Abstract

A gas turbine trip is an unplanned shutdown, of which the most relevant consequences are business interruption and a reduction of equipment remaining useful life. Thus, understanding the underlying causes of gas turbine trip would allow predicting its occurrence in order to maximize gas turbine profitability and improve its availability. In the ever competitive Oil & Gas sector, data mining and machine learning are increasingly being employed to support a deeper insight and improved operation of gas turbines. Among the various machine learning tools, random forests are an ensemble learning method consisting of an aggregation of decision tree classifiers. This paper presents a novel methodology aimed at exploiting information embedded in the data and develops random forest models, aimed at predicting gas turbine trip based on information gathered during a timeframe of historical data acquired from multiple sensors. The novel approach exploits time series segmentation to increase the amount of training data, thus reducing overfitting. First, data are transformed according to a feature engineering methodology developed in a separate work by the same authors. Then, Random Forest models are trained and tested on unseen observations to demonstrate the benefits of the novel approach. The superiority of the novel approach is proved by considering two real-word case studies, involving field data taken during three years of operation of two fleets of gas turbines located in different regions. The novel methodology allows values of precision, recall and accuracy in the range 75–85%, thus demonstrating the industrial feasibility of the predictive methodology.

References

1.
Lu
,
F.
,
Ju
,
H.
, and
Huang
,
J.
,
2016
, “
An Improved Extended Kalman Filter With Inequality Constraints for Gas Turbine Engine Health Monitoring
,”
Aerosp. Sci. Technol.
,
58
, pp.
36
47
.10.1016/j.ast.2016.08.008
2.
Qin
,
S. J.
, and
Chiang
,
L. H.
,
2019
, “
Advances and Opportunities in Machine Learning for Process Data Analytics
,”
Comput. Chem. Eng.
,
126
(
2019
), pp.
465
473
.10.1016/j.compchemeng.2019.04.003
3.
De Giorgi
,
M. G.
,
Campilongo
,
S.
, and
Ficarella
,
A.
,
2018
, “
A Diagnostics Tool for Aero-Engines Health Monitoring Using Machine Learning Technique
,”
Energy Procedia
,
148
(
2018
), pp.
860
867
.10.1016/j.egypro.2018.08.109
4.
Zhong
,
S.
,
Fu
,
S.
, and
Lin
,
L.
,
2019
, “
A Novel Gas Turbine Fault Diagnosis Method Based on Transfer Learning With CNN
,”
Measurement
,
137
(
2019
), pp.
435
453
.10.1016/j.measurement.2019.01.022
5.
Amozegar
,
M.
, and
Khorasani
,
K.
,
2016
, “
An Ensemble of Dynamic Neural Network Identifiers for Fault Detection and Isolation of Gas Turbine Engines
,”
Neural Networks
,
76
(
2016
), pp.
106
121
.10.1016/j.neunet.2016.01.003
6.
Bai
,
M.
,
Liu
,
J.
,
Chai
,
J.
,
Zhao
,
X.
, and
Yu
,
D.
,
2020
, “
Anomaly Detection of Gas Turbines Based on Normal Pattern Extraction
,”
Appl. Therm. Eng.
,
166
(
2020
), p.
114664
.10.1016/j.applthermaleng.2019.114664
7.
Naderi
,
E.
, and
Khorasani
,
K.
,
2018
, “
Data-Driven Fault Detection, Isolation and Estimation of Aircraft Gas Turbine Engine Actuator and Sensors
,”
Mech. Syst. Signal Process.
,
100
(
2018
), pp.
415
438
.10.1016/j.ymssp.2017.07.021
8.
Taylor
,
J. V.
,
Conduit
,
B.
,
Dickens
,
A.
,
Hall
,
C.
,
Hillel
,
M.
, and
Miller
,
R. J.
,
2019
, “
Predicting the Operability of Damaged Compressors Using Machine Learning
,”
ASME
Paper No. GT2019-91339.10.1115/GT2019-91339
9.
Wong
,
P. K.
,
Yang
,
Z.
,
Vong
,
C. M.
, and
Zhong
,
J.
,
2014
, “
Real Time Fault Diagnosis for Gas Turbine Generator System Using Extreme Learning Machine
,”
Neurocomputing
,
128
(
2014
), pp.
