The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
11.neural network precept diagnosis on petrochemical pipelines for quality maintenance
1. Control Theory and Informatics www.iiste.org
ISSN 2224-5774 (print) ISSN 2225-0492 (online)
Vol 2, No.2, 2012
Neural Network Precept Diagnosis on Petrochemical Pipelines
for Quality Maintenance
S.Bhuvaneswari1* R.Hemachandran2 R.Vignashwaran3
1. Reader, Department of Computer Science, Pondicherry University, Karaikal Campus, Karaikal
2. Faculty, N.I.T, Puducherry
3. Scholar, Department of Computer Science, Amrita University, Coimbatore
* E-mail of the corresponding author: booni_67@yahoo.co.in
Abstract
Pipeline tubes are part of vital mechanical systems largely used in petrochemical industries. They serve to
transport natural gases or liquids. They are cylindrical tubes and are submitted to the risks of corrosion due
to high PH concentrations of the transported liquids in addition to fatigue cracks. Due to the nature of their
function, they are subject to the alternation of pressure-depression along the time, initiating therefore in the
tubes’ body micro-cracks that can propagate abruptly to lead to failure by fatigue. On to the diagnostic
study for the issue the development of this prognostic process employing neural network for such systems
bounds to the scope of quality maintenance.
Keywords: Percept, Simulated results, Fluid Mechanics
1. Introduction
The pipelines tubes are manufactured as cylindrical tubes of radius R and thickness e. The failure by
fatigue is caused by the fluctuation of pressure-depression along the time t ( 0 ≤ P ≤ P0). These pipelines are
unfortunately usually designed for ultimate limits states (resistance).To be more realistic, a prognostic
model is proposed here based on analytic laws of degradation by fatigue (Paris’ law) in addition to the
cumulative law of damage (Miner’s law).This prognostic model is crucial in petrochemical industries for
the reason of favorable economic and availability consequences on the exploitation cost .
Fig. 1: Internal pressure diagram.
2. Paris Law
The Paris’ law allows determining the propagation speed of the cracks da/ dN at the time of their
5
2. Control Theory and Informatics www.iiste.org
ISSN 2224-5774 (print) ISSN 2225-0492 (online)
Vol 2, No.2, 2012
da
detection: = C.(∆K ) m where a is the crack length, N is the number of cycles, C and m are the Paris
dN
constants, and ∆K is the stress intensity factor.
We can distinguish:
- The long cracks that obey to Paris law
- The short cracks that serve to decrease the speed of propagation
- The short physical cracks that serve to increase the speed of propagation
da
The law can be written also as :
log = log C + m log(∆K )
dN
da
log
dN
Phase I
Phase II Final fracture
Low speed of
Stable Kc
propagation
propagation
Phase III
High speed of
propagation
da
= C (∆K )
m
dN
log (∆K )
Threshold
∆Kth
Fig. 2: The three phases of cracks growth, Paris’ law.
3. Pipelines under Pressure
A tube is considered thin when its thickness is of the order of one tenth of its radius: e ≤R/10
6
3. Control Theory and Informatics www.iiste.org
ISSN 2224-5774 (print) ISSN 2225-0492 (online)
Vol 2, No.2, 2012
Fig. 3: Cylindrical pipelines
Fig. 4: Stress type distribution
4. State of Stresses
Te tubes are cylindrical shells of revolution. when thin tubes of radius r and of thickness e are under
internal pressure p, the state of stresses is membrane-like under bending loads. the membrane stresses are
circumferential (hoop stress) σθ and longitudinal stresses (axial stress) σL
Fig. 5: Axial stresses and Hoop stresses in cylindrical pipelines
PR
σ θ = e
PR
These stresses are given by: σ L =
2e
The critical cracks are those which are perpendicular to maximal stressesσθ, that means longitudinal
cracks which are parallel to the axis of the tube. A crack is of depth a or of length a, if we measure in the
direction of the tube thickness e. Normally the ratio a/e is within the following range: 0.1 ≤ a/e ≤0.99
7
4. Control Theory and Informatics www.iiste.org
ISSN 2224-5774 (print) ISSN 2225-0492 (online)
Vol 2, No.2, 2012
Fig. 6: Crack length in radial view
Fig. 7: Cracked pipeline section
The stress intensity factor KI represents the effect of stress concentration in the presence of a flat crack.
Fig. 8: Non-uniform distribution of stresses near the crack
The stress intensity factor is given [6] by:
K I = y (a ) × πa σ θ
⇒ K I = 0.6 × g (a )× πa × P.
