IEA-EBC Annex 86 Energy Efficient IAQ Management in Residential Buildings

AIVC April Workshop

Marc Delghust co-lead ST5 Series of four webinars IEA-EBC Annex 86 organised in collaboration with IEA-EBC Annex 86 ‘Energy efficient IAQ management’

April 1, Building ventilation: How does it affect SARS-CoV-2 transmission? April 8, IAQ and ventilation Metrics April 13, Big data, IAQ and ventilation - part 1 April 21, Big data, IAQ and ventilation - part 2

www.iea-ebc.org

IEA EBC Annex XX – May 2020

Jelle Laverge, Ghent University

Annex text Energy Efficient IAQ Management in residential buildings

This annex will propose an integrated rating method for the performance assessment and optimization of energy efficient strategies of managing the indoor air quality (IAQ) in new and existing residential buildings. IEA-EBC Annex 86 To achieve this, we will gather the existing scientific knowledge and data on pollution sources in buildings, look at the opportunities that spring from the rise of IoT connected sensors, study current and innovative use cases of IAQ management strategies and develop road maps to ensure the continuous performance of the proposed solutions over their lifetime. In the annex, experts from different fields including mechanical engineering, building science, chemistry, data Energy Efficient IAQ Management science and environmental health will work together with other stakeholders towards consensus on the basic assumptions that underlie such a performance assessment and practical guidelines and tools to bring the results to in residential buildings practice. The goal is to accelerate the development of better and more energy efficient IAQ management strategies to address rapidly changing expectations of the home environment due to challenges such as peak oil, climate change or pandemics.

1. Background, research issues and justification

The energy performance of new and existing residential buildings needs to be radically improved to meet ambitious climate change goals and residential buildings are by far the largest component in the total building stock. A central boundary condition in constructing energy efficient buildings is doing so while maintaining a healthy, acceptable and desirable indoor environment. In its mission statement, the IEQ-Global alliance (IEQ_GA) states that Indoor Environmental Quality includes the thermal environment, the indoor air quality, lighting and the acoustic environment that occupants experience in buildings. While ventilation is the main strategy that is adopted for IAQ management, other technologies influencing IAQ (e.g. air filtration) are available as well and a large number of ventilation strategies exist. There is, however, no coherent assessment framework to rate and compare the performance of IAQ management strategies. This annex will therefore focus on assessing the performance trade- off between and identifying the optimal solutions for maximizing energy savings while guaranteeing a high level of indoor air quality in new, renovated and existing residential buildings. 2

1.1 Background The annex will source from the knowledge on general residential ventilation that has been developed in completed EBC annexes, namely Annexes 8, 9, 18, 23, 27, 66, 68 and the AIVC (Annex 5). Below, the most important input from the final reports of these annexes and the open questions in light of the context given above are summarized.

v2.4 1 www.iea-ebc.org

1.3 Justification From the overview of the state of the art, it is clear that the issues raised in the previous section can’t be solved directly from existing knowledge. Partial answers to each of these issues are available, but a new annex is needed to address the gaps and integrate the available solutions in a coherent and operable rating method. International collaboration is a prerequisite for this effort since market access for innovative IAQ management strategies is currently blocked in many countries due to all kinds of prescriptive regulatory constraints. With the methods developed in the annex, we will be able to generate the necessary body of evidence to take regulatory action to overcome these barriers, generate consensus, open these markets and create a level playing field, which today is limited by very sparse and inconsistent approaches in the different member states.

2. Aim, Objectives and Scope

As mentioned above, the goal of this proposed annex is to work in an international collaboration to create an integrated general assessment method to operationalize the air quality approach suggested by IEA EBC Annex 9 to support the development, rating and implementation of innovative and highly energy efficient IAQ management strategies.

An IAQ management strategy is understood to be any coherent set of measures by a stakeholder in the building that aims to improve IAQ. www.iea-ebc.org

Annex 86 2.1 Aim The annex proposal aims to improve the energy efficiency of the IAQ management strategies in operation and to improve their acceptability, control, installation quality and long-term reliability.

Key objectives IEA EBC Annex XX – May 2020 The Annex has the following specific key objectives: - To select metrics to assess energy performance and indoor environmental quality of an IAQ Jelle Laverge, Ghent University Scope and Goals management strategy and study their aggregation - To improve the acceptability, control, installation quality and long-term reliability of IAQ management Annex text strategies by proposing specific metrics for these quality issues Energy Efficient IAQ Management in residential buildings Provide a framework to improve - To set up a coherent rating method for IAQ management strategy that takes into account the selected

energy efficiency of IAQ management for metrics This annex will propose an integrated rating method for the performance assessment and optimization of energy - To identify or further develop the tools that will be needed to assist designers and managers of efficient strategies of managing the indoor air quality (IAQ) in new and existing residential buildings. buildings in assessing the performance of an IAQ management strategy using the rating method To achieve this, we will gather the existing scientific knowledge and data on pollution sources in buildinrgesidentials, look at buildings - To gather existing or provide new standardized input data for the rating method the opportunities that spring from the rise of IoT connected sensors, study current and innovative use cases of IAQ - To study the potential use of smart materials as (an integral part of) an IAQ management strategy management strategies and develop road maps to ensure the continuous performance of the proposebd othsolut inewons construction and refurbishment - To develop specific IAQ management solutions for retrofitting existing buildings over their lifetime. - To benefit from recent advances in sensor technology and cloud-based data storage to systematically In the annex, experts from different fields including mechanical engineering, building science, chemistry, data improve the quality of the implemented IAQ management strategies, ensure their operation and science and environmental health will work together with other stakeholders towards consensus on the basic improve the quality of the rating method as well as the input data assumptions that underlie such a performance assessment and practical guidelines and tools to bring the results to - To improve the availability of these data sources by exploring use cases for their providers practice. - To disseminate about each of the above findings. The goal is to accelerate the development of better and more energy efficient IAQ management strategies to address rapidly changing expectations of the home environment due to challenges such as peak oil, climate change or Relation to the EBC strategic plan 2019-2024 pandemics.

1. Background, research issues and justification v2.4 The energy performance of new and existing residential buildings needs to be radically improved to meet ambitious 5 climate change goals and residential buildings are by far the largest component in the total building stock. A central boundary condition in constructing energy efficient buildings is doing so while maintaining a healthy, acceptable and desirable indoor environment. In its mission statement, the IEQ-Global alliance (IEQ_GA) states that Indoor Environmental Quality includes the thermal environment, the indoor air quality, lighting and the acoustic environment that occupants experience in buildings. While ventilation is the main strategy that is adopted for IAQ management, other technologies influencing IAQ (e.g. air filtration) are available as well and a large number of ventilation strategies exist. There is, however, no coherent assessment framework to rate and compare the performance of IAQ management strategies. This annex will therefore focus on assessing the performance trade- off between and identifying the optimal solutions for maximizing energy savings while guaranteeing a high level of indoor air quality in new, renovated and existing residential buildings. 3

IEA EBC Annex 8 (1984-1987) dealt with the impact of occupant behavior on ventilation, its consequences from an energy point of view, and the suitability and applicability of different methods of influencing inhabitants' behaviour. It focused on window airing. It concluded that: Annex 86 - ventilation behaviour (its frequency and duration and its underlying motives) is related to the type of room in which it occurs; List of annex participants per country: Australia: CSIRO Austria: University of Innsbruck Partners Belgium: UGent, KUL, BBRI, University of Antwerp v2.4 1 Brazil: Pontfical Catholic University of Parana Canada: NRC Chile: PUC 42 institutes from 24 countries China: Nanjing University, BUCE and Tsinghua University Denmark: DTU and Aalborg University Copenhagen Active participation by companies Finland: Aalto University encouraged! France: La Rochelle University, ENS PSL, CEREMA, Université de Lille, UPJV and CETIAT Germany: TH Rosenheim Ireland: NUIG Italy: EURAC research center New Zealand: BRANZ Netherlands: Technical University of Eindhoven, BBA/TUDelft and Zehnder Norway: Oslo Metropolitan University and SINTEFF Portugal: University of Coimbra, Polytechnic Institute of Viseu and University of Porto Singapore: National University of Singapore Spain: Eduardo Torroja Institute for Constructon Sciences – CSIC Sweden: Chalmers University and KTH Switzerland: ETH Turkey: TTMD United Kingdom: University of Strathclyde, Lancaster University and University of Nottingham USA: Syracuse University, UMD, UTexas and LBL

4 www.iea-ebc.org

In addition to a rating method, the annex will provide the target audience with tools for the development, design and quality testing based on the performance and practical results of a proposed IAQ management strategy. With these tools, manufacturers and practitioners will be able to identify quality products and select the most appropriate systems for their applications and projects.

