Table Of ContentLecture Notes in Electrical Engineering 919
Sio-Iong Ao
Len Gelman Editors
Transactions
on Engineering
Technologies
Proceedings of World Congress
on Engineering 2021
Lecture Notes in Electrical Engineering
Volume 919
SeriesEditors
LeopoldoAngrisani,DepartmentofElectricalandInformationTechnologiesEngineering,UniversityofNapoli
FedericoII,Naples,Italy
MarcoArteaga,DepartamentdeControlyRobótica,UniversidadNacionalAutónomadeMéxico,Coyoacán,
Mexico
BijayaKetanPanigrahi,ElectricalEngineering,IndianInstituteofTechnologyDelhi,NewDelhi,Delhi,India
SamarjitChakraborty,FakultätfürElektrotechnikundInformationstechnik,TUMünchen,Munich,Germany
JimingChen,ZhejiangUniversity,Hangzhou,Zhejiang,China
ShanbenChen,MaterialsScienceandEngineering,ShanghaiJiaoTongUniversity,Shanghai,China
TanKayChen,DepartmentofElectricalandComputerEngineering,NationalUniversityofSingapore,
Singapore,Singapore
RüdigerDillmann,HumanoidsandIntelligentSystemsLaboratory,KarlsruheInstituteforTechnology,
Karlsruhe,Germany
HaibinDuan,BeijingUniversityofAeronauticsandAstronautics,Beijing,China
GianluigiFerrari,UniversitàdiParma,Parma,Italy
ManuelFerre,CentreforAutomationandRoboticsCAR(UPM-CSIC),UniversidadPolitécnicadeMadrid,
Madrid,Spain
SandraHirche,DepartmentofElectricalEngineeringandInformationScience,TechnischeUniversität
München,Munich,Germany
FaryarJabbari,DepartmentofMechanicalandAerospaceEngineering,UniversityofCalifornia,Irvine,CA,
USA
LiminJia,StateKeyLaboratoryofRailTrafficControlandSafety,BeijingJiaotongUniversity,Beijing,China
JanuszKacprzyk,SystemsResearchInstitute,PolishAcademyofSciences,Warsaw,Poland
AlaaKhamis,GermanUniversityinEgyptElTagamoaElKhames,NewCairoCity,Egypt
TorstenKroeger,StanfordUniversity,Stanford,CA,USA
YongLi,HunanUniversity,Changsha,Hunan,China
QilianLiang,DepartmentofElectricalEngineering,UniversityofTexasatArlington,Arlington,TX,USA
FerranMartín,Departamentd’EnginyeriaElectrònica,UniversitatAutònomadeBarcelona,Bellaterra,
Barcelona,Spain
TanCherMing,CollegeofEngineering,NanyangTechnologicalUniversity,Singapore,Singapore
WolfgangMinker,InstituteofInformationTechnology,UniversityofUlm,Ulm,Germany
PradeepMisra,DepartmentofElectricalEngineering,WrightStateUniversity,Dayton,OH,USA
SebastianMöller,QualityandUsabilityLaboratory,TUBerlin,Berlin,Germany
SubhasMukhopadhyay,SchoolofEngineering&AdvancedTechnology,MasseyUniversity,Palmerston
North,Manawatu-Wanganui,NewZealand
Cun-ZhengNing,ElectricalEngineering,ArizonaStateUniversity,Tempe,AZ,USA
ToyoakiNishida,GraduateSchoolofInformatics,KyotoUniversity,Kyoto,Japan
LucaOneto,DepartmentofInformatics,Bioengineering.,Robotics,UniversityofGenova,Genova,Genova,
Italy
FedericaPascucci,DipartimentodiIngegneria,UniversitàdegliStudi“RomaTre”,Rome,Italy
YongQin,StateKeyLaboratoryofRailTrafficControlandSafety,BeijingJiaotongUniversity,Beijing,China
GanWoonSeng,SchoolofElectrical&ElectronicEngineering,NanyangTechnologicalUniversity,
Singapore,Singapore
JoachimSpeidel,InstituteofTelecommunications,UniversitätStuttgart,Stuttgart,Germany
GermanoVeiga,CampusdaFEUP,INESCPorto,Porto,Portugal
HaitaoWu,AcademyofOpto-electronics,ChineseAcademyofSciences,Beijing,China
WalterZamboni,DIEM-UniversitàdeglistudidiSalerno,Fisciano,Salerno,Italy
JunjieJamesZhang,Charlotte,NC,USA