249
257
.10.1016/j.neucom.2013.03.059
10.
Zhao
,
Y.
,
Huang
,
G.
,
Hu
,
Q.
,
Tan
,
J.
,
Wang
,
J.
, and
Yang
,
Z.
,
2019
, “
Soft Extreme Learning Machine for Fault Detection of Aircraft Engine
,”
Aerosp. Sci. Technol.
,
91
, pp.
70
81
.10.1016/j.ast.2019.05.021
11.
Yan
,
W.
,
2006
, “
Application of Random Forest to Aircraft Engine Fault Diagnosis
,”
The Proceedings of the Multiconference on Computational Engineering in Systems Applications
, Vol.
1
,
IEEE
, Beijing, China, Oct. 4–6, pp.
468
475
.10.1109/CESA.2006.4281698
12.
Maragoudakis
,
M.
,
Loukis
,
E.
, and
Pantelides
,
P. P.
,
2008
, “
Random Forests Identification of Gas Turbine Faults
,”
19th International Conference on Systems
Engineering,
IEEE
, Las Vegas, NV, Aug. 19–21, pp.
127
132
.10.1109/ICSEng.2008.81
13.
Wang
,
C.
,
Lu
,
N.
,
Cheng
,
Y.
, and
Jiang
,
B.
,
2018
, “
Deep Forest Based Multivariate Classification for Diagnostic Health Monitoring
,” Chinese Control and Decision Conference (
CCDC
), Shenyang, Shenyang, China, June 9–11, pp.
6233
6238
.10.1109/CCDC.2018.8408224
14.
Batayev
,
N.
,
2018
, “
Gas Turbine Fault Classification Based on Machine Learning Supervised Techniques
,” 14th International Conference on Electronics Computer and Computation (
ICECCO
), Kaskelen, Kazakhstan, Nov. 29–Dec. 1, pp.
206
212
.10.1109/ICECCO.2018.8634719
15.
Di Maio
,
F.
,
Zio
,
E.
,
Pecht
,
M.
,
Tse
,
P.
, and
Tsui
,
K.
,
2011
, “
Ensemble of Unsupervised Fuzzy C-Means Classifiers for Clustering Health Status of Oil Sand Pumps
,”
ESREL
, Troyes, France, Sept. 18–22, pp.
419
427
.10.1201/b11433-60
16.
Baraldi
,
P.
,
Di Maio
,
F.
,
Rigamonti
,
M.
,
Zio
,
E.
, and
Seraoui
,
R.
,
2015
, “
Clustering for Unsupervised Fault Diagnosis in Nuclear Turbine Shut-Down Transients
,”
Mech. Syst. Signal Process.
,
58–59
(
2015
), pp.
160
178
.10.1016/j.ymssp.2014.12.018
17.
Al-Dahidi
,
S.
,
Di Maio
,
F.
,
Baraldi
,
P.
,
Zio
,
E.
, and
Seraoui
,
R.
,
2018
, “
A Framework for Reconciliating Data Clusters From a Fleet of Nuclear Power Plants Turbines for Fault Diagnosis
,”
Appl. Soft Comput.
,
69
(
2018
), pp.
213
231
.10.1016/j.asoc.2018.04.044
18.
Wang
,
Z.
,
Zhao
,
N.
,
Wang
,
W.
,
Tang
,
R.
, and
Li
,
S.
,
2015
, “
A Fault Diagnosis Approach for Gas Turbine Exhaust Gas Temperature Based on Fuzzy C-Means Clustering and Support Vector Machine
,”
Math. Probl. Eng.
, 2015, Article ID. 240267, p.
11
.10.1155/2015/240267
19.
Hora Fontes
,
C.
, and
Pereira
,
O.
,
2016
, “
Pattern Recognition in Multivariate timeseries - A Case Study Applied to Fault Detection in a Gas Turbine
,”
Eng. Appl. Artif. Intell.
,
49
(
2016
), pp.
10
18
.10.1016/j.engappai.2015.11.005
20.
Hora Fontes
,
C.
, and
Budman
,
H.
,
2017
, “
A Hybrid Clustering Approach for Multivariate Time series - A Case Study Applied to Failure Analysis in a Gas Turbine
,”
ISA Trans.
,
71
(
2017
), pp.
513
529
.10.1016/j.isatra.2017.09.004
21.