Building owners and occupants, facility managers, environmental health professionals and researchers not participating in the annex are secondary stakeholders for the proposed annex

3. Means

By mapping the existing work on the pollution sources in residential context (including occupant activities and the penetration of outdoor pollution), we want to develop a consistent set of metrics that allows us to assess the performance of the various technologies in terms of energy efficiency, comfort and health for the occupants. This includes extending the framework developed in Annex 68 with specific metrics for energy efficiency and particulate matter, explicitly including moisture control and HVAC component modeling as well as creating a common methodology for IAQ data sharing among smart devices. By pooling and analyzing the data, the range of conditions in dwellings can be better understood and the most appropriate energy efficient IAQ management strategies can be identified. The application of the metrics over time will also allow continuous commissioning to ensure that the predicted energy efficiency is actually maintained over the lifetime of the system. Another advantage is the possibility of conveying the information towards building occupants, building operators and facility managers in a coherent way. Ensuring performance over the lifetime of the system, however, requires continuous commissioning of the technical components of the system as well to prevent deterioration of the performance. These activities will be structured in 6 subtasks as described below.

3w.1w Mwe.tiheoad-eoblocg.yo rg

Aw rwanwg.eie ao-fe mbce.tohrogd s will be used to accom plish the work in this annex, ranging from literature review to e xperimental lab work. There will be face to face consent workshops, for the development of the rating method. www.iea-ebc.org There will also be a considerable amount of n umerical modeling to compare system options in identical conditions.

3 .2 Research Strategy TAh3e.3 g eneArsasle rsessineagr ecnhe srtgrya tseagvyin ogf athned aenxnpeoxs uisr ecy rcelidcuacl tcions peontte bnutiladl ing. The 6 subtasks proceed in parallel, each with NJU: Small scale chamber tests will be conducted to obtain SVOCs and VOCs emission data. their speNcuifmic ewriocrakl psliamnu, lmateiothnos dtso a sntdu dgoy atlhs,e b eunt eyregty c osoavrdininga aten dcl oesxeplyo swuhree rree ndueectdioend tpoo atechntieiavle otfh et hoev enreawll gsomala ortf TU: Emissions from occupants are available from past research projects (new data may also be available from their Dth1e. 2A nneAmx la.i tAteelrtrahiatoulsur geinh r retehvseiied sweun batntiadsl ksbysu nwitldhoiernksg ilsa ruegnpedloyer rit n do pinfaf aerpraepllnertol ,cp tlrhimieatiaret w imco corodkn eidsli inctligo tsnoesol.y l sin, rtelrecvoannnte mctoedde, lwinigth a tshsue mrepstuioltns sf raonmd o cwcwupwa.nite tae-setbincg. ochrgam ber--please check with Prof. Xudong Yang). the diffemreonnti toarsiknsg f eperodtinogco inlst o the work of others. This requires an iterative process of interaction with the whole BDB TRUI :a wnde UcaMn Dco: Enxtrpiberuitmee wnittahl (aSn)dV OmCo edmelilsinsgio snt uddayta o ffr tohme ienntergioyr spaaviintgs paontde nvatiranl ioshf eths.e new functional materials Dgr1o.u3 p. CAo nretrpibourtt ionfg u fsreo mca tshees ior fo cwonh earcetinvtit siests, tohfe p seurbfotarmskasn wceo rmk etotrgiecst haenrd t omwaatcrdhsin ag steoto olsf 4 common deliverables. U inA rSeosSid: ewnet iwali lbl ureilvdiienwgs a unndd teers dt iaffvearielanbtl ecl ismouartcees m odels and coordinate further developments TBhUeC E6A-m aonndt hDlTyU m: Eexeptienrgims (ein tps eornso enx poor svuiare i nretedruncettio) na rpeo utesendti atlo o ef xtchhea negwe smtaattuesr iaclsr oins sr ethsied edniftfiearle bnuti lsduinbgtass ks and Pwarwtnwer.is e ian-veoblvce.do:r g Pm aortnniteorrs tinhveo olvedra: l l convergence of the work. At the end of these meetings, the work plans for the next 6 months Ua roeN d (isUcKu)s, sPeUdC in ( Cphleilnea),r By BaRnId ( BuEp)d, aUtGede nbta (sBeEd) ,o Uno tSh e(U laKt)e, sUtL fRin (dFiRn)g, sL iUn (tUhKe )o, tDhTeUr s(DubKt)a sks. U PaArStonSe r(sA iTn)v, oUlvLRed (:F R ), ULille (FR), LU (UK), NU IG (Ire), UoS (UK), UoN (UK), EPFL (CH), SU (USA), NJU (CN), TU (CN), DTU (DK), SU (USA), UMD (USA), UPJV (FR), NJU (CN), BUCEA (CN), BBRI (BE) BStBaRkIe (hBoEl)Td,h eCirsEs Ta iIncAtviTov il(tvFyeR dw),: i Pll UfoCc (uCsH oilne )t,h BeR sAtaNtZe -(oNfZ-t)h e-art on improved and continuous commissioning and optimisation

