NeerajPriyadarshi,SanjeevikumarPadmanaban,KamalKantHiran, JensBoHolm-Nielson,RameshC.Bansal(Eds.) ArtificialIntelligenceandInternetofThingsforRenewableEnergySystems De Gruyter Frontiers in Computational Intelligence Edited by Siddhartha Bhattacharyya Volume 12 Artificial Intelligence and Internet of Things for Renewable Energy Systems Edited by Neeraj Priyadarshi, Sanjeevikumar Padmanaban, Kamal Kant Hiran, Jens Bo Holm-Nielson, Ramesh C. Bansal Editors Dr.NeerajPriyadarshi Dr.KamalKantHiran CTiFGlobalCapsule DepartmentofComputerScienceEngineering DepartmentofBusinessDevelopmentand SirPadampatSinghaniaUniversity Technology Udaipur-ChittorgarhRd AarhusUniversity Bhatewar313601 Herning Rajasthan Denmark India CommunicationandNetworkingDepartment [email protected] RenewableEnergyLaboratory PrinceSultanUniversity Dr.JensBoHolm-Nielson Riyadh DepartmentofEnergyTechnology SaudiArabia AalborgUniversity [email protected] 6700Esbjerg Denmark Dr.SanjeevikumarPadmanaban [email protected] CTiFGlobalCapsule DepartmentofBusinessDevelopmentand Dr.RameshC.Bansal Technology DepartmentofElectricalandComputer AarhusUniversity Engineering Herning UniversityofSharjah Denmark UniversityCityRd CommunicationandNetworkingDepartment UniversityCity RenewableEnergyLaboratory UnitedArabEmirates PrinceSultanUniversity [email protected] Riyadh SaudiArabia [email protected] ISBN978-3-11-071379-4 e-ISBN(PDF)978-3-11-071404-3 e-ISBN(EPUB)978-3-11-071415-9 ISSN2512-8868 LibraryofCongressControlNumber:2021945435 BibliographicinformationpublishedbytheDeutscheNationalbibliothek TheDeutscheNationalbibliothekliststhispublicationintheDeutscheNationalbibliografie; detailedbibliographicdataareavailableontheInternetathttp://dnb.dnb.de. ©2022WalterdeGruyterGmbH,Berlin/Boston Coverimage:shulz/E+/gettyimages Typesetting:IntegraSoftwareServicesPvt.Ltd. Printingandbinding:CPIbooksGmbH,Leck www.degruyter.com Preface Thisbook presentsartificialintelligenceand machinelearning framework for renew- ableenergysystems.Themachinelearningmodelswithrespecttosolarenergystorage systempredictionsareanalyzedinChapter1.ThefourthcomponentoftheInternetof things(IoT)systemistheuserinterface;thishelpstheuserstocontrolIoT.Chapter2 highlights the study and implementation of various types of fuzzy structures using rule-basedinterfacesforsteady-stateandtransientanalysis.Chapter3providesasur- veyoftherole,impact,andchallenges,and recommended solutionsof IoTforsmart buildings.Chapter4presentsacomprehensivedesignofalow-costsmartsingle-phase energy metermonitoring system. Chapter 5 explains the IoT-basedsmart grid. Chap- ter6presentsmaximumpowerpointtrackingcontrolusingparticleswarmoptimiza- tionalgorithmforphotovoltaic(PV) panelaffectedbypartialshadingduetoshadow casting.PartialshadingcastedonaPVpanelwillproducemultiplepeaks’powerchar- acteristic curve, thustracking the global peak becomesa challengeespecially under dynamicchangingpartialshadingcondition.Chapter7discussesawirelesssystemfor monitoringandcontrollingofanelectricalsubstationusingNodeMCU(Wi-Fimodule), Arduino Uno (microcontroller), ThingSpeak server, and Blynk. The different pa- rameters of the substation such as current, voltage, power factor, frequency, and temperature are monitored using various sensors and electronic components. Smart grid–based big data analytics using machine learning and artificial intelli- gence has been discussed in Chapter 8. Chapter 9 presents the IoT-based intelli- gentsolarenergy-harvestingtechniquewithimprovedefficiency. https://doi.org/10.1515/9783110714043-202 Contents Preface V AbinayaInbamani,PrabhaUmapathy,KathirvelChinnasamy, VeerapandiyanVeerasamy,S.VenkateshKumar 1 ArtificialintelligenceandInternetofthingsforrenewableenergy systems 1 BibhuPrasadGanthia,SubratKumarBarik,ByamakeshNayak 2 PowercontrolofmodifiedtypeIIIDFIG-basedwindturbinesystemusing four-modetypeIfuzzylogiccontroller 41 B.Gunapriya,A.Singaravelan,M.Karthik,M.Mahesh,K.Sudhapriya 3 AnIoT-basedapproachforefficienthomeautomation 91 ZhangRuiChoo,HungyangLeong,RodneyH.G.Tan 4 DesignandimplementationofIoT-enabledsmartsingle-phaseenergy metermonitoringsystem 123 MonikaYadav,AmritanshMehrotra,DevenderKumarSaini 5 Internetofthings(IoT)-basedsmartgrids 165 JiaShunKoh,RodneyH.G.Tan,WeiHongLim 6 Maximumpowerpointtrackingcontrolunderpartialshadingconditions usingparticleswarmoptimizationalgorithm 185 VarshaSingh,SimranBajaj,ShreyashGanvir,SwapnilSahu,MahendraGavel 7 WirelessmonitoringofsubstationusingIoT 215 SubinKoshy,S.Rahul,R.Sunitha,ElizabethP.Cheriyan 8 Smartgrid–basedbigdataanalyticsusingmachinelearning andartificialintelligence:asurvey 241 EmdadulHoque,DipKumarSaha,DibakarRakshit 9 IoT-basedintelligentsolarenergy-harvestingtechniquewithimproved efficiency 279 Editor’sBriefBiographies 303 Index 307 Abinaya Inbamani,Prabha Umapathy,KathirvelChinnasamy, Veerapandiyan Veerasamy,S.Venkatesh Kumar 1 Artificial intelligence and Internet of things for renewable energy systems Abstract: The