BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Battery degradation model for mission assignment in a fleet of ele
ctric vehicles
DTSTART:20220519T145000Z
DTEND:20220519T151000Z
DTSTAMP:20240621T033100Z
UID:indico-contribution-380@conferences.enbis.org
DESCRIPTION:Speakers: Pedro Dias Longhitano (Volvo Group)\, Benjamin Echar
d (Volvo Trucks)\, Khaoula Tidriri (Gipsa-lab)\, Christophe Bérenguer (Un
iv. Grenoble Alpes)\n\nBattery prognostics and health management has recen
tly become a very important and strategic topic specially with the rise of
electric vehicles and electric mobility in general\, which is seen as a k
ey tool to reduce the impact of global warming. In order for battery heal
th management to be viable\, it is necessary to quantify and understand ba
ttery state of health (SOH) and its degradation mechanisms. A lot of resea
rch has been developed to understand the different degradation processes i
n a battery [1]\, to identify and understand the impacts of stress factors
[2]\, and to quantify tendencies of degradation and predict remaining use
ful life [3]. In this presentation\, an overview on battery degradation mo
delling and prognostics is given\, exploring the definitions of the main s
tress factors\, and covering the most common degradation models in the lit
erature. Particular attention is given to an empirical model based on stre
ss cycle decomposition linking the degradation to the history of charge an
d discharge cycles\, which is flexible and accurate [4]. This model can be
used to estimate the SoH variation of the battery based on the way the ve
hicle is driven. A three-step method is thus proposed to develop a compreh
ensive model for degradation induced by driving conditions\, combining the
aforementioned degradation model\, with battery and vehicle dynamics mode
ls\, as well as information on road topography and vehicle parameters. The
first step of this overall degradation model consists in using the availa
ble information topography\, speed limits and crossroads to estimate the e
lectrical power required to carry on a given displacement. In the second s
tep\, the required electrical power is used to infer a state of charge (So
C) trajectory. Finally\, in the last step\, the SoC profile is decomposed
into stress cycles that serve as input for the degradation model. Such a c
omprehensive SOH evolution model linking route profiles and driving condit
ions to battery degradation is suitable for decision-making problems relat
ed to the optimal management of electric vehicles. The use of this degrada
tion model is illustrated on a use-case of a fleet of electric vehicles th
at must perform a set of missions: it is shown how the order of those miss
ions can be decided by optimizing not only energy consumption but also bat
tery degradation. \n\n[1] Anthony Barré\, Benjamin Deguilhem\, Sébastien
Grolleau\, Mathias Gérard\, Frédéric Suard\, Delphine Riu\, A review o
n lithium-ion battery ageing mechanisms and estimations for automotive app
lications\, Journal of Power Sources\,Volume 241\,\n[2] Saxena\, Saurabh\,
Darius Roman\, Valentin Robu\, David Flynn\, and Michael Pecht. 2021. "Ba
ttery Stress Factor Ranking for Accelerated Degradation Test Planning Usin
g Machine Learning" Energies 14\, no. 3: 723. \n[3] Jiaming Fan et al.
“A novel machine learning method based approach for Li-ion battery progn
ostic and health management”. In: IEEE Access 7 (2019)\, pp. 160043–\n
160061 \,2013\, Pages 680-689\,\n[4] Xu \,B.\, Oudalov\, A.\, Ulbig\, A.\,
Andersson\, G.\, and Kirschen\, D.S. (2018). Modeling of lithium-ion batt
ery degradation for cell life assessment. IEEE Transactions on Smart Grid\
, 9(2)\, 1131–1140. doi:10.1109/TSG.2016.2578950\n\nhttps://conferences.
enbis.org/event/16/contributions/380/
LOCATION:Grenoble
URL:https://conferences.enbis.org/event/16/contributions/380/
END:VEVENT
BEGIN:VEVENT
SUMMARY:A goodness-of-fit test for homogeneous gamma process under a gener
al sampling scheme
DTSTART:20220520T091000Z
DTEND:20220520T093000Z
DTSTAMP:20240621T033100Z
UID:indico-contribution-340@conferences.enbis.org
DESCRIPTION:Speakers: Christian Paroissin (Université de Pau et des Pays
de l'Adour)\n\nDegradation models are more and more studied and used in pr
actice. Most of these models are based on Lévy processes. For such models
\, estimation methods have been proposed. These models are also considered
for developing complex and efficient maintenance policies. However\, a ma
in issue remains: goodness-of-fit (GoF) test for these models. In this tal
k\, we propose a GoF test for the homogeneous gamma process under a genera
l sampling scheme.\n\nhttps://conferences.enbis.org/event/16/contributions
/340/
LOCATION:Grenoble
URL:https://conferences.enbis.org/event/16/contributions/340/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Quantile Regression via Accelerated Destructive Degradation Modeli
ng for Reliabilty Estimation
DTSTART:20220520T081000Z
DTEND:20220520T083000Z
DTSTAMP:20240621T033100Z
UID:indico-contribution-366@conferences.enbis.org
DESCRIPTION:Speakers: Munwon Lim (Hanyang University)\, Suk Joo Bae (Hanya
ng University)\, Gyu Ri Kim (Department of Industrial Engineering\, Hanyan
g University)\n\nAlong with the shortening of production period\, manufact
uring industry utilizes an accelerated degradation test (ADT) to estimate
the reliability of newly developed products as quickly as possible. In ADT
\, the stress factor which is related to the failure mechanisms is imposed
to cause the failure of products faster than those under normal use condi
tion. By increasing the degree of stress such as voltage\, temperature\, h
umidity or other external factors\, the performance of new products contin
uously degrades and leads to the failure. In some applications\, accelerat
ed destructive degradation test (ADDT) is conducted when testing units sho
uld be destroyed to measure the performance of the degrading product. \nFo
r general ADT and ADDT models\, the mean estimators have been considered a
s the location measurement. However\, the estimation result using mean est
imator can be inappropriate for highly skewed data\, because the lifetime
estimation of degradation data with outliers or irregularity can be distor
ted.\nIn this paper\, the ADDT modeling based on quantile regression (QR)
is suggested as a comprehensive approach for asymmetric observations. QR-b
ased ADDT requires fewer assumptions than the general parametric methods t
o construct the model\, and enables the interpretation to be more flexible
. Through the data application\, our approach provides a great advantage b
y inferring a nonlinear degradation path without bias and partiality.\n\nh
ttps://conferences.enbis.org/event/16/contributions/366/
LOCATION:Grenoble