- Poof lIiAcyQ mmaakneargs eamnden rte sgyusltaetmorsy u bsoindgie Ios Tw sielln bseo rasb alned t ocl ouusde dthaitsa m. Eextahmodp lefos ro tfh mee rtahtoindgs eoxf pelnoreerdgy in e tfhfiicsi eancti viaitqy 3St.3a kSeuhbotladsekrss ainnvdo dlvievdis:i o n of work www.iea-ebc.org Stakeholmcdaeanrns ab igene vmoolevnneitdt o:s ryisntge msyss tems based on IoT sensors and/or internet enabled ventilation systems, which allow S- u b task M1 Maneutfraiccst uarnedrs d oevf ebloupildminegn tm oaf taenr iIaAlsQ smhalnl abgee imnveonltv setdr arteeggayr draintign gte msteinthg oadn d possible co-development of Regu- latoMtrhyae nb buoufdaiilecdtsiun, rgee nrovspi r(eeorgna.mt oeern nVttoea rilid dhe e(nUatliStfhAy )po) prwoeifrlela sbtsieoio nan ballneso tamon adul sirees st heliaekre mc huenetrbhsao wlda intlloc be fedu ratbihrleflro t wdoe srv,e emfleoarp ltf ouan tcdht eibo ecnoninlclhgecm bteayd-rpk ia ntshpseu itnr Annex 86 This subptarsokd iusc dtse vthoatet dh atvoe t hfuen dcetivoenl otpom abesnotr obf ina dgoeonre praoll lruattaingts m. ethod for the benchmarking of the performance data phreoadtu rcetc poovertrfyo oliro faulty sensors in particular rooms. These methods can be based on automated analysis of o- f IAQ mBuainldaignegm deenstig nsyesrtse, mhesa. lItnh oardgdaitniiozna titoon sre alnevda tnetc hmneotlroigcisc, aal insestti touft easp wprhoop mriatkee toesotlisn, gc foonrs iinstdeunstt rmy aondde lriunng sensor data, potentially complemented with cloud-based interactions with the user (e.g. via apps). assumpttihoenisr alanbde mllinogn istyosrtinegm psr aorteo caolslos aarme oanlsgo p portoepnotsiaeld s.t akeholders. D eliverables: v2.4 DSDu2Tb.U1ta wskil Rl2 ec Spooonurtrtrc iobenu c thseoa uwrarictehte cdrhiazatarata iofcrtnoe marinz adset tivoyepnri acalan lcd ae tsxyepp osisctuarld eeie xinsp oirnes suairdpeea inrntt miraeel sbnidut eiblndutiiniladgli snb gusil dinincglusd. ing both decentralized and ADcetliivietrieasb:l e s: 8 IEA EBC Annex XX – May 2020 DTrho2io.s2m S T-b caArsenea ditn etvsee rcnoteniltsa-ibtsitaoesnne tds iyonspteumnt vasa.c lcuWeesess fdaorare tta hcbeua rasreses noetfsl seym weisonsrtik omine grt ahotoends modfee ivtnehdloodposer dfso oirnu raScTne as1l. y asnisd acnodn trvoisl usatrliaztaetgioiens oinf SdTa 4ta. AD13..1 SAe dleacttaibnag spee orfo ermiasnsicoen m anedtr itcrsa.n sport properties of the smart materials developed in the project for active DIcto 2s.ltl3ae rcttse Afdr odfrmaotm ainb fraoesrsemid oaentni orteinas lai dvveaniltailbaalt eIiA oinQn l.si tyesrtaetmurse. ,T ahdisd winogr nke wwil el cxopnetrinmueen utanld rers uthltes awnhneerxe. n eeded and reviewing and SIAtaQr tcionng trforol.m a thorough analysis of the metrics currently used in different partner countries, those dCeEvTeIAloTp icnogu mldo bdee lisn (temrepstireidca tl,o s ecomnit-reimbuptieri ctaol tohri sp htayskic ably m woodreklisn)g f oorn c haorwa cttheeri zdinatga r efrloevma ntth ere csoidnetnrotila sl esonusorcrse sc.a n Jelle Laverge, Ghent University Workplan D3.2 Mechanistic models for estimating the energy saving and exposure reduction potential of the new materials Aprreoavsid oef d ipneafsorctriimcbuealdati roi ninn l tiuteesrereafsuttul arfoere rd :t ehdeic matoendi ttor sinmga arnt dve mntailiantieonna anncde ionf Athnen esxy s6t8e,m thsi.s activity addresses how to include, www.iuenad-ebr cre.oalrigst ic environmental conditions. The database and models will be published in scientific journal SUuLbiltlaes dk ei3nv eSamldodapiritnti ogmn ra ettoae lhrtiemaalsel t ahos,n oalitnh eIeA srQ yfsa mtcetamonrssa fgsouercm IhAe aQnst m csotormantfiteoogrrtyi,n pgr,o odcuccutpivaitnyc,y u…s evri sfuriaelnizdaltinioens,s ,a dnuarlyasbiisl.i ty, sustainability, Annex text - Saortuicrlceess anfrdo min ab purilodjiencgt rmepaotertr.i als and pieces of furniture (review and consolidation of models and T his ST idreesniltiiefinecse o apnpdo retcuonnitoiemsi cto f aucsteo trhs e( ibnuviel dstimnge snttr uacntdu roep aenradt (ibonioa-lb caosestds)) binu iald irnegs imdeantteiraila lcso (nfotecxuts. sTinhgis o in chluemdeps Energy Efficient IAQ Management in residential buildings D3.3 pCahraammbeetre rtse)s t data to validate the simulation model and confirm the effectiveness of selected smart cPoanrctnreetresp) ri naovnpodol svtiehndeg: na o mvelt hfuondc ttoio angagl rmegaateteri athlse i nsesliedcet eitd t mo eatcrtiicvse flyo/rp oapstsiimveizlya tmioann paugrep tohsee sI.A Q, for example, through 6 Subtasks - Imndaoteorri aslso uorc perso fdruocmts .o ccupants and occupant activities (cooking, cosmetics, cleaning…) and occupancy active pa UinGte, nwta (lBlbEo),a DrdTsU, t(eDxKt)il,e LsU c (oUaKte),d U wLiltlhe (aFdRv)a, nCcEeTdIA sTo r(FbRe)n ts or hemp concrete, and quantifies their potential This annex will propose an integrated rating method for the performance assessment and optimization of energy patterns (partly drawing on the outputs of Annex 79) efficient strategies of managing the indoor air quality (IAQ) in new and existing residential buildings. b- ased onHU teohaNet:i nawsgis lael nlsesdam cdeo tnohtke if nrliagtme arpeapwtuloirarenk cr deevsi ve(ewwlo aponedd a sinnud mS gTma 1sa. s r itzoev eits f,o frir peupblalciceast…io)n U oN: will carry out a health impact study of known pollutant in houses and prioritize them to determine criteria To achieve this, we will gather the existing scientific knowledge and data on pollution sources in buildings, look at -S ubtask OS4tu Eatnkdesouhororinl dsgoe uprsre crinefosv ro(mflvraoenmdc: ea iorf a snmda frto vme ngtriolautniodn) ApDocetllliiuvvit etairenastb:s l e( tsh: i s will link with ST2) the opportunities that spring from the rise of IoT connected sensors, study current and innovative use cSTase s1 o fand IAQ 2: methodology -T his -subBtmaiosaklno fugoficauaclts uearcse toirvnsi tpyor f(affecutrnicngagal l cIgornTo dwcitoihon)n se cthteadt aIsAsQur em roelniaitbolrei,n cgo satn edf femcatinvaeg aenmde rnotb udsetv iimcepsl ermeseenatracthio nin osft istmutaerst, management strategies and develop road maps to ensure the continuous performance of the proposed solutions A-D 34..11 MSPRgeUoeacvCtoie:en rwdniaiam llrl r epeycpr oneontmparttelri isoarbstgnuieo tIsnAe cs aQt inode ad sutn ehacd ehnt oadelirn tiraneecrdgtagoeutyorlu aiprzrt eaoct rhrrifyoeo vmnbrim oeoisdwaft inrt eyhcs e( e po.g rf o.s,d mOua3cr-ttisn viteinatieladt rioena c(taicotnivsi)t y 4.1) vDe4n.2ti latioInnv.e Tshtiigs ainticolnu dreesp boort h: inIAstQa llantido ne annedrg oyp epreartfioornm. Aan pcoeos,r pdeurrfaobrimlitayn caen do f osmccaurpta vnetn tinilatetiroanc tsioysnt eomf s5 c asnm naort over their lifetime. -f or LBiUeted: rwraoitolul mrceo ensnutrvriivbreouynt ema nteodn ltliatse broartautroer yre tveisetwin g to gather relevant data and existing knowledge about properties ST 3 and 4: application to technology only leadv etnot iwlaatsioten ostfr eatneegrigeys and aggravated IAQ. It can also create a bad reputation of smart ventilation among In the annex, experts from different fields including mechanical engineering, building science, chemistry, data -U oS -c anfEc opocrion dtncrealtmurnsibsiioopulnotesrgt yf,tr roaes mtpteh entecht tieosl i ndt(ea aaritranabtd oua enrnema eliyr stessriviasoi nensw mo foi scnshi eoImAn Qipc aptlh assrua)b msteatnecres/s amnedt mricosi satunrde imn entehwo dfuonlocgtyio ndaelv melaotpemriaenlst ( efogr. rDe4le.3v antI nstsapkeechtioolnd eprrso t-o dceosl ifgonre rress, idinesntatilalel rssm aasr tw velnl tailsa toioccnu epnasnutrsi.n Tgh tihse, ignu tahraen etneedd, cpaenr floeramda tnoc ea doovpetri othne o liff emtiomre science and environmental health will work together with other stakeholders towards consensus on the basic a sses- sinMgu sIeAetQ acla -inoser egsn aoenfr itgch yef reiarf fmdicaeitewan otr hkos m(MesO Fs), precise humidity control material (PHCM), hemp concrete, etc.). A ST 5: new opportunities through IoT p rimitive, less efficient (in terms of energy use) and less effective (in terms of IAQ) forms of IAQ management. The assumptions that underlie such a performance assessment and practical guidelines and tools to bring the results to AB BctRivI:i tcieasnt: as retainrcgh p fooinr tc writilel rbiae bthaes efdur othne cr usrtruednyt orfe tshuelt si monil afirlittriaetsi obne towfe PeMn V, OpeCr faonrdm manociset uarses edsisffmuseinotn .o f ventilation s ubtask defines a smart ventilation according to the AIVC: “Smart ventilation is a process to continually adjust the practice. AsDy2es.tl1iev mersa L(biItlAeQsra: at un rde E rneevriegwy) , and possibly LCA analysis ST 6: dissemination and management vSeunbttilaastkio 5n Esnyestregmy sianv itnimgse ,a anndd I AoQpt:i iomnaplrloyv beym elonctast iaonnd, vtoa lipdraotviiodne tthhreo udgehsi rcelodu IdA dQa tbae naenfdit sIo wT hciolen nmeicntiemdi zdienvgi ceense rgy The goal is to accelerate the development of better and more energy efficient IAQ management strategies to address DUc5oTG.1nUe s nuwmt:il plwT Ctdihiloleailns vsc ,esa oilucflolitacpiilabv itiotyhiyor ea bni tni/elcrel eslwu padi otfnehurdst nB iancoBt gtrRih oegIenv uria iedlnw emo lnoian-ftIe eAe roQxiian sl tcsIio nasTgnt-s dd a(cstthuaaac frhroae rcag tesae nrrtdehizirenegrg ymit iesan flad fidocsiioseorcnr osptmo tIiuAfoorQncrt/e mcsoo aron nfvn areogeriesiemido)e.en nn (ct…tai asp)lt” arb.ab utiSlieilttdgaiiernetsgsi.n sg a fnrdo min ptuhtis This subtask is exploring the potential of the new generation of IoT connected devices (both standalone and rapidly changing expectations of the home environment due to challenges such as peak