The book series Lecture Notes in Electrical Engineering (LNEE) publishes the
latest developments in Electrical Engineering - quickly, informally and in high
quality. While original research reported in proceedings and monographs has
traditionallyformedthecoreofLNEE,wealsoencourageauthorstosubmitbooks
devoted to supporting student education and professional training in the various
fieldsandapplicationsareasofelectricalengineering.Theseriescoverclassicaland
emergingtopicsconcerning:
(cid:129)
CommunicationEngineering,InformationTheoryandNetworks
(cid:129)
ElectronicsEngineeringandMicroelectronics
(cid:129)
Signal,ImageandSpeechProcessing
(cid:129)
WirelessandMobileCommunication
(cid:129)
CircuitsandSystems
(cid:129)
EnergySystems,PowerElectronicsandElectricalMachines
(cid:129)
Electro-opticalEngineering
(cid:129)
InstrumentationEngineering
(cid:129)
AvionicsEngineering
(cid:129)
ControlSystems
(cid:129)
Internet-of-ThingsandCybersecurity
(cid:129)
BiomedicalDevices,MEMSandNEMS
For general information about this book series, comments or suggestions, please
[email protected].
To submit a proposal or request further information, please contact the
PublishingEditorinyourcountry:
China
JasmineDou,Editor([email protected])
India,Japan,RestofAsia
SwatiMeherishi,EditorialDirector([email protected])
SoutheastAsia,Australia,NewZealand
RameshNathPremnath,Editor([email protected])
USA,Canada:
MichaelLuby,SeniorEditor([email protected])
AllotherCountries:
LeontinaDiCecco,SeniorEditor([email protected])
**ThisseriesisindexedbyEICompendexandScopusdatabases.**
·
Sio-Iong Ao Len Gelman
Editors
Transactions on Engineering
Technologies
Proceedings of World Congress on
Engineering 2021
Editors
Sio-IongAo LenGelman
InternationalAssociationofEngineers SchoolofComputingandEngineering
IAENGSecretariat TheUniversityofHuddersfield
HongKong,HongKong Huddersfield,UK
ISSN 1876-1100 ISSN 1876-1119 (electronic)
LectureNotesinElectricalEngineering
ISBN 978-981-19-3578-7 ISBN 978-981-19-3579-4 (eBook)
https://doi.org/10.1007/978-981-19-3579-4
©TheEditor(s)(ifapplicable)andTheAuthor(s),underexclusivelicensetoSpringerNature
SingaporePteLtd.2023
Thisworkissubjecttocopyright.AllrightsaresolelyandexclusivelylicensedbythePublisher,whether
thewholeorpartofthematerialisconcerned,specificallytherightsoftranslation,reprinting,reuse
ofillustrations,recitation,broadcasting,reproductiononmicrofilmsorinanyotherphysicalway,and
transmissionorinformationstorageandretrieval,electronicadaptation,computersoftware,orbysimilar
ordissimilarmethodologynowknownorhereafterdeveloped.
Theuseofgeneraldescriptivenames,registerednames,trademarks,servicemarks,etc.inthispublication
doesnotimply,evenintheabsenceofaspecificstatement,thatsuchnamesareexemptfromtherelevant
protectivelawsandregulationsandthereforefreeforgeneraluse.
Thepublisher,theauthors,andtheeditorsaresafetoassumethattheadviceandinformationinthisbook
arebelievedtobetrueandaccurateatthedateofpublication.Neitherthepublishernortheauthorsor
theeditorsgiveawarranty,expressedorimplied,withrespecttothematerialcontainedhereinorforany
errorsoromissionsthatmayhavebeenmade.Thepublisherremainsneutralwithregardtojurisdictional
claimsinpublishedmapsandinstitutionalaffiliations.