R
≤ K IC
e
with Y (a ) = 0.6 × g (a ) : is the geometric factor ;
a
1 + 2
g (a ) = e J IC ⋅ E
3 K IC =
a 2 1 − (ν ) 2
1 − 8
e
5. Control Theory and Informatics www.iiste.org
ISSN 2224-5774 (print) ISSN 2225-0492 (online)
Vol 2, No.2, 2012
K IC : is the tenacity of material (critical stress intensity factor) and is given by:
J IC ⋅ E
Note that the factor KI must not exceed the value of KIC . K IC =
1 − (ν ) 2
5. Proposed Percept Model
Consider a pipeline of radius R = 240 mm and of thickness e = 8 mm transporting natural gases, the
parameters related to materials and to the environment are taken as being equal to
: [5] m= 3 et C = ε = 5.2.10 −13
e e
The length of the crack is denoted by a with an initial value a 0 = 0.2 mm a0 ≤ a ≤ a N = ⇒ =8
8 aN
We have to respect the following ratio:
a e
0 .1 ≤ ≤ 0.99 ⇒ 1.01 ≤ ≤ 10
e a
Take a similar form to
da
as a = εφ 1 ( a ) φ 2 ( p )
&
dN
with: ε = C ; (
φ1 (a) = Y (a ) π a ) m
; p = ∆σ and
φ 2 ( p) = p m = (∆σ)m
The initial damage is: a(0) = a0
A recurrent form of crack length gives:
a i = εφ1 ( a i −1 ) φ 2 ( p i ) + a i −1
And the corresponding degradation is given by:
Di = Di −1 + ηφ1 ( Di −1 )φ 2 ( pi )
for m = 3 ⇒ φ 2 ( p i ) = p i3 = (∆ σ θ i )3
ε
Morevor η =
a N − a0
9
6. Control Theory and Informatics www.iiste.org
ISSN 2224-5774 (print) ISSN 2225-0492 (online)
Vol 2, No.2, 2012
da j
dj =
We define the damage fraction by:
aN − a0
Therefore, we get the cumulated total damage:
i
i i da j
∑ da j ai
j =1
Di = ∑ d j = ∑ = =
j =1 j =1 a N − a0 a N − a0 a N − a0
N
DN = ∑ d j = 1
We can easily prove that:
j =1
D
Failure
1
Reliable
ni/Ni
0
Fig. 9: Miner’s law of damage
where :
0 ≤ n ≤ N , a0 ≤ a ≤ a N ;
N
D0 ≤ D ≤ 1 = D N ; D N = ∑ d j = 1
j =1
a0 D a
D0 = ⇒ a0 = 0 N
a N − a0 1 + D0
10
7. Control Theory and Informatics www.iiste.org
ISSN 2224-5774 (print) ISSN 2225-0492 (online)
Vol 2, No.2, 2012
The other sequences are :
a0
D0 =
a N − a0
a1
D1 =
a N − a0
a2
D2 =
a N − a0
M
an
Dn =
a N − a0
6. Percept simulation of levels
Fig. 10: Triangular simulation of internal pressure
Table :1 Statistical Characteristics of Each Pressure Mode
Mean of p i
Pressure mode C.o.v. of p i in % Law
( p i in MPa)
High (mode 1) 8 10 % Triangular
Middle (mode 2) 5 10% Triangular
Low (mode 3) 3 10% Triangular
11
8. Control Theory and Informatics www.iiste.org
ISSN 2224-5774 (print) ISSN 2225-0492 (online)
Vol 2, No.2, 2012
We study three levels of maximal pressures in pipelines which are: 3 MPa, 5 MPa, and 8 MPa that are
repeated within a specific interval of time T=8 hours. At each level, we deduce the degradation trajectory D
in terms of time or in terms of the number of cycles N.
The failure by fatigue is obtained for a certain critical number of cycles: pressure-depression or for a certain
time period. Therefore, the lifetime of the pipeline for each level of maximal pressure is deduced at D=1.
7. Results and Discussion on Simulation
The Monte Carlo one level percept simulations for 1000 times for the pipeline system and under the 3
modes of internal pressure (high, middle and low) gives the degradation trajectory which are represented in
the following 3 figures.
Fig. 11: Degradation evolution for mode 1
12
9. Control Theory and Informatics www.iiste.org
ISSN 2224-5774 (print) ISSN 2225-0492 (online)
Vol 2, No.2, 2012
Fig. 12: Degradation evolution for mode 2
Fig. 13: Degradation evolution for mode 3
13
10. Control Theory and Informatics www.iiste.org
ISSN 2224-5774 (print) ISSN 2225-0492 (online)
Vol 2, No.2, 2012
Fig. 14: Degradation evolution for All three modes
We deduce from the percept interrogation that the pipeline lifetime is nearly 115 hours for mode 1 (high
pressure), nearly 160 hours for mode 2 (middle pressure), and nearly 240 hours for mode 3 (low pressure).
From these curves, we can see that our prognostic model, using analytic laws, gives the remaining lifetime
of pipelines at any instant.
8. Conclusion and Scope for Future Work
The percept neural network sustains in predicting the life time effectiveness on field efficiency for the
radial pipelines by which the user is able to read the rear and bear happenings on fluid mechanics in
industries. The study also helps in predicting the sustainability feature of turbines in heavy alloy plants
which could be scope for the work in future.