oil, climate change or Sd5Ue.2 f&i n iDtiTorUOneL ,vgw Read:irlr elwvd mjiioeinlalwig nnc tordtleyhn- gcetionri sionvbturetuyrtsto dteoilogl feto aotdrth e teshv o temhun elroitc tsiaeltapsr itap.mi otlEiupcnmrao eibr(s DtrislaeiCtonvyVnti e )o o wrfnia s tlt hienceseo s nodisfma iditinaleadsrreoeitodysr fpasorsori n ueacrn cipeslprseg e wytco iiefl litf chfb iecesi u encbneostwm eI Atpf uQiolne fmcd ts iaiomnntaoagrl te a mmnvae eitnnettrie ilsartnltrsieao. tn-eb.g aiTeshesi ds embedded in eg. AHU’s) for smart IAQ management. What can we learn from big data? Can we benchmark system pandemics. Sd5Ue.3 f&i n iUtiPodDJRnVaTe toUpwafo: bi lswrlamt js iolealin nrca totv hllvyaleae it lbnsaetotsbairtllta eaet n fieood fnwr tm thiihtneohec d a Ulsuercotdli eNtienhns eto ia fh ficlwei nmcideop emI o crmToa niumcngrpeiet lyteo eimfn ms eoynapstteeaenrtmi aoslcn acscen ufsodrsr .se einmnteulyrl gaaytve ea ifitflsaic bpieloent tein nIA ttQiha elm i mlaitnpeargacetu morne ndatan smdtr paoitneg gt hiees energy and IAQ performance based on this data? How can we make sure that the data is available and can be tmhear VkOetC 5dc.oe4pn ecenndtiRnraegtp ionrt a otshn we t ehtleyl pRsetHa tionef voesfe ntnhtiselai natger td p isnap roaancmelines.te e IrosT (iCmOp2le, mheunmtiadtiitoyn, s ofcocru epnaenrcgyy, e…ff)ic, iethnet IAtyQp em aonf asgeenmsienngt accessed? Can we update what we think we know about what happens in dwellings based on what we see in big 1. Background, research issues and justification UAco1Am.S2ob Sin: awSstetilortla tnlitensea,g gdt iuhe apesn atdy rpcaoet oionrfgd iimnsaettatehl lolaitdei.o rna t(ucreen rteravliiezwed /decentralized) and the types of control algorithms. The subtask data rollouts? What are the best protocols and ontologies? How to create viable services out of the data/business AUw 3Lili.ll2 lae l s: owMT eho xewdt erielanll itpdnina gtgr hot eimfc sitephctaoehtp oebed ei n thdo ate hvisniecocr uliuibtrde uersna atdthuleterer e rrt neyreaqptvuiicivieareewl mra eiersfnildotesw nf otdirais ltt hcrioeb numdtiootdinoe nplisna gtt etornosls a, nthde s ymstaetmch icnogn foigf ureraletvioannst. m o deling plans? How can we integrate data with smart grids? The energy performance of new and existing residential buildings needs to be radically improved to meet ambitious L U: can cMaosonsudtrmeibl pustetioetun tpso a laintneddr m altaoubnroietr oraertivonirgey wp tr eosttosc otols .a Rnaatlhyezer tthhaen pselrefoctrimnga n1c sep eocfi ftich eo pnteiown fmora teearciahl os ff tohre IsAeQ, t hcios natcrtoivl iitny These issues will be addressed by reporting experiences from a series of implementation case studies and overview climate change goals and residential buildings are by far the largest component in the total building stock. A central UASucLtRbi vt:ai twsiekeri ss 6e:w sf Di od lilcs epusneastemriatdiil cn obiapnuta isltodyenis n,itn gem msta.h aTnetha ilceigta eblmrleyahe taunnvrteai oa lrynuezrdvi oninefg wtt eth hreae mc ptairootnep reiartlise os vaenr dt ime cuhnadneirs mdisf foerf etnhte c IlAimQa mteas nwaiglle bme eannta slytrzaetde agnieds of available (types of) datasets. boundary condition in constructing energy efficient buildings is doing so while maintaining a healthy, acceptable EATP4hF.e1L : f iwnaicuRlllo nascrdutoriebnrstgt aprc eisobkxnu idsattisiednisn etgugor r ca estotmshin oitastnhr rs oetau l vn cbseldtonra msatseitklae a atgtlciiohgenisn mfsgot erta nhItAete sQogef i m ewtshai tenh a tgchteeimv aitevineastil awbiilltleh b ionep dtthieoevn easl.on pnedx .a nd the interaction with the AIVC. and desirable indoor environment. In its mission statement, the IEQ-Global alliance (IEQ_GA) states that Indoor This sub task includes the outreach of the annex, eg. by managing the dedicated section of the IEA EBC webpage. It STUh:e o imn pseTlechmoisne danactatriytvi ioetmny cisnascsileou nsdtseu,s dV aiOe sCr eswv, iiclelo wsoh kooiwnf gce axesimset iissntsgaio tkenn-soo fw-tlheed-gaer ta ubsoeu otf IcAloQu da ndda tean ine rtghye pcoenrftoerxmt oafn IcAeQs omfa rneasgideemnetinatl Environmental Quality includes the thermal environment, the indoor air quality, lighting and the acoustic UuPoseJNVs: tawhnieldl d cSiofUfne:s rMiedneotrd paelplainptgfro trphmreisa e ttefhf aemtc ott hnoeift oARrHIiVn Cog npd riVaoOgvniCdo eassnt idtcos m rineotqiesurtiaurcreetd w ptoit thde etnhtteeiar bml brionuaeffd eaernri ntIAga rQcga empt aeactuirtdiicei esn ocfe .h Tehmisp t caoskn cwreiltl ea lisno NsyJsUt:e omns s SrmVeOgaarCtrs d vainengnd to iVlnaOetCi osnr ,m tohreeir o fc othste, faonllodw tihneg gcohoalics:e of several (max. 5-7) smart ventilation strategies being environment that occupants experience in buildings. While ventilation is the main strategy that is adopted for IAQ bTPeUuni:Csl duE: irmwnegi lst lmhs cieoa ntceossr infidrateoilnsrm, u as apottupciodcrounyp otrahfin aett hese f fm leincotkn oiwtfo ictrhoin a(gt hidneiga rgleansyoueslrtt si(c ffsor ror emq)u olitrtie-hldae yrt oeor ndegedot weinramgl la)in noedn a eInAn QdIAe aQdn madn etnhtereixrcem sa, le csopmecfioarllty annex 68. management, other technologies influencing IAQ (e.g. air filtration) are available as well and a large number of SU and NJU: Chamber tests will be conducted to evaluate the effectiveness of selected functional materials for BD RTAUN-: wZ i:l Clc coaonn tcisnoiduneotrui bas pucptoerm otpmor liiisatsetieora nmtiunorgne i( trsoerveiin eligwn dk iwagitnho SstTi4cs): r feaquulti rdeida gtnoo dseist,e (rpmreinvee natniv IeA)Q m maienttreicn ance ventilation strategies exist. There is, however, no coherent assessment framework to rate and compare the V OC adsorption/conversion to validate the model prediction. BABcRtiIv-:v i2tw.i4ee Cs :oc annti ncuoonutrsi bouptteim tois atthieo nli t(eseraet luinrek wreivthie SwT 4w):it sh e laf -fleoacurnsi nogn aind o-coar lisboruartcioens , ipnr Bedeligctiaivne dcwonetllrionlg s and (S)VOC performance of IAQ management strategies. This annex will therefore focus on assessing the performance trade- eAAm16..i31s- s ionU2D r-siawsets aecya mf srueiosn metaert -siconteitnnen gort.iefo rtreh pdea crieonpmt so martnusd na ivncadart nrioeissnhu: elgtssi,v oianfn gtd ha aenl dsou g boetntat sitnkhgse a foesue TtdedbcoahocNrk o stotoeu sar,nc Wed1s 3ef arbonimnda PrusMs eo riss Vs ueen.t ilation nformation off between and identifying the optimal solutions for maximizing energy savings while guaranteeing a high level of Pape- rs (TLdheeiapsr eanncintdigivn,i tgdy eo cvnoe ntlohsipes itnssu gob faj cenhcdte )ac kssinegss tihnagt: timhep nroeevidnsg o sfc tiheen tsitfaick ekhnolwdelerds gaere, amsseet swsmithe nthte f rparmopeowsoedrk rsa atinndg pmreotdhuocdt indoor air quality in new, renovated and existing residential buildings. 5 v2.4 A6.2 tdAhelrivgoenulmgohpe munste eon cft atahsneds A twnahrngeeerxet im nthgee e tpirnogpso wseitdh mAIeVtCh omdes eatrien gasp plied to current questions of the stakeholders. 12 A6.3-v 2.4 NDewdi cbautesidn esess scioasness d ourr ianpgp tlhicea AtiIoVnCs :c oe.ngf.e relna cte d to aggregation and availability of IAQ or OA Q data-sets, 1.1 Background 10 UAo6N.4: WioRll feafglpiunplelay ra mnudpe dtirnai ctrese sat ool- ntai m tchaees oe p nsrlotiungderey, s fposr rto hRbr&loeDumg o htro tr heqeau lan-tneimtwifseyl ectthoteme rhmse aiasnlstdiho sneoifncfegiac/lot msp oteifdm aiains a intcitoievnri/vt…ieen s tion in homes The annex will source from the knowledge on general residential ventilation that has been developed in completed PAU6C.5: w illL aiapispolyn mweitthri cIEsQ to-G aA c ansed sottuhdeyr parnonbelxeems to quantify the health effects of an intervention in homes EBC annexes, namely Annexes 8, 9, 18, 23, 27, 66, 68 and the AIVC (Annex 5). Below, the most important input from BS Bo mRIe: coafn t haep palbyo tvoe a m ceanseti ostnueddy g oals are strongly related to some of the aims and activities of other subtasks. Within the final reports of these annexes and the open questions in light of the context given above are summarized. DSDuTebUlitv:a eWsrkai lb5l ,lte rtsyh: e t o f oacpupsl yw mill ebteri cosn t toh ae csapseec isfitcu dchya pllreonbglesm f rtoom q uthaen tuifsye t ohfe I ohTe,a altsh IeofTf eocftfse rosf garne aint tpeortveennttiiaoln t oin s uhpopmoerst (tdheopsen adcetinvtit oiens .a vAadildaebdle c phraolljenctgse).s studied in this subtask will include: IEA EBC Annex 8 (1984-1987) dealt with the impact of occupant behavior on ventilation, its consequences from an energy point of view, and the suitability and applicability of different methods of influencing inhabitants' behaviour. Annex 86 Deliv-e v2r.a4 bDliescs:e r ning cause and effects (e.g. system b iased results, availability of control groups or control(led) It focused on window airing. It concluded that: D1.1 Ap elirtieordast…ur) e review on rel evant performance metrics. 18 - ventilation behaviour (its frequency and duration and its underlying motives) is related to the type of room - Using external or indirect/deduced values (e.g. data from other, local or remote data sources to in which it occurs; complement embedded sensors (from IAQ monitoring units, presence information from security systems, v2.4 OAQ, weather forecasts...), balancing optima l use of those external sources and robustness/fa ll-backs) - Practical concerns: GDPR & IT (soft- & hardware): focussing on specifici9t ies of IAQ control strategies and related metrics (e.g. location and durability/calibration/… of sensors, instantaneousness) (for data Workplan communication and management protocols, see other IEA-EBC Annex 81) v2.4 1 - …