ThisSpringerimprintispublishedbytheregisteredcompanySpringerNatureSingaporePteLtd.
The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721,
Singapore
Preface
AlargeinternationalconferenceonAdvancesinEngineeringTechnologiesandPhys-
icalSciencewasheldinLondon,UK,July7–9,2021,undertheWorldCongresson
Engineering 2021 (WCE 2021). The WCE 2021 is organized by the International
AssociationofEngineers(IAENG);theCongressdetailsareavailableat:http://www.
iaeng.org/WCE2021.IAENGisanon-profitinternationalassociationforengineers
andcomputerscientists,whichwasfoundedoriginallyin1968.TheWorldCongress
on Engineering serves as good platforms for the engineering community to meet
witheachotherandtoexchangeideas.Theconferenceshavealsostruckabalance
betweentheoreticalandapplicationdevelopment.Theconferencecommitteeshave
beenformedwithoverthreehundredcommitteememberswhoaremainlyresearch
center heads, faculty deans, department heads, professors, and research scientists
fromover30countries.Thecongressistrulyglobalinternationaleventwithahigh
levelofparticipationfrommanycountries.TheresponsetotheCongresshasbeen
excellent.Therehavebeenmorethanfourhundredmanuscriptsubmissionsforthe
WCE 2021. All submitted papers have gone through the peer-review process, and
theoverallacceptancerateis50.73%.
This volume contains eleven revised and extended research articles written
by prominent researchers participating in the conference. Topics covered include
mechanicalengineering,engineeringmathematics,computerscience,electricalengi-
neering,andindustrialapplications.Thebookoffersthestateoftheartoftremendous
advancesinengineeringtechnologiesandphysicalscienceandapplicationsandalso
servesasanexcellentreferenceworkforresearchersandgraduatestudentsworking
onengineeringtechnologiesandphysicalscienceandapplications.
HongKong Prof.Sio-IongAo
Huddersfield,UK Prof.LenGelman
v
Contents
Piecewise Monotonic Fitting for Covid-19 Data Analysis
oftheUnitedKingdomDuring31-01-2020to19-11-2021 .............. 1
EvangelosE.Vassiliou,IoannisN.Perdikas,DemetriusE.Davos,
andIoannisC.Demetriou
OntheFreeBoundaryProblemfortheLowReynoldsNumber ........ 17
NinoKhatiashvili
CommunityDetectiononProximityNetworks ........................ 33
GünceKezibanOrmanandOnurAkdog˘an
A Machine-Learnt Approach to Market Segmentation
andPurchasePredictionUsingPoint-Of-Sale(POS)Data ............. 49
IromiR.Paranavithana,ThashikaD.Rupasinghe,andDanielD.Prior
FeasibilityandBenefitsofSmallPhotovoltaicSystemsinGhana ....... 63
KingsfordAtanga,GeorgeEduful,andDavidPulfrey
DesignofMultipleBearingPlacementsforBrushCutterVibration
Reduction ........................................................ 77
YukiKoikeandShingoTsuruoka
ExperimentalDEMHybridApproachforPredictionofLandslide
inUghelliSandstone ............................................... 87
AdeolaGraceOlugbenga,SimonJosephAntony,AbdulkarimNasir,
MohammedUmarGarba,andMuibatDiekolaYahya
ExtractingtheEfficientModelofCylindricalandConicalBasins
fromDifferentAspectRatios ....................................... 101
ShahadatHossainZehad, SadmanAlFaiyaz, MdRedwanIslam,
andIrfanAhmed
vii
Piecewise Monotonic Fitting for Covid-19
Data Analysis of the United Kingdom
During 31-01-2020 to 19-11-2021
EvangelosE.Vassiliou,IoannisN.Perdikas,DemetriusE.Davos,
andIoannisC.Demetriou
Abstract The least squares piecewise monotonic data approximation method is
applied to daily Covid-19 new cases and deaths data of the UK for the period 31-
01-2020to19-11-2021.Thedatademonstratewidevariationinpartsandnoticeable
peaks over time. We are interested in estimating turning points of the data in that
thefitisusefultoanalyzingtheprogressofthepandemic.Anenormousnumberof
combinationsofturningpointsneedbeconsideredinordertofindanoptimalcombi-
nation,butthemethodprovidesquiteefficientlyaglobalsolution.Ourresultsshow
theefficacyofthepiecewisemonotonicitymethodinlocatingoptimalturningpoints
that are significant to the Covid-19 analyses. We consider the facts that influence
thechoiceofthenumberofpeaks.Ouranalysisprovideduswithinsightsregarding
thedrivingforcesbehindtheturningpointsthatthemethoddetected,whichfurther
maybehelpfultomanagement,aspartoftheinformationonwhichdecisionswill
bemade.