URL:https://conferences.enbis.org/event/16/contributions/366/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Statistical inference for a Wiener-based degradation model with im
perfect maintenance actions under different observation schemes
DTSTART:20220520T085000Z
DTEND:20220520T091000Z
DTSTAMP:20240621T033100Z
UID:indico-contribution-331@conferences.enbis.org
DESCRIPTION:Speakers: Margaux Leroy (Univ. Grenoble Alpes)\, Laurent Doyen
(Univ. Grenoble Alpes)\, Olivier Gaudoin (Université Grenoble Alpes)\, C
hristophe Bérenguer (Univ. Grenoble Alpes)\n\nIn this article\, technolog
ical or industrial equipment that are subject to degradation are considere
d. These units undergo maintenance actions\, which reduce their degradatio
n level.\nThe paper considers a degradation model with imperfect maintenan
ce effect. The underlying degradation process is a Wiener process with dri
ft. The maintenance effects are described with an Arithmetic Reduction of
Degradation ($ARD_1$) model. The system is regularly inspected and the deg
radation levels are measured.\n \nFour different observation schemes are
considered so that degradation levels can be observed between maintenance
actions as well as just before or just after maintenance times. In each sc
heme\, observations of the degradation level between successive maintenanc
e actions are made. In the first observation scheme\, degradation levels j
ust before and just after each maintenance action are observed. In the sec
ond scheme\, degradation levels just after each maintenance action are not
observed but are observed just before. On the contrary\, in the third sch
eme\, degradation levels just before the maintenance actions are not obse
rved but are observed just after. Finally\, in the fourth observation sche
me\, the degradation levels are not observed neither just before nor just
after the maintenance actions.\n \nThe paper studies the estimation of th
e model parameters under these different observation schemes. The maximum
likelihood estimators are derived for each scheme.\nSeveral situations are
studied in order to assess the impact of different features on the estima
tion quality.\nAmong them\, the number of observations between successive
maintenance actions\, the number of maintenance actions\, the maintenance
efficiency parameter and the location of the observations are considered.
These situations are used to assess the estimation quality and compare the
observation schemes through an extensive simulation and performance study
.\n\nhttps://conferences.enbis.org/event/16/contributions/331/
LOCATION:Grenoble
URL:https://conferences.enbis.org/event/16/contributions/331/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Phase-type models for competing risks
DTSTART:20220520T122000Z
DTEND:20220520T124000Z
DTSTAMP:20240621T033100Z
UID:indico-contribution-342@conferences.enbis.org
DESCRIPTION:Speakers: Bo Henry Lindqvist (Norwegian University of Science
and Technology)\n\nA phase-type distribution can be defined to be the dist
ribution of time to absorption for an absorbing finite state Markov chain
in continuous time. Phase-type distributions have received much attention
in applied probability\, in particular in queuing theory\, generalizing th
e Erlang distribution. Among other applications\, they have for a long tim
e been used in reliability and survival analysis. Particular interest has
been in the use of so-called Coxian phase-type models. Their usefulness st
ems from the fact that they are able to model phenomena where an object go
es through stages (phases) in a specified order\, and may transit to the a
bsorbing state (corresponding to the event of interest) from any phase. I
t is noteworthy that Coxian phase-type models have recently\, in a number
of papers\, been successfully applied to model hospital length of stay in
health care studies. These authors typically claim the superiority of Coxi
an phase-type models over common parametric models like gamma and lognorma
l for this kind of data. Similar models are apparently appropriate for rel
iability modeling of complex degrading systems. \n\nThe main purpose of th
e present talk is to study how the phase-type methodology can be modified
to include competing risks\, thereby enabling the modeling of failure dist
ributions with several failure modes\, or\, more generally\, event histori
es with several types of events. One then considers a finite state Markov
chain with more than one absorbing state\, each of which corresponds to a
particular risk. Standard functions from the theory of competing risks can
now be given in terms of the transition matrix of the underlying Markov c
hain. We will be particularly concerned with the uniqueness of parameteriz
ations of phase-type models for competing risks\, which is of particular i
nterest in statistical inference. We will briefly consider maximum likelih
ood estimation in Coxian competing risks models\, using the EM algorithm.
A real data example will be analyzed for illustration.\n\nhttps://conferen
ces.enbis.org/event/16/contributions/342/
LOCATION:Grenoble
URL:https://conferences.enbis.org/event/16/contributions/342/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Modeling Spatially Clustered Failure Time Data via Multivariate Ga
ussian Random Fields
DTSTART:20220520T120000Z
DTEND:20220520T122000Z
DTSTAMP:20240621T033100Z
UID:indico-contribution-385@conferences.enbis.org
DESCRIPTION:Speakers: akim adekpedjou (University of Missouri)\, Sophie Da
bo\, zarah Sainul\n\nConsider a fixed number of clustered areas identified
by their geographical coordinate that monitored for the occurrences of an
event such as pandemic\, epidemic\, migration to name a few. Data collect
ed on units at all areas include time varying covariates and environmental
factors. The collected data is considered pairwise to account for spatial
correlation between all pair of areas. The pairwise right censored data
is probit-transformed yielding a multivariate gaussian random field preser
ving the spatial correlation function. The data is analyzed using counting
process machinery and geostatistical formulation that led to a class of w
eighted pairwise semiparametric estimating functions. Estimators of model
s unknowns are shown to be consistent and asymptotically normally distrib
uted under infill-type spatial statistics asymptotic. Detailed small samp
le numerical studies that are in agreement with theoretical results are pr
ovided. The foregoing procedures are applied to leukemia survival data in
Northeast England.\n\nhttps://conferences.enbis.org/event/16/contributions
/385/
LOCATION:Grenoble
URL:https://conferences.enbis.org/event/16/contributions/385/
END:VEVENT
BEGIN:VEVENT
SUMMARY:RELSYS : A new method based on damage physical-chemical processes
with uncertainties and hazard.