It is not the aim of this subtask to deepen the knowledge on the separate fields of research related to these 6 Subtasks challenges in general (e.g. developing new state-of-the art communication protocols: see IEA-EBC Annex 81), but to focus on the specificities related to energy efficient IAQ management strategies in residential buildings. Different types of activities will take place in this subtask. The first (Activity 5.1) will focus on classification and registration of existing data-sets relevant for the goals mentioned above (e.g. IAQ, OAQ, climate data…, real-time ST 1 and 2: methodology

ST 3 and 4: application to technology v2.4 16 ST 5: new opportunities through IoT ST 6: dissemination and management

6 www.iea-ebc.org

Annex 86 ► ST 5

IEA EBC Annex XX – May 2020 Energy savings and IAQ: improvements and validation through

Jelle Laverge, Ghent University cloud data and IoT connected devices

Annex text Energy Efficient IAQ Management in residential buildings

This annex will propose an integrated rating method for the performance assessment and optimizatio-n oSmartnessf energy efficient strategies of managing the indoor air quality (IAQ) in new and existing residential buildings. (e.g. smart ventilation incl. continuous commissioning & optimization, use of remote data, ST4) To achieve this, we will gather the existing scientific knowledge and data on pollution sources in buildings, look at the opportunities that spring from the rise of IoT connected sensors, study current and innovative use c-aseKnowledges of IAQ & data-sets management strategies and develop road maps to ensure the continuous performance of the proposed solutions over their lifetime. (e.g. for defining metrics (ST1), typical exposures (ST2) In the annex, experts from different fields including mechanical engineering, building science, chemistry, data science and environmental health will work together with other stakeholders towards consensus on the basic assumptions that underlie such a performance assessment and practical guidelines and tools to bring thapplicationse results to practice. The goal is to accelerate the development of better and more energy efficient IAQ management strategies to address real-time & delayed, on-line & off-line, new business cases? rapidly changing expectations of the home environment due to challenges such as peak oil, climate change or pandemics. challenges 1. Background, research issues and justification • real-life, uncontrolled environments (cause/effect?)