sustainability incredibly insists in having innovations in renewable energy. To obtain an unsullied and a well-grounded environment, innovations in the present mechanisms have to be uplifted ensuring a predictive framework and anenormousoutcome canbeexpected.The goalneedstobemettoestablishnew research concepts and to move on the energy requirements in optimization of the existingmachinelearning(ML)framework.Withthelotmorerenewableenergyex- istinginnature,thetwomostvariableandcommonlyusedrenewableenergiesare solarandwind.Itsconsarenonuniformityofpoweranditsdependencyonexternal environmental factors. Due to this, sole dependence on renewable energy is not possible,andhenceconventionalpowergridisalsotobeconsideredwhenanysort of predictive analysis needs to be done. Hence, more concentration is to be made onforecastingofrenewableenergyandonsmartgridsensuringcontinuousequilib- rium and balance within renewable energy and conventional grid. The electricity demandandsupplyofpowercanbepredictedusingMLalgorithmsensuringbetter savings with operational costs. The two-way electricity and information flow will provesmartgridinfutureensuringcontinuousmonitoringofthenetworkbringing more requirements for ML framework. The early warning systems incorporated by Germany symbolize a very good example of how ML algorithms analyze the real- timedatafromvariousrenewableenergysourcestoanalyzethetotalamountofenergy requirementofthecountry.GoogleinventiononDeepMindprovesthattheenergyeffi- ciencyis3.5timesmorecomparedtotheenergydemandsofthelast5years.Theinno- vations on intelligent home energy management systems prove promising energy usagewithMLinrealtime.Theconsumerorthecustomerbehaviorcanbepredicted easilywithintelligenttechniquesalongwithvariousotherfeatureslikeweatherorcli- mate modeling ensuring a complete and interoperable frameworkthatsuits the con- ventional grids. Artificial intelligence helps in sustainability of the grid with more focusondemandresponse.Theenergymanagementandoperationcostmanagement insmartgridisapromisingfeatureincorporatingMLalgorithmsinrenewableenergy. AbinayaInbamani,DepartmentofElectricalandElectronicsEngineering,SriRamakrishna EngineeringCollege,Coimbatore,e-mail:[email protected] PrabhaUmapathy,KathirvelChinnasamy,S.VenkateshKumar,DepartmentofElectrical andElectronicsEngineering,SriRamakrishnaEngineeringCollege,Coimbatore VeerapandiyanVeerasamy,AdvancedLightning,PowerandEnergyResearch(ALPER),Department ofElectricalandElectronicsEngineering,FacultyofEngineering,UniversityPutraMalaysia (UPM),UPMSerdang,Selangor43400,Malaysia https://doi.org/10.1515/9783110714043-001 2 AbinayaInbamanietal. Theoptimizationonmanagingtheassetalongwithitsmaintenance ensuresefficient managementofpower.Thelargeamountofdatacollectedbringsdataminingandpre- diction,therebyenablingmoreanalyticsondata.Toestablishsuchakindofplatform, varioustransfermodelsforsolarandwindneedtobeappliedalongwithirradianceto powermodels incorporating blending of information along with its categorization in MLalgorithms.FeatureselectionmethodsalongwiththediversifiedalgorithmsinML for diversified applications ensurecontinuous upgradation in framework ratherthan integratedapproach. Keywords:renewableenergy,artificialintelligence,machinelearningframework 1.1 Solar energy storage systems and grid- connected PV systems To consider being an infinite feature, renewable energy is significantly acquired fromnaturalresources.Sincethenaturalresourcesarecapableofprovidingrenew- ableenergyendlessly,itcanbeutilizedinfinitely.Regardlessofanyvulnerablepol- lutant’semission,therenewableenergycanbeharnessed,whichareconsideredas a significant feature. Renewable energy generation sources are very essential for providing green energy to remote rural areas as well as urban areas. Some of the renewableenergyresourcesaresolarpower,windpower,biomass,oceanwaveen- ergy, geothermal energy, hydroelectric power, and tidal power. Renewable energy sourceisoftentermedasalternativeenergyresources,sinceitisconsideredbeinga replacementtofossilfuelsandnuclearpower.Thereareafewaspectsofrenewable energyasfollows: – Existenceofrenewableenergyisperpetual,anditisinfiniteintheenvironment. – Alwaysabletobeharnessed,unlimitedly. – Knownbeingacleanalternativeenergythanfossilfuels. – Canbereplenishedconsistently. 1.1.1 Overview of solarenergy system Solar energy is one amid massively available renewable energy source and best cleanest energy source. Solar energy’s key source is sun’s radiation obtained from the Sun. Solar-powered electrical generation depends on photovoltaic (PV) and heat engines. The solar energy usage is always anticipated to proliferation around theworldintheforthcomingcenturies.Mainly,middleandlowlatitudes mightbe thechiefenergysourcewhichimproveseconomyafterglobalandlocalcrises.Solar energy is pollution free and comparatively no greenhouse gases are emitted. Solar technologies are generally characterized by two methods: (i) passive solar and (ii)