DTSTART:20220520T124000Z
DTEND:20220520T130000Z
DTSTAMP:20240621T033100Z
UID:indico-contribution-389@conferences.enbis.org
DESCRIPTION:Speakers: Jerome de Reffye (individual researcher ( retired))\
n\nJerome de Reffye\nEngineering Degrees from Gustave Eiffel University (
ESIEE ) \nand Pierre et Marie Curie University ( Paris VI )\nPhD in applie
d mathematics and theoretical physics \nfrom Pierre et Marie Curie Univers
ity\n\nOn opposite way of empirical approachs we develop an analytic metho
d in Reliability - Maintainability - Availability - Safety (RAMS) area whi
ch led to the RELSYS model\, (RELiability of SYStems)\, allowing to take i
nto account all the physical-chemical parameters of the degradation models
of the system components as well as their uncertainties in a model of ran
domized physical-chemical evolution allowing the complete calculation of t
he probabilities of their failures. This calculation is sufficiently compl
ete to be compatible with the actuarial calculation of insurance\, both of
which allow the association between the RAMS and the calculation of the g
uarantee of the system costs. \nWe show that it is possible\, with precise
calculations\, to evaluate the risks of uninsurable systems by time serie
s because of the lack of data due to the rarity of the feared events. The
probabilities of occurrence of these events being very low\, the calculati
ons must be based on justified models. \nIt uses the Langevin’s equation
for phenomena that evolve slowly over time. The notions of Limit State in
Service and Ultimate Limit State are introduced.\nThis model gives access
to dynamic reliability which studies time-dependent phenomena and provide
s failure probabilities as functions of time through numerical simulation.
\nWe obtain random functions of time whose parameters are themselves rand
om. The uncertainties are thus divided into two parts according to their o
rigin: Uncertainties on the physical-chemical parameters (randomness conce
ntrated at the origin) and uncertainties on the realization of the degrada
tion processes (randomness distributed in time). \nWe thus obtain the fail
ure probabilities with their confidence intervals. The numerous examples s
how that RELSYS can be applied to any man-made system. We show the importa
nce of taking into account the probabilistic aspect of the problem from th
e beginning of the modeling and to develop determinism within the random m
odel. Finally\, for the solution of particular problems\, one will find or
iginal methods in signal processing.\nThe RELSYS application into system m
aintenance is using classical theory of random processes. The preventive m
aintenance parameters can be calculated by RELSYS from the failure probabi
lities and the technical specifications about the residual failure probabi
lities. The corrective maintenance cost can be deduced from the previous a
nalysis. \nThe application of RELSYS to the calculation of the cost of the
commitments of guarantees of a program uses the concept of Value at Risk.
The used techniques are derived from reinsurance.\nWe will show numerous
examples illustrating the theory by practice in engineering. RELSYS suppli
es a whole tool to dynamical RAMS analysis.\n\nhttps://conferences.enbis.o
rg/event/16/contributions/389/
LOCATION:Grenoble
URL:https://conferences.enbis.org/event/16/contributions/389/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Combining AI with Model based Design: battery State-of-charge esti
mator using Deep Learning
DTSTART:20220520T130000Z
DTEND:20220520T132000Z
DTSTAMP:20240621T033100Z
UID:indico-contribution-344@conferences.enbis.org
DESCRIPTION:Speakers: Moubarak GADO (MathWorks)\, PIERRE HAROUIMI\n\nAcros
s industries\, the growing dependence on battery pack energy storage has u
nderscored the importance of battery management systems (BMS) whose role i
s to monitor battery state\, ensure safe operation and maximize performanc
e. For example\, the BMS helps avoid overcharging and over discharging\, i
t manages the temperature of the battery and so on\, and it does so by col
lecting information from sensors on the battery for current\, voltage\, te
mperature etc. So\, this is a closed-loop system by design.\nOne of the th
ings that cannot be directly measured but is required for many of these op
erations is the battery state of charge (SOC). So\, this quantity needs to
be estimated somehow. One way to solve this problem is using recursive es
timation based on a Kalman filter. However\, the Kalman filter requires a
dynamical model of the battery – which may or may not be accurate – an
d is very time-consuming. Besides\, handling just the algorithm is not en
ough. Models need to be incorporated into an entire system design workflow
to deliver a product or a service to the market. The bridge between engin
eering and science workflows is one of the most important pieces of such a
n application. Combining Model-Based-Design with Artificial Intelligence w
ill enrich the model and make collaboration between teams robust and more
automated.\n\nWe will explore\, in detail\, the workflow involved in devel
oping\, testing\, and deploying an AI-based state-of-charge estimator for
batteries using Model-Based Design:\n - Designing and training deep learni
ng models \n - Demonstrate a workflow for how you can research\, develop\,
and deploy your own deep learning application with Model-Based Design\n -
Integrating deep learning and machine learning models into Simulink for s