Losi
,
E.
,
Venturini
,
M.
,
Manservigi
,
L.
,
Ceschini
,
G. F.
,
Bechini
,
G.
,
Cota
,
G.
, and
Riguzzi
,
F.
,
2020
, “
Structured Methodology for clustering gas turbine Transients by Means of Multi-Variate Time Series
,”
ASME
Paper No. GT2020-14751.10.1115/GT2020-14751
22.
Bhargava
,
R. K.
,
2017
,
Technical Dictionary on the Gas Turbine Technology
,
Innovative Turbomachinery Technologies Corp
,
Katy, TX
.
23.
Morini
,
M.
,
Pinelli
,
M.
, and
Venturini
,
M.
,
2007
, “
Development of a One-Dimensional Modular Dynamic Model for the Simulation of Surge in Compression Systems
,”
ASME J. Turbomach.
,
129
(
3
), pp.
437
447
.10.1115/1.2447757
24.
Morini
,
M.
,
Pinelli
,
M.
, and
Venturini
,
M.
,
2009
, “
Analysis of Biogas Compression System Dynamics
,”
Appl. Energy
,
86
(
11
), pp.
2466
2475
.10.1016/j.apenergy.2009.03.008
25.
Pezzini
,
P.
,
Tucker
,
D.
, and
Traverso
,
A.
,
2013
, “
Avoiding Compressor Surge During Emergency Shut-Down Hybrid Turbine Systems
,”
ASME
Paper No. GT2013-94810.10.1115/GT2013-94810
26.
Kurz
,
R.
,
Garceau
,
S.
,
Ji
,
M.
, and
Brun
,
K.
,
2019
, “
Compressor Speed Decay During Emergency Shutdowns
,”
ASME
Paper No. GT2019-90020.10.1115/GT2019-90020
27.
Losi
,
E.
,
Venturini
,
M.
,
Manservigi
,
L.
,
Ceschini
,
G. F.
,
Bechini
,
G.
,
Cota
,
G.
, and
Riguzzi
,
F.
,
2021
, “
Data Selection and Feature Engineering for the Application of Machine Learning to the Prediction of Gas Turbine Trip
,”
ASME
Paper No. GT2021-58914.10.1115/GT2021-58914
28.
Louppe
,
G.
, and
Geurts
,
P.
,
2012
, “
Ensembles on Random Patches
,”
Joint European Conference on Machine Learning and Knowledge Discovery in Databases
,
Springer
,
Berlin, Heidelberg
.
29.
Ho
,
T. K.
,
1998
, “
The Random Subspace Method for Constructing Decision Forests
,”
IEEE Trans. Pattern Anal. Machine Intell.
,
20
(
8
), pp.
832
844
.
30.
Breiman
,
L.
,
1996
, “
Bagging Predictors
,”
Mach. Learn.
,
24
(
2
), pp.
123
140
.10.1007/BF00058655
31.
Breiman
,
L.
,
2001
, “
Random Forest
,”
Mach. Learn.
,
45
(
1
), pp.
5
32
.10.1023/A:1010933404324
32.
Ho
,
T. K.
,
1995
, “
Random Decision Forests
,”
Proceedings of Third International Conference on Document Analysis and Recognition
, Vol.
1
.
IEEE
, Montreal, QC, Canada, Aug. 14–16, pp.
278
282
.10.1109/ICDAR.1995.598994
33.
Breiman
,
L.
,
Friedman
,
J. H.
,
Olshen
,
R. A.
, and
Stone
,
C. J.
,
1984
,
Classification and Regression Trees
,
Wadsworth & Brooks/Cole Advanced Books & Software
,
Monterey, CA
.
34.
Rokach
,
L.
, and
Maimon
,
O.
,
2014
,
Data Mining With Decision Trees: Theory and Applications
, 2nd ed.,
World Scientific Publishing
, Hackensack, NJ.
35.
Domingos
,
P.
,
2012
, “
A Few Useful Things to Know About Machine Learning
,”
Commun. ACM
,
55
(
10
), pp.
78
87
.10.1145/2347736.2347755
36.
Kuhn
,
M.
, and
Johnson
,
K.
,
2019
,
Feature Engineering and Selection: A Practical Approach for Predictive Models
,
CRC Press
, London, UK.
You do not currently have access to this content.