The energy performance of new and existing residential buildings needs to be radically improved to meet ambitious • data quality: often limited number and lower cost sensors climate change goals and residential buildings are by far the largest component in the total building stock. A central • GDPR boundary condition in constructing energy efficient buildings is doing so while maintaining a healthy, acceptable and desirable indoor environment. In its mission statement, the IEQ-Global alliance (IEQ_GA) states that Indoor • IT Environmental Quality includes the thermal environment, the indoor air quality, lighting and the acoustic • … environment that occupants experience in buildings. While ventilation is the main strategy that is adopted for IAQ management, other technologies influencing IAQ (e.g. air filtration) are available as well and a large number of ventilation strategies exist. There is, however, no coherent assessment framework to rate and compare the performance of IAQ management strategies. This annex will therefore focus on assessing the performance trade- off between and identifying the optimal solutions for maximizing energy savings while guaranteeing a high level of indoor air quality in new, renovated and existing residential buildings. 7

v2.4 1 Series of four webinars organised in collaboration with IEA-EBC Annex 86 ‘Energy efficient IAQ management’

April 1, Building ventilation: How does it affect SARS-CoV-2 transmission? April 8, IAQ and ventilation Metrics

April 13, Big data, IAQ and ventilation - part 1 (academics)

April 21, Big data, IAQ and ventilation - part 2 (industry)

Wouter Borsboom Benjamin Hanoune Pieter Pauwels TNO University of Lille TU Eindhoven The Netherlands France The Netherlands

Webinar management

Maria Kapsalaki Valérie Leprince (INIVE, BE) (INIVE, BE)

Big data, IAQ and ventilation – part 1 webinar 2021.04.13

Objectives: To address • the applications of IoT devices and big data in IAQ and ventilation and • discuss the possibilities they provide for research. To set the starting stage for subtask 5 of IEA-EBC Annex 86

17:00 | Introduction Marc Delghust – Ghent University, Belgium 17:10 | Improving IAQ with BIM based Predictive Twins Wouter Borsboom – TNO, Netherlands 17:30 | Online personal IAQ monitoring, Benjamin Hanoune – University of Lille, France 17:50 | Brains for buildings: where to find all the relevant smart building data? Pieter Pauwels – Eindhoven University of Technology, Netherlands

18:10 | Questions and Answers 18:30 | Closing & End of webinar . 10 webinar 2020.04.13

How to ask questions during the webinar Note: Please DO NOT use the chat box to ask Locate the Q&A box your questions!

Select All Panelists | Type your question | Click on Send

webinar 2020.04.13

NOTES: • The webinar will be recorded and published at www.aivc.org in a few days, along with the presentation slides.

• After the end of the webinar you will be redirected to our post event survey. Your feedback is valuable so take some minutes of your time to fill it in.

Organized by: www.aivc.org Facilitated by

Disclaimer: The sole responsibility for the content of presentations and information given orally during AIVC webinars lies with the authors. It does not necessarily reflect the opinion of AIVC. Neither AIVC nor the authors are responsible for any use that may be made of information contained therein. 12 13

Big data, IAQ and ventilation – part 1 webinar 2021.04.13 Q&A ?

. 14 AIVC April Workshop

Series of four webinars organised in collaboration with IEA-EBC Annex 86 ‘Energy efficient IAQ management’

WOUTER BORSBOOM

TNO (www.tno.nl) is an independent and not-for-profit organization. TNO connects people and knowledge to create innovations that boost the competitive strength of industry and the well-being of society in a sustainable way. This is our mission and it is what drives us, the over 3,400 professionals at TNO, in our work every day. We work in collaboration with partners and focus on nine domains.

Wouter Borsboom Senior Business Consultant TNO

Energy Built Environment, Monitoring and assestment of dwellings and offices, energy, ventilation and health, Country Towards Networks of predictive twins in the Built representative IEA-ANNEX V: Environment, Arjen Adriaanse, Wouter Borsboom, AIVC.org, Board Member Rob Roef, 2021 https://repository.tudelft.nl/islandora/object/uui INIVE.org, BDTA. d:ba8043dd-1dfc-4469-bfeb-53006de6e88a

2 THERE IS A CLEAR NEED TO ASSESS INDOOR AIR QUALITY

DIFFERENT PROTOCOLS TO ASSESS IAQ

Example of a Dutch IAQ requirement, which parameters to use, how to measure, where to measure. www.pvegezondekantoren.nl

Theme Survey Continious Inspection Total monitoring

Thermal Noise IAQ

Light

Total IEQ

4 BUILDINGS PROVIDE TONS OF DATA, BUT DO WE HAVE INFORMATION TO IMPROVE IAQ ? WHAT CAN BIM AND PREDICTIVE TWINS MEAN? Temperature 02-Feb-2016 25

10 00:00 03:00 06:00 09:00 12:00 15:00 18:00 21:00 00:00

600 room temperatures, which one to choose?

Wouter Borsboom, Ruud van der Linden, TNO, 2021

BUT HOW ARE DIGITAL SOLUTIONS REALLY GOING TO MAKE A DIFFERENCE TO THESE CHALLENGES?

Living lab TNO (NL)

Sustainability Health

Energy Maintenance performance

Physical building Predictive building Digital Twin

6 Predictive twins are predictive digital replicas of physical structures such as bridges, tunnels, homes and offices. With these twins, the future behaviour and use of structures and networks of structures can be predicted and influenced

Wouter Borsboom, Ruud van der Linden, TNO, 2021

6 BIM 2 PREDICTIVE TWIN

How can we use BIM information to structure and analyze data from building systems to check the current performance, detect events, predict, control and optimize?

How to make it operational in the form of workflows in both design commissioning and operation?

How to use it to improve IAQ, event detection, optimize building processes & commissioning?

Development of predictive twin methodologies and tools to meet these challenges for clients.

Wouter Borsboom, Ruud van der LInden TNO, 2021

WHY USE BIM DATA, WE HAVE AI

Heatflow kW

Power kW Measurement data: generally a lot but usually important factors unmeasured Many parameters in a data driven model with low quality data → Flow m3/s overfitting -> poor estimation

Advantages of using BIM in combination with physical models:

What you already know you don’t have to estimate. You can do with less informative data to calibrate the model Logical boundaries on parameters in a physical model

Wouter Borsboom, Ruud van der Linden, TNO, 2021

8 EXAMPLE BIM 2 PREDICTIVE TWIN: BIM 2 BEM CHALLENGES

BIM models not made for simulation models Decisions have to be made for zoning, and also for space 8-zone BIM model 2-zone BEM model boundaries. Inconsistencies especially on the space boundaries can give issues for building model. Different standards: IFC, gbXML, IDF.

Space boundaries issues

Wouter Borsboom, Ruud van der Linden, TNO, 2021

BIM 2 PREDICTIVE TWIN FOR IAQ AND ENERGY USE

Building Information Model (BIM) Building Energy Model (BEM)

BIM-Information

Revit gbXML & IDF

TNO SirenE Predictive Twin

User behaviour model (data en machine learning) Physical model energy use

Dynamic data (measuring data) HayStack

Wouter Borsboom, Ruud van der Linden, TNO, 2021

10 BIM INFORMATION: EXAMPLE OF GBXML 2 MODEL

gbXML data container standard 6.1 & IDF file File import / data selection

SirenE: Generation of a general struct containing all information needed for building simulation

Dwelling: two zone IDF HVAC model gbXML model -

TNO AirMaps ventilation model TNO Heat transfer model

This project has received funding from the European Union’s Horizon 2020 research and innovation Wouter Borsboom, Ruud van der Linden, TNO, 2021 programme under grant agreement No. 869918 11

OCCUPANT MODEL , WORK IN PROGRESS

Adaptive triggers: environmental trigger (e.g. due to discomfort) Non-adaptive triggers: intentional actions (e.g. showering) Contextual factors: influence the magnitude of the triggers (e.g. occupants clothing)

Occupancy: - usually modeled with time profiles

Behavior: - usually modelled with Markov chains, - TNO research on federated learning Schweiker, 2017