ystem-level simulation\n - Generate optimized C code and Performed Process
or-in-the-loop (PIL) test\n\nhttps://conferences.enbis.org/event/16/contri
butions/344/
LOCATION:Grenoble
URL:https://conferences.enbis.org/event/16/contributions/344/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Maintenance policies for items with their repair processes modelle
d by the extended Poisson processes
DTSTART:20220519T120000Z
DTEND:20220519T122000Z
DTSTAMP:20240621T033100Z
UID:indico-contribution-379@conferences.enbis.org
DESCRIPTION:Speakers: Shaomin Wu (University of Kent)\, Jiaqi Yin (Univers
ity of Kent)\n\nOptimisation of maintenance policies for items with their
repair processes modelled by the geometric process (GP) has received a goo
d amount of attention. The extended Poisson process (EPP)\, one of the ext
ensions of the GP\, can be used to model the repair process with its times
-between-failures possessing a non-monotonic trend. A central issue in the
applications of the EPPs in maintenance policy optimisation is to find wh
en the EPP has an increasing times-between-failures. This paper aims to an
swer this question. Numerical examples are provided to illustrate the prop
osed maintenance policies.\n\nhttps://conferences.enbis.org/event/16/contr
ibutions/379/
LOCATION:Grenoble
URL:https://conferences.enbis.org/event/16/contributions/379/
END:VEVENT
BEGIN:VEVENT
SUMMARY:The long road from data collection to maintenance optimization of
industrial equipment
DTSTART:20220519T141000Z
DTEND:20220519T143000Z
DTSTAMP:20240621T033100Z
UID:indico-contribution-383@conferences.enbis.org
DESCRIPTION:Speakers: Emmanuel REMY (EDF R&D)\, Yves LE BRIS (EDF DTEAM
– CIST INGEUM)\n\nIn the coming years\, with the development of intermit
tent renewable energy sources and the gradual phasing out of coal-fired po
wer plants\, combined cycle gas turbines (CCGT) will play an essential rol
e in regulating electricity production. This need for flexibility will inc
rease the demands on the equipment of these CCGT and the issue of optimizi
ng their maintenance will become increasingly important. To this end\, the
use of statistical tools to enhance the value of data from the operation
and maintenance of these plants' equipment is a possible approach to provi
de decision support elements.\nIt is in this industrial context that an im
portant collection of data was carried out for several conventional repair
able equipment (turbines\, pumps...) of three EDF CCGT. The second step co
nsisted in a pre-processing / cleaning of these raw data with the support
of field experts\, an essential requirement for the statistical modeling s
tage. A wide range of imperfect maintenance models implemented in the free
R VAM (for Virtual Age Models) package (https://rpackages.imag.fr/VAM#) w
as tested to evaluate the ability of these models\, on the one hand to rep
roduce the field reality\, on the other hand to bring useful insights to h
elp the development of equipment maintenance plans.\nThe communication wil
l present this work\, illustrating it on a piece of equipment and insistin
g on its industrial application dimension.\n\nhttps://conferences.enbis.or
g/event/16/contributions/383/
LOCATION:Grenoble
URL:https://conferences.enbis.org/event/16/contributions/383/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Imperfect condition-based maintenance for a gamma degradation proc
ess in presence of unknown parameters
DTSTART:20220520T083000Z
DTEND:20220520T085000Z
DTSTAMP:20240621T033100Z
UID:indico-contribution-338@conferences.enbis.org
DESCRIPTION:Speakers: Franck Corset (LJK - Université Grenoble Alpes)\, M
itra Fouladirad (Centrale Marseille)\, Christian Paroissin (Université Pa
u et Pays de l'Adour)\n\nA system subject to degradation is considered. Th
e degradation is modelled by a gamma process. A condition-based maintenanc
e policy with perfect corrective and an imperfect preventive actions is pr
oposed. The maintenance cost is derived considering a Markov-renewal proce
ss. The statistical inference of the degradation and maintenance parameter
s by the maximum likelihood method is proposed. A sensibility analysis to
different parameters is carried out and the perspectives are detailed.\n\n
https://conferences.enbis.org/event/16/contributions/338/
LOCATION:Grenoble
URL:https://conferences.enbis.org/event/16/contributions/338/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Stochastic Drift Model for Discrete Parameters
DTSTART:20220519T085000Z
DTEND:20220519T091000Z
DTSTAMP:20240621T033100Z
UID:indico-contribution-341@conferences.enbis.org
DESCRIPTION:Speakers: Lukas Sommeregger (Infineon Technologies Austria AG)
\, Horst Lewitschnig (Infineon Technologies Austria AG)\n\nIn the context
of semiconductor reliability\, predictive maintenance and calculation of r
esidual useful life are important topics under the greater umbrella of pro
gnostics and health management.\nEspecially in automotive applications\, w
ith higher expected usage times of self-driving autonomous vehicles\, it b
ecomes more and more important to recognize degradation processes early\,
so that preventive maintenance actions can be taken automatically. For sem
iconductor producers\, it is important to account for life-time degradatio
n of electronic devices when guaranteeing quality standards for their cust
omer.\nFor this\, accurate and fast statistical models are needed to ident
ify degradation by parameter drift. Typically\, electrical parameters have
specified limits in which they need to stay over their whole life cycle.