References
G. Vachtsevanos, F. Lewis, M. Roemer, A. Hess, B. Wu, Intelligent Fault Diagnosis and Prognosis
for Engineering Systems, John Wiley & Sons, Inc., 2006, ch. 5,6 and 7.
J. Lemaitre and J. Chaboche, Mechanics of Solid Materials. New York: Cambridge University Press,
1990.
M. Langon, Introduction a la Fatigue et Mécanique de la Rupture, Centre d’essais aéronautique de
Toulouse, ENSICA April,1999
K. El-Tawil, S. Kadry, Fatigue Stochastique des Systèmes Mécaniques Basée sur la Technique de
Transformation Probabiliste, internal report, Lebanese University, grant research program, 2010
J. Lemaitre, R. Desmorat, Engineering Damage Mechanics, New York: Springer-Verlag, 2005, ch. 6.
14
11. Control Theory and Informatics www.iiste.org
ISSN 2224-5774 (print) ISSN 2225-0492 (online)
Vol 2, No.2, 2012
K. El-Tawil, A. Abou Jaoude and S. Kadry, “Life time estimation under probabilistic fatigue of
cracked plates for multiple limit states”, ICNAAM, 2009.
K. El-Tawil, Mécanique Aléatoire et Fiabilité, cours de master2r mécanique, Ecole doctorale des
sciences et technologies EDST Université libanaise, Beyrouth 2004
A. Abou Jaoude, K. El-Tawil, S. Kadry, H. Noura and M. Ouladsine, ”Analytic prognostic model for
a dynamic system”, European Conference of Control, 2010, submitted for publication.
15
12. International Journals Call for Paper
The IISTE, a U.S. publisher, is currently hosting the academic journals listed below. The peer review process of the following journals
usually takes LESS THAN 14 business days and IISTE usually publishes a qualified article within 30 days. Authors should
send their full paper to the following email address. More information can be found in the IISTE website : www.iiste.org
Business, Economics, Finance and Management PAPER SUBMISSION EMAIL
European Journal of Business and Management EJBM@iiste.org
Research Journal of Finance and Accounting RJFA@iiste.org
Journal of Economics and Sustainable Development JESD@iiste.org
Information and Knowledge Management IKM@iiste.org
Developing Country Studies DCS@iiste.org
Industrial Engineering Letters IEL@iiste.org
Physical Sciences, Mathematics and Chemistry PAPER SUBMISSION EMAIL
Journal of Natural Sciences Research JNSR@iiste.org
Chemistry and Materials Research CMR@iiste.org
Mathematical Theory and Modeling MTM@iiste.org
Advances in Physics Theories and Applications APTA@iiste.org
Chemical and Process Engineering Research CPER@iiste.org
Engineering, Technology and Systems PAPER SUBMISSION EMAIL
Computer Engineering and Intelligent Systems CEIS@iiste.org
Innovative Systems Design and Engineering ISDE@iiste.org
Journal of Energy Technologies and Policy JETP@iiste.org
Information and Knowledge Management IKM@iiste.org
Control Theory and Informatics CTI@iiste.org
Journal of Information Engineering and Applications JIEA@iiste.org
Industrial Engineering Letters IEL@iiste.org
Network and Complex Systems NCS@iiste.org
Environment, Civil, Materials Sciences PAPER SUBMISSION EMAIL
Journal of Environment and Earth Science JEES@iiste.org
Civil and Environmental Research CER@iiste.org
Journal of Natural Sciences Research JNSR@iiste.org
Civil and Environmental Research CER@iiste.org
Life Science, Food and Medical Sciences PAPER SUBMISSION EMAIL
Journal of Natural Sciences Research JNSR@iiste.org
Journal of Biology, Agriculture and Healthcare JBAH@iiste.org
Food Science and Quality Management FSQM@iiste.org
Chemistry and Materials Research CMR@iiste.org
Education, and other Social Sciences PAPER SUBMISSION EMAIL
Journal of Education and Practice JEP@iiste.org
Journal of Law, Policy and Globalization JLPG@iiste.org Global knowledge sharing:
New Media and Mass Communication NMMC@iiste.org EBSCO, Index Copernicus, Ulrich's
Journal of Energy Technologies and Policy JETP@iiste.org Periodicals Directory, JournalTOCS, PKP
Historical Research Letter HRL@iiste.org Open Archives Harvester, Bielefeld
Academic Search Engine, Elektronische
Public Policy and Administration Research PPAR@iiste.org Zeitschriftenbibliothek EZB, Open J-Gate,
International Affairs and Global Strategy IAGS@iiste.org OCLC WorldCat, Universe Digtial Library ,
Research on Humanities and Social Sciences RHSS@iiste.org NewJour, Google Scholar.
Developing Country Studies DCS@iiste.org IISTE is member of CrossRef. All journals
Arts and Design Studies ADS@iiste.org have high IC Impact Factor Values (ICV).