Wouter Borsboom, Ruud van der Linden, TNO, 2021

12 EXAMPLE: PREDICTIVE TWIN TO ESTIMATE FLOOR TEMPERATURE

TKI Officecomfort

DYNAMIC DATA FOR PREDICTIVE TWINS: DECISION MAKING NEEDS RELIABLE AND INFORMATIVE DATA

room temperatures: 8ste 25 26

24 15

10 22 5

0 “Data Tsunami” 20 -5 03/06 03/13 03/20 03/27 04/03

LBK 1 16-Feb-2016 : 01-Apr-2016 : air supply 18 200 control-signal-supply-fan /10 100 pressure-difference-airfilter-supply

0 14 21 28 06 13 20 27 03 16 LBK 1: air return 100 control-signal-return-fan /10 50 pressure-difference-airfilter-retour control-signal-heat-wheel 0 14 14 21 28 06 13 20 27 03 03/06 03/13 Temperature03/20 02-Feb-2016 03/27 04/03 25 LBK 1: temperatures 30 supply-temperature-air 20 return-temperature-air temp-air-supply-after-heat-wheel 10 outlet-watertemperature-heating-coil 20 14 21 28 06 13 20 27 03

LBK 1: pressures 400 supply-pressure 200 return-pressure

15 0 14 21 28 06 13 20 27 03

10 00:00 03:00 06:00 09:00 12:00 15:00 18:00 21:00 00:00 time stamp What is monitored? database(s) • Data from Building Automation System: room thermostats, Air Handling Units, DID’s, Coolers etc. Data consistency • Energy meters. • IOT data Data ≠ Information • Additional systems: (entry gates), sun shades, complaints etc. • AIM: services, retrofitting

Wouter Borsboom, Ruud van der Linden, TNO, 2021

14 DATA ISSUES

LBK 1 16-Feb-2016 : 01-Apr-2016 : air supply 200 room temperatures: 8ste control-signal-supply-fan /10 26 100 pressure-difference-airfilter-supply

0 14 21 28 06 13 20 27 03 24

LBK 1: air return 100 22 control-signal-return-fan /10 50 pressure-difference-airfilter-retour control-signal-heat-wheel 20 0 14 21 28 06 13 20 27 03

LBK 1: temperatures 18 30 supply-temperature-air 20 return-temperature-air temp-air-supply-after-heat-wheel 16 10 outlet-watertemperature-heating-coil 14 21 28 06 13 20 27 03

14 LBK 1: pressures 03/06 03/13 03/20 03/27 04/03 400 supply-pressure 200 return-pressure An outlyer or a problem?

0 14 21 28 06 13 20 27 03

Lots of parameters in the AHU, but will it give me insight?

Wouter Borsboom, Ruud van der Linden, TNO, 2021

DEMONSTRATION USE CASE GUI SCALABLE MODELS

Wouter Borsboom, Ruud van der Linden, TNO, 2021

16 17

WHAT IS NEEDED?

Optimal IAQ & Energy efficient office

Optimi- zation Only a process that is controlled can be optimized

Prediction & Control Only a process that is monitored can be controlled

Intelligent Monitoring, event detection Data quality assessment and data repair IOT, Near Real-time Data

Wouter Borsboom, Ruud van der Linden, TNO, 2021

18 DATA INFORMATION FLOW Assessment of current status Prediction of Model-based asset future analysis of data development and root causes Combining with Smart data financial / business analysis for criteria correlations Data quality Recommendations assessment & for action filtering Planning of actions Structuring of data in time from multiple sources Standardized Real-time tagging training of staff of data Actions in workflow Measurement of management current peformance system (sensors or human) Execution of Evaluation of Logging of Naar Henk Akkermans, maintenance / WCM/TiU 2015 effectiveness maintenance / operations actions Wouter Borsboom, Ruud van der Linden, TNO, 2021 actions operations actions 19

MANY STAKEHOLDERS INVOLVED,

INTEGRATION WITH BIM IN OPERATION AND WORKFLOW ORGANIZATION IS CRUCIAL TO EFFECTIVELY IMPROVE IAQ

20 THANK YOU FOR YOUR ATTENTION

21 Online personal IAQ monitoring – personal exposure to indoor air pollutants. B. Hanoune

[email protected]

https://www.apolline.science

• Development of small-size, low-cost, pollutant monitoring system adapted to indoor and outdoor environments • U Lille campus-wide sensor network, can be deployed in other environments

• Research objectives : to understand the drivers of the dynamics of pollution inside buildings (U Lille and other environments), and to quantify the exposure of people

• Educational and awareness tool for students, staff, academics and general public

2 Strategies to monitor personal IAP exposure

Strategy #1 : Personal sensors - Size, weight, autonomy, communications are critical factors - Few sensors inside the device - Access to indoor and outdoor personal exposure

Strategy #2 : Fixed indoor air sensors - Electrical plug and ethernet available - Sensor box can be somewhat large - Access to room/building air concentration, not exposure

From sensors to data

Individual sensors Sensing node Communication Server (storage infrastructure) protocol

Delayed or real-time web-based display

4 « Wearable sensors » strategy

Do we really want to have sensors on everyone ? (would be great also, but…)

24 hours with a sensor

Challenges : - Discrimination indoor/outdoor : jumps in T, RH, number of satellites… - Metrology of sensors (changes in T, RH, speed) - Data transfer - Autonomy - Time and space variability - Inter-individual variability - Are the sensor people-proof ?

6 Apolline in the subway

Indoor Outdoor Metro Outdoor

Indoor concentrations ~ 5 µg/m3 Outdoor concentrations ~ 10 µg/m3

PM10 Subway concentrations up to 80 µg/m3

PM1 Subway concentrations up to 50 µg/m3

« Fixed sensors » strategy

Do we really want to have sensors in every room of every building ? (would be great, but…)

8 What inter-dwellings variability ?

All investigated environments behave similarly.

Main difference is T and RH, driven by behavior, not by pollution sources.

The « statistical » outliers correspond to pollution episodes driven by activities.

Identification of patterns associated to activities

Self-reported activities : cooking (various methods), some cleaning.

Some events detected but not reported.

« Chemical signature » of activities. Need to construct a database for such signatures.

10 Sensors and ventilation

AER Students in their office at the University CO2 as a proxy to trigger ventilation only when necessary.

From sensors to usable information

Sensing node Server (storage infrastructure)

Can we / should we analyze the data before transmission ? - Partial information - Computation power - Less transmissions Data analysis : - Energy efficiency ongoing work, and the key to putting sensors everywhere

12 Take home messages

• Wearable and fixed sensors allow to monitor or evaluate personal exposure.

• Provided you can fix all technical issues : sensor response, size, weight, autonomy, cost, communication protocols, storage infrastructure…

• Not that much low cost, if you consider putting sensors everywhere, framework infrastructure, calibration and operation, environmental cost

• For exposure, recommandation for a network of wearable sensors / fixed sensors / reference monitoring stations

• Issue #1 : make sure what the goal is : exposure characterization, building-related pollution, activity-related pollution, building management, alerting system…

• Issue #2 : what are we looking for ? Environmental parameters, particles, (speciated) gases, exposure duration, pre-identified events, unexpected events…

• Issue #3 : how and when to analyze ? Delayed/real-time. Server/on-board sensor. Time-series, classification, artificial intelligence…

https://annex86.iea-ebc.org/

14 Thanks to :

• the participants in the studies

• the U Lille APOLLINE team that designed and implemented the sensors network infrastructure

• the APOLLINE funding partners : CLIMIBIO, IREPSE, CaPPA, Rincent Air, I-SITE ULNE

• the AIVC Webinar organizers

• the audience

15 Brains for buildings: where to find all the relevant smart building data?

TUESDAY 13 APRIL 2021 – WEBINAR – BIG DATA, IAQ AND VENTILATION – PART 1

Pieter Pauwels, Associate Professor

Department of the Built Environment, Information Systems in the Built Environment

Who am I?