\nEfficient life-time simulations are performed by so-called accelerated s
tress tests. In those tests\, electrical parameters are measured before\,
during\, and after higher-than usual stress conditions. These stress test
data represent the expected life-time behavior of these parameters.\nUsing
models based on these data\, tighter limits\, so called test limits\, are
then introduced at production testing to guarantee life-time quality of t
he devices for the customer.\nBased on this data\, quality control measure
s like guard bands are introduced. Guard bands are the differences between
specification and test limits and account\, amongst others\, for lifetime
drift effects of electrical parameters.\nModels to calculate lifetime dri
ft have to be flexible enough to accurately represent a large number of st
ress test behaviors while being computationally light-weight enough to run
on edge devices in the vehicles.\nWe present a statistical model for disc
rete parameters based on nonparametric interval estimation of conditional
transition probabilities in Markov chains that allows for flexible modelli
ng and fast computation. We then show how to use the model to formulate an
integer optimization problem to calculate optimal test limits. Calculatio
n for both arbitrary parameter distributions at production testing as well
as defined initial distributions are shown. Finally\, we give an approach
to calculate remaining useful lifetime for electronic components. \nThe
work has been performed in the project ArchitectECA2030 under grant agreem
ent No 877539. The project is co-funded by grants from Germany\, Netherlan
ds\, Czech Republic\, Austria\, Norway and - Electronic Component Systems
for European Leadership Joint Undertaking (ECSEL JU).\nAll ArchitectECA203
0 related communication reflects only the author’s view and ECSEL JU and
the Commission are not responsible for any use that may be made of the in
formation it contains.\n\nhttps://conferences.enbis.org/event/16/contribut
ions/341/
LOCATION:Grenoble
URL:https://conferences.enbis.org/event/16/contributions/341/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Data-driven Maintenance Optimization Using Random Forest Algorithm
s
DTSTART:20220519T124000Z
DTEND:20220519T130000Z
DTSTAMP:20240621T033100Z
UID:indico-contribution-346@conferences.enbis.org
DESCRIPTION:Speakers: HASAN MISAII (University of TEHRAN and University of
Technology of TROYES)\, MITRA FOULADIRAD (Aix Marseille Université et Un
iversité de Technologie de Troyes)\, Firoozeh HAGHIGHI (University of Teh
ran )\n\nIn this paper\, a multi-component series system is considered whi
ch is periodically inspected and at inspection times the failed components
are replaced by a new one. Therefore\, this maintenance action is perfect
corrective maintenance for the failed component\, and it can be considere
d as imperfect corrective maintenance for the system. The inspection inter
val is considered as a decision parameter and the maintenance policy is op
timized using long-run cost rate function. It is assumed that there is no
information related to components' lifetime distributions and their parame
ters. Therefore\, an optimal decision parameter is derived considering his
torical data (a data storage for the system that includes information rela
ted to past repairs) using density estimation and random forest algorithms
. Eventually\, the efficiency of the proposed optimal decision parameter a
ccording to available data is compared to the one derived when all informa
tion on the system is available.\n\n keywords: Maintenance Optimization\,
Data-driven Estimation\, Random Forest Algorithm.\n\nhttps://conferences.e
nbis.org/event/16/contributions/346/
LOCATION:Grenoble
URL:https://conferences.enbis.org/event/16/contributions/346/
END:VEVENT
BEGIN:VEVENT
SUMMARY:On the modelling of dependence between univariate Lévy wear proce
sses and impact on the reliability function
DTSTART:20220519T083000Z
DTEND:20220519T085000Z
DTSTAMP:20240621T033100Z
UID:indico-contribution-345@conferences.enbis.org
DESCRIPTION:Speakers: Sophie MERCIER (University of Pau and Pays de l'Adou
r)\, Ghislain VERDIER (University of Pau and Pays de l'Adour)\n\nUnivariat
e Lévy processes have become quite common in the reliability literature f
or modelling accumulative deterioration. In case of correlated deteriorati
on indicators\, several possibilities have been suggested for modelling th
eir dependence. The point of this study is the analysis and comparison of
three different dependence models considered in the most recent literature
: 1. Use of a regular copula\, where the dependence in a multivariate incr
ement is modelled through a time-independent regular copula\; 2. Superposi
tion of independent univariate Lévy processes\, where each marginal proce
ss is constructed as the sum of independent univariate Lévy processes $\\
{X_j(t)\, t\\geq 0\\}$ with possibly common $\\{X_j(t)\, t\\geq 0\\}$ betw
een margins\; 3. Use of a Lévy copula. The three methods are first presen
ted and analysed. As for the model based on a regular copula\, it is shown
that the corresponding multivariate process cannot have independent incre
ments in general\, so that it is not a Lévy process. This means that the
distribution of the multivariate process is not fully characterized in thi
s way. The second and third models both lead to a multivariate Lévy proce
ss\, with a limited dependence range for the second superposition-based mo
del\, which is not the case for the third Lévy copula-based model. Howeve
r\, this last model requires a higher technicity for its use and numerical
methods (such as Monte-Carlo simulations) have to be used for its numeric
al assessment. Practical details are given in the paper and two Monte-Car
lo simulation procedures are compared.\n\nA two-component series system is
next considered\, with joint deterioration level modelled by one of the t
hree previous models. Each component is considered as failed as soon as it
s deterioration level is beyond a given failure threshold. The impact of a
wrong choice for the model is explored\, based on data simulated from one
of the three models and next adjusted to all three models. It is shown th
at a wrong choice for the model can lead to either surestimate or underest
imate the reliability function of the two-component series system\, which
could be problematic in an applicative context.\n\nhttps://conferences.enb
is.org/event/16/contributions/345/
LOCATION:Grenoble
URL:https://conferences.enbis.org/event/16/contributions/345/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Reliability degradation and optimal maintenance for information eq
uipment installed on railway cars
DTSTART:20220519T130000Z
DTEND:20220519T132000Z
DTSTAMP:20240621T033100Z
UID:indico-contribution-367@conferences.enbis.org
DESCRIPTION:Speakers: Haim Livni (Oslo Reliability)\n\nA Reliability degra
dation model was developed in a project\, which involved LCD TV screens in
stalled on railway. One critical item in the project was an LED strip. \nA
ccelerated Life tests data are the data source for the determination of LE
D Reliability. The expected life of LEDs being about 10-15 years - testing
them until death is not practical. \nLuminosity (vs time measurements pr
ovided by the manufacturer – were fitted to a degradation model. As oppo
sed to [1] and [2]\, where an exponential degradation model for the avera
ge luminosity was applied our degradation model assumed degradation equat
ions\, based on the second law of thermodynamics developed by A. Einstein
(1905)\, Fokker (1919) \, Planck (1930) and Kolmogorov (1931). The main di
fference in the approaches is that the degradation function's Taylor expan
sion contains terms of the first and second derivative\, while the models
of [1] and [2] contain only the first.\nAs opposed to [2] we did not fit t
he results to an assumed Reliability function (Weibull\, Normal\, Lognorma
l) and left it in a tabular form. The table allows to determine required r
eliability and maintenance information. \n\nThe following results are
derived:\n1. PDF of the failure rate as a function of time\n2. Reliabilit
y as a function of time and temperature (for simple and complex components
)\n3. MTBF for a device used for limited and unlimited life. \nA model de
veloped for maintenance costs and spare parts provisioning allows develop
ment Optimal Preventive Maintenance Policy :\n1. Optimal preventive maint
enance without individual monitoring.\n2. Optimal preventive maintenance b
ased on Rest of Useful Life (with monitoring)\n\nReferences\n\n1. Ott \,Me
lanie "Capabilities and Reliability of LEDs and Laser Diodes" Internal NAS
A Parts and Packaging Publication (1996).\n2. J.Fan K_C Yung\,M.Pecht Lif
etime Estimation of High-Power White LED Using Degradation-Data-Driven Met
hod IEEE TRANSACTIONS ON DEVICE AND MATERIALS RELIABILITY\, VOL. 12\, NO.
