smith
,
D. J.
,
2016
, “
Critical Analysis of the Prediction of Stress Relaxation From Forward Creep of Type 316H Austenitic Stainless Steel
,”
Mater. Des.
,
95
, pp.
656
668
.10.1016/j.matdes.2016.01.118
4.
Ge
,
Z.
,
2017
, “
Review on Data-Driven Modeling and Monitoring for Plant-Wide Industrial Processes
,”
Chemom. Intell. Lab. Syst.
,
171
, pp.
16
25
.10.1016/j.chemolab.2017.09.021
5.
Carney
,
C.
, and
Gilman
,
T.
,
2011
, “
Stress-Based Fatigue Monitoring: Methodology for Fatigue Monitoring of Class 1 Nuclear Components in a Reactor Water Environment
,” EPRI, Palo Alto, CA, Report No.
1022876
.https://www.nrc.gov/docs/ML1200/ML12004A053.pdf
6.
Trumpler
,
W. E.
,
1941
, “
Relaxation of Metals at High Temperatures
,”
J. Appl. Phys.
,
12
(
3
), pp.
248
253
.10.1063/1.1712902
7.
Harmathy
,
T. Z.
,
1967
, “
A Comprehensive Creep Model
,”
ASME J. Basic Eng.
,
89
(
3
), pp.
496
502
.10.1115/1.3609648
8.
May
,
D. L.
,
Gordon
,
A. P.
, and
Segletes
,
D. S.
,
2013
, “
The Application of the Norton-Bailey Law for Creep Prediction Through Power Law Regression
,”
ASME
Paper No. GT2013-96008. 10.1115/GT2013-96008
9.
Feltham
,
P.
,
1961
, “
Creep and Stress Relaxation in Alpha-Brass at Low Temperatures
,”
Philos. Mag.
,
6
(
62
), pp.
259
270
.10.1080/14786436108243315
10.
Miller
,
A. K.
,
1987
,
Unified Constitutive Equations for Creep and Plasticity
,
Springer
, Dordrecht,
The Netherlands
.
11.
Wang
,
W. Z.
,
Buhl
,
P.
, and
Klenk
,
A.
,
2015
, “
A Unified Viscoplastic Constitutive Model With Damage for Multi-Axial Creep-Fatigue Loading
,”
Int. J. Damage Mech.
,
24
(
3
), pp.
363
382
.10.1177/1056789514537007
12.
Benaarbia
,
A.
,
Rouse
,
J. P.
, and
Sun
,
W.
,
2018
, “
A Thermodynamically-Based Viscoelastic-Viscoplastic Model for the High Temperature Cyclic Behaviour of 9–12% Cr Steels
,”
Int. J. Plast.
,
107
, pp.
100
121
.10.1016/j.ijplas.2018.03.015
13.
Abdul Ghafir
,
M. F.
,
Li
,
Y. G.
, and
Wang
,
L.
,
2013
, “
Creep Life Prediction for Aero Gas Turbine Hot Section Component Using Artificial Neural Networks
,”
ASME J. Eng. Gas Turbines Power
,
136
(
3
), p.
031504
.10.1115/1.4025725
14.
Yoo
,
Y. S.
,
Jo
,
C. Y.
, and
Jones
,
C. N.
,
2002
, “
Compositional Prediction of Creep Rupture Life of Single Crystal Ni Base Superalloy by Bayesian Neural Network
,”
Mater. Sci. Eng. A
,
336
(
1–2
), pp.
22
29
.10.1016/S0921-5093(01)01965-7
15.
Frolova
,
O.
,
Roos
,
E.
,
Maile
,
K.
, and
Müller
,
W.
,
2011
, “
Representation of the Heat Specific Creep Rupture Behaviour of 9% Cr Steels Using Neural Networks
,”
Trans. Mach. Learn. Data Min.
,
4
(
1
), pp.
1
16
.http://www.ibai-publishing.org/journal/issue_mldm/2011_july/mldm_4_1_1-16.php
16.