Associate Professor TU Eindhoven (2019)

Assistant Professor Ghent University (2016-2019)

Postdoc Ghent University (2014-2016)

Postdoc University of Amsterdam (2012-2014)

Master & PhD in Civil Engineering – Architecture @Ghent University (2008, 2012)

2 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

2 Presentation Outline

1. Brains 4 Buildings: why? 2. Building Data Semantics: BIM, IFC, LBD, BRICK, etc. 3. System Integration for scalability and feasibility

3 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

- Buildings are to serve people’s needs: - Occupants and FM: Health, comfort, ease of use & ease of operation, affordability - Humankind: energy efficiency, renewables - Building operation is key (Energy & Indoor climate systems) - Lots of occupants & FM dissatisfaction - Lots of energy wastage - Operation data & data analytics, ML, AI are key to: - Understand - Steer & Control optimally - Adapt to renewable energy 4 Brains for buildings: where to find all the relevant- Make smart better building designs data? - Pieter Pauwels

4 End aims and user stories

- Fault detection (improved building management systems – BMSs) - understanding what goes wrong in a system - predicting when faults occur and having clues about the why - live detection - User-centered systems - comfort levels per individual user DATA - learning from user interaction INTEGRATION - Flexible Energy (interfaces): - consumption of energy where it is being created - creation of energy where it is consumed - load balance in the local and regional grid, including the appliances used

5 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

Data Integration for Smart Buildings

To enable making our buildings smarter, advanced data integration is needed (among several other matters):

- Ensure data connectivity between applications

- Ensure security, ethical use and privacy of data

- Standardise data sets and approaches

- Aim for system integration at API level, between individual systems of diverse manufacturers

6 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

6 Presentation Outline

1. Brains 4 Buildings: why? 2. Building Data Semantics: BIM, IFC, LBD, BRICK, etc. 3. System Integration for scalability and feasibility

7 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

All sorts of data available …

8 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

8 BIM Data

9 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

BIM data: Revit, modelling guidelines, agreements, 3D modelling, and IFC

10 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

10 Data in the Industry Foundation Classes (IFC) - Overall building shape and topology easy - Classification of elements possible, but not many classes => extension with bSDD classes and properties possible - Difficult (not impossible) to include sensor data (timeseries data) - Availability in STEP, XML, RDF, and JSON

11 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

W3C Linked Building Data

12 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

12 Emergence of W3C LBD Community Group: Mission Statement Bring together experts in the area of Building Information Modeling (BIM) and Web of Data technologies to: 1. define existing and future use cases and requirements for Linked Data based applications across the life cycle of buildings. 2. discuss best practices for publishing building data on the Web propose ontology models to describe: 1. Buildings and building elements (topology, associate values to properties) 2. Products and product properties 3. discuss how they can be used together with other specifications: 1. existing standards (IFC, GeoSPARQL, Semantic Sensor Network, ...) 2. separate initiatives (schema.org, Haystack, BRICK, ...)

13 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

Scope of BOT: just the start, allowing to extend

- Limited set of classes - Extensible and easy to combine with other ontologies and data sets - Comprehensible

14 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

14 Modular ontology modelling advocated by LBD group

BOT SOSA

OPM Schema GEOM level - DOG SEAS PROPS OM

ifcOWL GEO Schema

bot4osh osh level - geom4osh ifc4osh

link4osh Instance • Implemented using Semantic Web Technologies -> Web-scale, queryable • Reuse of existing ontologies -> Modular • Linking at instance level -> Multi-model method Sample dataset available at: https://github.com/TechnicalBuildingSystems/OpenSmartHomeData

15 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

Modular approach to building data

16 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

16 Reference ontologies

BOT https://w3id.org/bot# BEO https://pi.pauwel.be/voc/buildingelement/ MEP https://pi.pauwel.be/voc/distributionelement/ OMG https://w3id.org/omg# FOG https://w3id.org/fog# BPO https://www.w3id.org/bpo# OPM https://www.w3id.org/opm#

Revit to LBD exporter: on demand IFC to LBD converter: on demand

17 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

BRICK, HTO, SAREF, and REC

18 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

18 BRICK, HTO, SAREF, REC, etc.

- Developments are rather disconnected from any BIM- or building-related area - Focus on systems, incl. operation and control - Focus on the sensor Point and Equipment types https://brickschema.org/ - Beware of biased overview tables

19 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

BRICK - A uniform metadata schema for buildings

https://brickschema.org/

20 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

20 BRICK, HTO, SAREF, REC, etc.

BRICK SAREF4BLDG

Real Estate Core

21 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

In short Nothing that cannot be included in a modular linked building data (LBD) cloud.

22 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

22 Presentation Outline

1. Brains 4 Buildings: why? 2. Building Data Semantics: BIM, IFC, LBD, BRICK, etc. 3. System Integration for scalability and feasibility

23 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

Mapping all into RDF

24 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

24 Available options for integration for software

OPTION 1: Transform all into semantic graphs (e.g. R2RML or custom data transformers) and do data integration • Plus: all in same format • Plus: inference possible • Minus: unfit storage • Minus: disconnect from origin • Minus: no ML algorithms nor procedural code possible • Minus: how to handle privacy and security (trust?)

25 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

Native data procedure ETL format Data management in Native data RDF application format stores back-end End user Native data application format development

Linking data RDF stores

26 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

26 Gonçal Costa, Alvaro Sicilia, Leandro Madrazo. Energy efficiency of buildings. 1st Intl. Workshop on Linked Data in Architecture and Construction. Ghent, BE, 2012. James O’Donnell, Edward Corry, Edward Curry, Marcus Keane. Building and using multi-domain data. 1st Intl. Workshop on Linked Data in Architecture and Construction. Ghent, BE, 2012.

27 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

Native data procedure ETL format Data management in Native data RDF application format stores back-end End user Native data application format development

Linking data

- Disconnect from source RDF - Specialised software and data handling lost stores Security and data access protocols?

28 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

28 Maintaining specialized data stores and deploying web technologies

29 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

Available options for integration in trustworthy manner OPTION 2: Store all in well-fit data stores (KV stores, graphDBs, relational DBs, timeseries stores, etc.) and perform data integration (also) on a system and API level (system integration) • Plus: apt data storage • Plus: data stays at source -> web-based connections needed • Plus: ML algorithms and procedural algorithms not blocked • Plus: Privacy and security can be easily handled at the gates of APIs and DBs. • Minus: multitude of systems requires lots of diverse software and expertise

30 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

30 Linking data in server Native data back-end across Data SQL format different types of management in databases application back-end Native data RDF format End user application back-end Native data RDF format

Data Access Layers (REST APIs)

End user application Standard web back-end Native data technologies PCD format (ACL)

31 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

Petrova, E., 32Pauwels,Brains P., Svidtfor buildings:, K., Jensen, where R.L. to (2018) find all From the relevantpatterns smartto evidence building: Enhancing data? - Pietersustainable Pauwelsbuilding design with pattern recognition and information retrieval approaches. Proceedings of the 12th ECPPM conference, pp. 391-399. 32 Petrova, E., Pauwels, P., Svidt, K., Jensen, R.L. (2018) From patterns to evidence: Enhancing sustainable building design with pattern recognition and information retrieval approaches. Proceedings of the 12th ECPPM conference, pp. 391-399.

33 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

Abox linking to point cloud data and geometry

Jeroen Werbrouck et al., Scan-to-graph: Semantic enrichment of existing building geometry, Automation in Construction, 119 (2020). https://doi.org/10.1016/j.autcon.2020.103286.

34 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

34 Increase simplicity, preserve semantics Simplicity Digital Twin System (DTLab) Topological map of floor 8 SCADA system

InfluxDB / … API E Physical building v Semantic 3D model (BIM) e Knowledge graph n Metric map of floor 8 t

GraphDB / b Neo4J / … r API o k e Full simulated world r Research Challenges: Security - - Trustworthy Data Integration Non-semantic 2D MongoDB Authentication - - Merging Models and Data & 3D geometry API

35 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

36 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels

36 Presentation Outline

1. Brains 4 Buildings: why? 2. Building Data Semantics: BIM, IFC, LBD, BRICK, etc. 3. System Integration for scalability and feasibility

37 Brains for buildings: where to find all the relevant smart building data? - Pieter Pauwels