2\, JUNE 2012\n3. Si\, Xiao-Sheng\, et al. "Remaining useful life estimat
ion based on a nonlinear diffusion degradation process." IEEE Transactions
on reliability 61.1 (2012): 50-67\n4. Livni\, Haim. "Life cycle maintenan
ce costs for a non-exponential component." Applied Mathematical Modelling
103 (2022): 261-286.\n\nhttps://conferences.enbis.org/event/16/contribu
tions/367/
LOCATION:Grenoble
URL:https://conferences.enbis.org/event/16/contributions/367/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Fuel cell stochastic deterioration modeling for energy management
in a multi-stack system
DTSTART:20220519T091000Z
DTEND:20220519T093000Z
DTSTAMP:20240621T033100Z
UID:indico-contribution-360@conferences.enbis.org
DESCRIPTION:Speakers: Catherine Cadet (Gipsa-lab)\, Rachid Outbib\, zhongl
iang Li\, Jian Zuo (Phd student)\, Christophe Bérenguer (Univ. Grenoble A
lpes)\n\nFuel cells use hydrogen and oxygen as reactants to produce electr
icity through electrochemical reactions with the only byproduct of water.
They are widely used in various applications\, e.g. transport\, due to the
ir high efficiency\, energy density\, and limited impact on environmental
resources\, however fuel cells deployment is held down by multiple barrier
s such as their high cost or their shorter than required lifetime. To cros
s over these barriers\, using multi-stack fuel cells (MFC) instead of a si
ngle one is a promising solution. Firstly\, MFC offers improved reliabilit
y thanks to the multi-stack structure. Another advantage is that the durab
ility of multi-stack FC can also be increased by optimally distributing th
e power demand among different stacks by an efficient Energy Management St
rategy (EMS)\, and thus avoiding degraded mode operation [1]. In short\, M
FC systems are relevant to meet this challenge if properly dimensioned and
managed by an appropriate EMS taking into account the deterioration of th
e cells. In order to implement such a degradation-aware EMS\, it is mandat
ory to build a degradation model that integrates the dynamic behavior of M
FC according to the operating conditions. Fuel cell performance degradatio
n is linked to complex electrochemical\, mechanical\, and thermal mechanis
ms\, which are difficult to model using a “white-box” approach\, relyi
ng on the exact laws of physics. Within this context\, the aim of the pres
ent work is to propose a fuel cell degradation model adapted for the energ
y management of MFC.\nThe deterioration behavior of an MFC is characterize
d by two main features : (i) it is load-dependent\, i.e. the degradation i
s affected by the load distributed by the stack \; (ii) it is stochastic a
nd exhibits a stack-to-stack variability. A degradation-aware energy manag
ement system allocates a load to deliver to the different stacks of the MF
C system as a function of their degradation state and of their predicted d
egradation behavior. The deterioration dynamics must thus be modeled as a
function of the load power. Another specificity of fuel cells is their ind
ividual deterioration variability\, which can be due to stochasticity in t
he intrinsic fuel cell deterioration phenomena. This stochasticity varies
the deterioration levels even for the identical stacks operating under ide
ntical load profiles.\nIn order to meet these modelling requirements\, thi
s work develops a load-dependent stochastic deterioration model for an MFC
. First\, the overall stack resistance is chosen as the degradation indica
tor\, as it carries the key aging information of a fuel cell stack. Then\,
a stochastic non-homogeneous Gamma process is used to model the deteriora
tion of the fuel cell\, i.e. the increase in the fuel cell resistance. The
shape parameter of the considered Gamma process is further modeled by an
empirical function of the fuel cell operation load in order to make the re
sistance deterioration load-dependent. Finally\, to model the individual d
eterioration heterogeneity\, a random effect is added to the Gamma process
on its scale parameter\, taken as a random variable following a probabili
ty distribution (a Gamma law is chosen in this work). \nResistance degrada
tion paths can then be simulated based on the proposed deterioration model
\, based on which the first hitting time distribution of a failure thresho
ld (or equivalently a remaining useful life distribution) can be estimated
and the reliability of the system can be analyzed. The proposed model can
also be used to optimize the load allocation strategy for an MFC [2].