Alrashydah
,
E. I.
, and
Abo-Qudais
,
S. A.
,
2018
, “
Modeling of Creep Compliance Behavior in Asphalt Mixes Using Multiple Regression and Artificial Neural Networks
,”
Constr. Build. Mater.
,
159
, pp.
635
641
.10.1016/j.conbuildmat.2017.10.132
17.
Dominiczak
,
K.
,
Rządkowski
,
R.
,
Radulski
,
W.
, and
Szczepanik
,
R.
,
2015
, “
Online Prediction of Temperature and Stress in Steam Turbine Components Using Neural Networks
,”
ASME J. Eng. Gas Turbines Power
,
138
(
5
), p.
052606
.10.1115/1.4031626
18.
Liang
,
T.
,
Liu
,
X.
,
Fan
,
P.
,
Zhu
,
L.
,
Bi
,
Y.
, and
Zhang
,
Y.
,
2020
, “
Prediction of Long-Term Creep Life of 9Cr–1Mo–V–Nb Steel Using Artificial Neural Network
,”
Int. J. Pressure Vessel Piping
,
179
, p.
104014
.10.1016/j.ijpvp.2019.104014
19.
Mazur
,
Z.
, and
Hernandez-Rossette
,
A.
,
2015
, “
Steam Turbine Rotor Discs Failure Evaluation and Repair Process Implementation
,”
Eng. Failure Anal.
,
56
, pp.
545
554
.10.1016/j.engfailanal.2014.10.010
20.
Hong
,
H.
,
Wang
,
W.
, and
Liu
,
Y.
,
2019
, “
High-Temperature Fatigue Behavior of a Steam Turbine Rotor Under Flexible Operating Conditions With Variable Loading Amplitudes
,”
Int. J. Mech. Sci.
,
163
, pp.
105
121
.10.1016/j.ijmecsci.2019.105121
21.
Wang
,
W. Z.
,
Buhl
,
P.
,
Klenk
,
A.
, and
Liu
,
Y. Z.
,
2016
, “
Study of Creep–Fatigue Behavior in a 1000 MW Rotor Using a Unified Viscoplastic Constitutive Model With Damage
,”
Int. J. Damage Mech.
,
25
(
2
), pp.
178
202
.10.1177/1056789515576766
22.
Wang
,
W. Z.
,
Buhl
,
P.
,
Klenk
,
A.
, and
Liu
,
Y. Z.
,
2016
, “
The Effect of in-Service Steam Temperature Transients on the Damage Behavior of a Steam Turbine Rotor
,”
Int. J. Fatigue
,
87
, pp.
471
483
.10.1016/j.ijfatigue.2016.02.040
23.
Rumelhart
,
D.
,
Hinton
,
G.
, and
Williams
,
R.
,
1985
, “
Learning Internal Representations by Error Propagation
,” Institute for Cognitive Science, Boulder, CO, Report No.
8506
.https://web.stanford.edu/class/psych209a/ReadingsByDate/02_06/PDPVolIChapter8.pdf
24.
Nair
,
V.
, and
Hinton
,
G. E.
,
2010
, “
Rectified Linear Units Improve Restricted Boltzmann Machines
,”
Proceedings of the 27th International Conference on Machine Learning
, Haifa, Isreal, June 21–24, pp.
1
8
. https://www.cs.toronto.edu/~fritz/absps/reluICML.pdf
25.
Maas
,
A. L.
,
Hannun
,
A. Y.
, and
Ng
,
A. Y.
,
2013
, “
Rectifier Nonlinearities Improve Neural Network Acoustic Models
,”
Proceedings of the 30th International Conference on Machine Learning
, Atlanta, GA, June 16–21, pp.
1
6
.https://ai.stanford.edu/~amaas/papers/relu_hybrid_icml2013_final.pdf
26.
He
,
K.
,
Zhang
,
X.
,
Ren
,
S.
, and
Sun
,
J.
,
2015
, “
Delving Deep Into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification
,”
Proceedings of the IEEE International Conference on Computer Vision
, Santiago, Chile, Dec. 13–-16, pp.
1026
1034
.10.1109/ICCV.2015.123
You do not currently have access to this content.