· · ·
Keywords Approximation Combinatorialproblem Covid-19pandemicdata
· · · ·
Divideddifferenceoffirstorder Leastsquaresfit Peak Piecewisemonotonic
·
Turningpoint UnitedKingdom
B
E.E.Vassiliou( )
DepartmentofFinancialandManagementEngineering,UniversityoftheAegean,41
KountouriotouStreet,Chios82100,Greece
e-mail:[email protected]
I.N.Perdikas·D.E.Davos·I.C.Demetriou
DepartmentofEconomics,NationalandKapodistrianUniversityofAthens,1Sofokleousand
AristidouStreet,Athens10559,Greece
e-mail:[email protected]
D.E.Davos
e-mail:[email protected]
I.C.Demetriou
e-mail:[email protected]
©TheAuthor(s),underexclusivelicensetoSpringerNatureSingaporePteLtd.2023 1
S.-I.AoandL.Gelman(eds.),TransactionsonEngineeringTechnologies,LectureNotes
inElectricalEngineering919,https://doi.org/10.1007/978-981-19-3579-4_1
2 E.E.Vassiliouetal.
1 Introduction
There has been an explosion of scientific publications on the Covid-19 pandemic
(see,e.g.,[7]).Theapplicationandanalysisofthepiecewisemonotonicmethodto
peakestimationofCovid-19data,whichweaddresshere,areconductedforthefirst
time. Daily Covid-19 new cases and deaths data of the UK for the period 31-01-
2020to19-11-2021demonstratewidevariationintime,aswellassomenoticeable
peaksofdifferentmagnitudealongthetimerange.Weareinterestedinestimating
peaks and trends of these data that can assist in the decision making policy. Since
the underlying laws are unknown, the estimation problem is made rather difficult.
Weassume,though,thatiftheunderlyingfunctionispiecewisemonotonic,thenthe
numberofitsturningpointsissubstantiallylowerthanthenumberofturningpoints
ofthedata.
Therefore, we apply the method of Demetriou and Powell [6] that makes least
changestothedatabyimposingalimitonthenumberofsignchangesinthesequence
of the first differences. Specifically, if k−1 is the limit, then the approximated
values consist of at most k monotonic sections. Thus, if some value of k can be
derived from the form of the underlying function, or if the user is willing to try
several values of k, then no other knowledge of this function is required. In this
paper, we define “least changes” to the data by minimizing the sum of squares of
changes subject to the smoothness condition. Further, two consecutive monotonic
sectionsmeetataturningpoint.Thepositionsoftheturningpointsarealsointeger
variables of the optimization calculation whose optimal values have to be found
automatically.However,aboutO(nk−1)combinationsofpositionscanoccur,which
makes it impossible to test each and every one separately. Demetriou and Powell
[6]havestudiedthisproblem,and,furthermore,Demetriou[1, 2, 4]hasdeveloped
algorithms and software that obtain a global solution in only O(n2+knlog n)
2
computeroperations.Whenk =1(monotoniccase)ork =2(unimodalcase),this
workisreducedtoO(n).Thisexcellentefficiencyisduetoadecompositionproperty
ofthesolution.Indeed,themethodpartitionsthedataintoatmostk disjointsetsof
adjacentvaluesandsolvesak =1problemforeachset.