\n \nKeywords: Multi-stack fuel cells\, load-dependent deteriorat
ion model\, stochastic modelling\, Gamma process\, random effect. \n\nRef
erences:\n[1]Marx\, Neigel\, et al. "On the sizing and energy management o
f an hybrid multistack fuel cell–Battery system for automotive applicati
ons." International Journal of Hydrogen Energy 42.2 (2017): 1518-1526.\n[2
] Zuo\, J.\, C. Cadet\, Z. Li\, C. Bérenguer\, and R. Outbib (2022). Post
-prognostics decision-making strategy for load allocation on a stochastica
lly deteriorating multi-stack fuel cell system. To appear in Proc. Inst. M
ech. Eng - Part O: Journal of Risk and Reliability.\n\nhttps://conferences
.enbis.org/event/16/contributions/360/
LOCATION:Grenoble
URL:https://conferences.enbis.org/event/16/contributions/360/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Change-level detection for Lévy subordinators
DTSTART:20220519T093000Z
DTEND:20220519T095000Z
DTSTAMP:20240621T033100Z
UID:indico-contribution-361@conferences.enbis.org
DESCRIPTION:Speakers: Zeina Al Masry (FEMTO ST\, Université de Franche Co
mté)\, Ghislain Verdier (LMAP\, Université de Pau et des pays de l'Adour
)\, Landy Rabehasaina (Laboratoire de Mathématiques\, Université Franche
Comté)\n\nLet $\\boldsymbol{X}=(X_t)_{t\\ge 0}$ be a process behaving as
a general increasing Lévy process (subordinator) prior to hitting a give
n unknown level $m_0$\, then behaving as another different subordinator on
ce this threshold is crossed. We address the detection of this unknown thr
eshold $m_0\\in [0\,+\\infty]$ from an observed trajectory of the process.
These kind of model and issue are encountered in many areas such as relia
bility and quality control in degradation problems. More precisely\, we co
nstruct\, from a sample path and for each $\\epsilon >0$\, a so-called det
ection level $L_\\epsilon$ by considering a CUSUM inspired procedure. Und
er mild assumptions\, this level is such that\, while $m_0$ is infinite (i
.e. when no changes occur)\, its expectation $ \\mathbb{E}_{\\infty}(L_{\\
epsilon})$ tends to $+\\infty$ as $\\epsilon$ tends to $0$\, and the expec
ted overshoot $ \\mathbb{E}_{m_0}([L_{\\epsilon} - m_0]^+)$\, while the th
reshold $m_0$ is finite\, is negligible compared to $ \\mathbb{E}_{\\infty
}(L_{\\epsilon})$ as $\\epsilon$ tends to $0$. Numerical illustrations are
provided when the Lévy processes are gamma processes with different shap
e parameters. This is joint work with Z.Al Masry and G.Verdier.\n\nhttps:/
/conferences.enbis.org/event/16/contributions/361/
LOCATION:Grenoble
URL:https://conferences.enbis.org/event/16/contributions/361/
END:VEVENT
BEGIN:VEVENT
SUMMARY:A condition-based maintenance policy in a system with heterogeneit
ies
DTSTART:20220519T122000Z
DTEND:20220519T124000Z
DTSTAMP:20240621T033100Z
UID:indico-contribution-363@conferences.enbis.org
DESCRIPTION:Speakers: Luis Landesa (Universidad de Extremadura)\, Inma T.
Castro (Universidad de Extremadura)\, Lucia Bautista (Universidad de Extr
emadura)\n\nModels that describe the deterioration processes of the compon
ents are key to determining the lifetime of a system and play a fundamenta
l role in predicting the system reliability and planning the system mainte
nance. In most systems there is heterogeneity among the degradation paths
of the units. This variability is usually introduced in the model through
random effects\, that is\, considering random coefficients on the model. \
nA degrading system subject to multiple degradation processes whose initia
tion times follow a shot-Cox noise process is studied. The growth of these
processes is modeled by a homogeneous gamma process. A condition based ma
intenance policy with periodic inspections is applied to reduce the impact
of failures and optimise the total expected maintenance cost. The heterog
eneities between components are included in the model considering that the
scale parameter of the gamma process follows a uniform distribution. Nume
rical examples of this maintenance policy are given comparing both models\
, with and without heterogeneities.\n\nhttps://conferences.enbis.org/event
/16/contributions/363/
LOCATION:Grenoble
URL:https://conferences.enbis.org/event/16/contributions/363/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Development of an Operational Digital Twin of a Locomotive Systems
DTSTART:20220519T135000Z
DTEND:20220519T141000Z
DTSTAMP:20240621T033100Z
UID:indico-contribution-364@conferences.enbis.org
DESCRIPTION:Speakers: Ron S. Kenett (KPA Ltd. and the Samuel Neaman Instit
ute for National Policy Research\, Technion City)\, Jacob Bortman (PHM Lab
oratory\, Department of Mechanical Engineering\, Ben-Gurion University of
the Negev)\, Gabriel Davidyan (Israel railway)\n\nA Digital Twin (DT) is a
new and powerful concept that maps a physical structure operating in a sp
ecific context to the digital space. The development and deployment of a
DT improves forecasting prognostic performance and decision support for
operators and managers. \, DT have been introduced in various industries
across a range of application areas including design\, manufacturing and
maintenance. Due to the large impact of maintenance on the proper function
ing of a system\, maintenance is one of the most studied DT applications.
In the case of trains\, poor maintenance can put the rolling carts out of
service or\, worse\, pose a safety risk to passengers and operators. Imple
menting intelligent maintenance strategies can therefore offer tremendous
benefits. This study addresses the development of an architecture for DT d
esigned to formulate and evaluate new hypotheses in predictive maintenance
by iterating between physical experiments and computational experiments.
The designed DT supports a broad perspective on statistical aspects of sim
ulations and experiments. In addition\, the DT enables real-time predictio
n and optimization of the actual behavior of a system at any stage of its
life cycle. Examples of safety valves and suspension systems will be given
.\n\nhttps://conferences.enbis.org/event/16/contributions/364/
LOCATION:Grenoble
URL:https://conferences.enbis.org/event/16/contributions/364/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Statistical process control versus deep learning for predictive ma
intenance of power plant process data
DTSTART:20220519T143000Z
DTEND:20220519T145000Z
DTSTAMP:20240621T033100Z
UID:indico-contribution-362@conferences.enbis.org
DESCRIPTION:Speakers: Henrik Hansen (DTU & Ørsted (Industrial PhD))\, Bo
Friis Nielsen (Department of Applied Mathematics and Computer Science\, Ap
plied Technical University of Denmark)\, Murat Kulahci (DTU)\n\n# Abstract
\n\nThis work is motivated by the non-documented\, practical learnings ga
ined by a predictive maintenance (PdM) development team in the Danish ener
gy company Ørsted. The team implements PdM solutions for power plant mach
inery to monitor for faults in the making. Their learnings support the hyp
othesis that there are not significant enough benefits to be gained from u
sing overly complicated condition monitoring models on selected machinery.
\n\nTo explore this hypothesis\, we set out to compare two different metho
dologies for detecting faults in process data. We compare a classical late
nt structure-based method from the field of statistical process control (S
PC) with a standard autoencoder deep neural network. Furthermore\, we comp
are the fault detection performance of these methods with two more experim
ental deep learning models recently proposed in literature [1]. The reason
for these specific models is that they a priori seem very well suited for
the modelling task the PdM team had undertaken due to the models’ alleg
ed ability to automate domain knowledge in a data-driven way.\n\nWe benchm
ark all methods against each other using first the well-known Tennessee Ea
stman Process (TEP) data\, and subsequently data collected from two feedwa
ter pumps (FWP) at a large Danish combined heat and power plant. \n\nThe T
EP data stems from a simulation tool for generating data from the process\
, and thus a large number of datasets are generated for each of 20 process
disturbances. For the FWP data\, six historical faults in the form of lea
ks are used to test the methods against each other in their ability to det
ect faults as they develop over time. Each methods’ ability to detect fa
ults is measured using a weighted combination of performance metrics such
as mean absolute error\, ROC AUC and average precision AP.\n\nPreliminary
results of the experiments suggest that detection performance is comparabl
e between the different models on both datasets\, but that each model seem
s to come with its own set of advantages in terms of fault detection perfo
rmance\, as in the case of reaction time to certain types of faults.\n\nBa
sed on the mentioned datasets and models\, we discuss the quantitative res
ults of these experiments\, as well as other pros and cons\, such as numbe
r of modelling decisions\, hyperparameters etc. of each paradigm that may
influence the choice of detection model in an industrial setting. \n\n# Re
ferences\n\n 1. Schulze\, J.-P\, Sperl\, P\, Böttinger\, K. 2022\, “Ano
maly Detection by Recombining Gated Unsupervised Experts”\, arXiv prepri
nt: https://doi.org/10.48550/arXiv.2008.13763\n\nhttps://conferences.enbis
.org/event/16/contributions/362/
LOCATION:Grenoble
URL:https://conferences.enbis.org/event/16/contributions/362/
END:VEVENT
BEGIN:VEVENT
SUMMARY:Degradation Model Selection Using Depth Functions
DTSTART:20220520T093000Z
DTEND:20220520T095000Z
DTSTAMP:20240621T033100Z
UID:indico-contribution-365@conferences.enbis.org
DESCRIPTION:Speakers: Arefe Asadi (University of technology of Troyes)\, M
itra Fouladirad (École Centrale Marseille\, Marseille\, France\, Aix Mars
eille Université (Aix-en-Provence)\, Université de Technologie de Troyes
)\, Diego Rodolpho Tomasi (Biofortis)\n\nDegradation modeling is an effe
ctive way for the reliability analysis of complex systems. For highly reli
able systems\, in which their failure is hard to observe\, degradation mea
surements often provide more information than failure time to improve syst
em reliability (Meeker and Escobar 2014). The degradation can be viewed a
s damage to a system that accumulates over time and eventually leads to fa
ilure when the accumulation reaches a failure threshold\, either random or
stipulated by industrial standards (Ye and. Xie 2015). Two large classes
of degradation models are stochastic processes and general path models. Th
e stochastic-process-based models show great flexibility in describing the
failure mechanisms caused by degradation (Lehmann 2009). \n\nThe aim of d
egradation modeling in presence of degradation data is to select a model f
rom a set of competing models\, capturing the features of the underlying d
egradation phenomenon. An efficient statistical tool is able to discard ir
relevant models. The concept of statistical depth could be employed as a
statistical tool for model selection. A depth function reflects the centra
lity of the observation to a statistical population (Staerman et al 2020).
\n\nTukey (1975) introduced a data depth to extend the notion of a median
to multi-variate random variables. Depth function have been extended by F
rairman and Muniz (2001) and Cuevas et al. (2006\, 2007) for functional da
ta\, the data which are recorded densely over time with one observed funct
ion per subject (Hall et al 2006). An alternative point of view based on t
he graphic representation of curves is proposed by Lopez-Pintado (2009).\n
\nIn this paper\, stochastic processes such as Lévy processes or stochast
ic differential equations are considered to model the degradation. After m
odel calibration in presence of data\, the models that show high values of
depth function are compared and a methodology to exploit and analyze the
depth function results is proposed.\n\nhttps://conferences.enbis.org/event
/16/contributions/365/
LOCATION:Grenoble
URL:https://conferences.enbis.org/event/16/contributions/365/
END:VEVENT
END:VCALENDAR