Besidesreductionofcomplexitywhenfindingtheoptimalturningpoints,piece-
wisemonotonicapproximationmethodhassomemoreadvantagescomparedtoother
approximationtechniques.First,itavoidstheassumptionthat f(x)hasaparametric
form.Second,theapproximationprocessisaprojectionbecause,ifitisappliedto
the approximated values, then no changes are made thereto. Third, the method of
[6]issofastinpracticethatitallowsrunningforasequenceofintegersk ifasuit-
ablevalueisnotknowninadvance.Fourth,duetothenatureoftheconstraintson
theapproximationcomponents,anyirregularerrorsinthedatadonotcauseripples
to the piecewise monotonic fit. Moreover, any peak in the data does not introduce
perturbationsawayfromthepeak.
ThepiecewisemonotonicapproximationmethodisoutlinedinSect.2.InSect.3,
themethodisappliedtobothCovid-19newcasesanddeathsdataoftheUK,some
technical aspects of the fit are provided, the estimation of peaks is illustrated, the
PiecewiseMonotonicFittingforCovid-19DataAnalysis… 3
resultsofthecalculationareexplained,andtheeffectivenessofthemethodisdemon-
strated.InSect.4,theresultsofSect.3areusedinfurtheranalyzingtheprogressof
theCovid-19pandemicintheUK,inconjunctionwithvariousimportantdatesand
eventspertainingtothetimelineofthepandemic.InSect.5,someconclusionsare
presentedandthepossibilityofexpandingthisresearchisdiscussed.
AlltheexperimentswerehandledroutinelybytheFortranprogramofDemetriou
[2],whichimplementsavariantofthemethodoutlinedinSect.2.Weprovidedjust
thedatasequencesandthevalueofk;theoptimalturningpointsandthesmoothed
valueswereautomaticallydeliveredbythemethod.
Piecewisemonotonicdataapproximationmaybeappliedtoavarietyofsituations
inwhichpeakestimationsarerequired,butsufficientinformationislackingtostate
a parametric form for the underlying function. Such situations are quite common,
so applications of piecewise monotonic approximation can occur in several fields.
Examplesarisefromtimeseries[8],fromspectroscopy[3, 5]andfromcurvefitting
to physical data, to mention few. More applications appear in the reference list of
[4].
2 PiecewiseMonotonicApproximation
Let {(x ,φ ): i =1,2,...,n} be pairs of real numbers, where the abscissae {x :
i i i
i =1,2,...,n} areinthestrictlyascending order x < x <···< x ,andwhere
1 2 n
φ is the measurement of some unknown underlying function, or a process, f(x)
i
at x .Duetoerrorsofmeasurement,itispossiblethatthenumberofsignchanges
i
inthesequence{φi+1−φi : i =1,2,...,n−1}ismuchgreaterthanthenumber
inthesequence{f(xi+1)− f(xi): i =1,2,...,n−1}.Thepiecewisemonotonic
approximation method makes the least change to the data so that the new values
{yi : i =1,2,...,n}haveatmostk−1signchangesinthesequence{yi+1−yi :
i =1,2,...,n−1} , where k is any positive integer. In this paper we define the
“least” change to the data that gives the smoothing condition by choosing the L
2
norm in Rn. Therefore, the piecewise monotonic method calculates numbers {y :
i
i =1,2,...,n}thatminimizetheexpression
(cid:2)n
(cid:3)(y ,y ,...,y )= (y −φ )2, (1)
1 2 n i i
i=1
subjecttotheconstraints
(cid:3)
ytj−1 ≤ ytj−1+1 ≤···≤ ytj, j isodd , (2)
ytj−1 ≥ ytj−1+1 ≥···≥ ytj, j iseven
where{t : j =1,2,...,k−1}areintegersthatsatisfytheconditions
j
1=t ≤t ≤···≤t =n. (3)
0 1 k
