South African building envelope thermal performance simulation: parameters and the role of moisture content

Meyer, Edmund Stanhope (2022-04)

Thesis (PhD)--Stellenbosch University, 2022.

Thesis

ENGLISH ABSTRACT: Literature tells of a connected, yet not well harmonised, narrative between building sustainability, building performance, building simulation and building regulations. Building simulation is actively used in industry to achieve targeted building ratings, but the targets do not always match reality due to the associated complexity and uncertainty associated with building simulation. This is further complicated by how the target is set for building performance, as many require an improvement over a benchmark value. For South Africa, the benchmark is provided by building standards. The need to investigate hygrothermal analysis for South African buildings is based on three observations. First, there is a lack of hygrothermal studies in South Africa. Second, conclusions drawn in the literature that focus on the thermal performance of South African low-income housing are based on 'what' environmental conditions are experienced and not 'why' specific environmental conditions are experienced. Third, although climate may be similar when comparing studies that focus on building performance, differences in results exist due to differences in the assumptions used for building simulation. The purpose of this study is to establish the case for improved building simulation and the regulation thereof in South Africa. To do so, four building models were adapted or created and modelled under specific conditions. The first two building models provide insight into how design choices influence the heating and cooling loads of two South African Green Star buildings when only considering heat-only analysis. Results indicate that providing high performance thermal insulation in combination with a large window-to- wall ratio will put additional strain on the air conditioning systems of the Green Star buildings. Furthermore, increasing the window-to-wall ratio of all glazed surfaces during analysis provides limited insight into the effect of the window-to-wall ratio of the Green Star buildings, unless considered alongside the changes to heat transfer through opaque building components, or considering specific surfaces. The third building model highlights the influence of moisture buffering materials on the environmental conditions of the building environment for a typical summer and winter week in South Africa. Results indicate that hygric materials directly influence the building environment. Furthermore, the sorption isotherm and initial moisture content of the material appear to influence simulation results significantly. Results indicate that if sufficient hygroscopic building materials are exposed (in terms of surface area and volume) to the South African environmental conditions, expected simulated results will be different, depending on the inclusion of moisture in heat transfer through building materials. The final building model serves as validation for the conclusions made regarding the importance of hygrothermal analysis. The model also highlights the need for improved modelling guidelines for South African buildings and additional requirements regarding buildings' material properties. The accuracy of the hygrothermal analysis, compared to heat-only analysis, confirms that hygrothermal analysis is required to accurately simulate the environmental conditions in South African buildings with moisture buffering components. Although results show the influence of moisture buffering materials on simulated temperature and relative humidity, further research is still required in South Africa to allow for independent hygrothermal analysis, i.e. hygrothermal analysis using South Africa specific data. Future research towards improved building simulation in South Africa should focus on expanding simulation input data provided for South African building performance simulation, as well as when hygrothermal modelling is needed.

AFRIKAANSE OPSOMMING: Literatuur vertel 'n samehangende, maar nie goed geharmoniseerde, verhaal oor volhoubare konstruksie van geboue, die gedrag en simulering van geboue, en bou-regulasies. Simulering van geboue word aktief in die industrie gebruik om geteikende gebou-graderings te behaal, maar die teikens pas nie altyd by die werklikheid nie as gevolg van die kompleksiteit en onsekerheid wat verband hou met die bousimulering. Dit word verder kompliseer deur die manier waarop die mikpunt vir gedrag van geboue gestel word, aangesien baie 'n verbetering verg ten opsigte van 'n bepaalde maatstaf. Vir Suid-Afrika word die maatstaf voorgeskryf in bou-standaarde. Die behoefte om higrotermiese analise vir Suid-Afrikaanse geboue te ondersoek, is baseer op drie waarnemings. Eerstens, is daar 'n gebrek aan higrotermiese studies in Suid-Afrika. Tweedens, word gevolgtrekkings gemaak in die literatuur wat fokus op die termiese prestasie van Suid-Afrikaanse lae- inkomste-behuising, baseer op slegs die omgewingstoestande en nie waarom spesifieke omgewingstoestande ondervind word nie. Derdens; hoewel die klimaat soortgelyk kan wees by die vergelyking van studies wat fokus op gedrag van geboue, is daar verskille in resultate as gevolg van verskille in die aannames wat gemaak word vir die bou-simulering. Die doel van hierdie studie is om ‘n saak te maak vir beter gebou-simulering en die regulering daarvan in Suid-Afrika. Om dit te doen, word vier bou-modelle onder spesifieke omstandighede aangepas of geskep en gemodelleer. Die eerste twee boumodelle bied insig in hoe ontwerpkeuses die verhitting- en verkoelingsladings van twee Suid-Afrikaanse Groen-Ster geboue beïnvloed wanneer slegs ontleding met warmte oorweeg word. Resultate dui aan dat die gebruik van hoë werkverrigting termiese isolasie in kombinasie met ‘n groot venster-tot-muur verhouding bykomende druk op die lugversorgingstelsels van die Groen-Ster geboue sal plaas. Verder bied die verhoging van die venster-tot-muur verhouding van alle geglasuurde oppervlaktes tydens ontleding beperkte insig in die effek van die venster-tot-muur verhouding van die Groen-Ster gebou, tensy dit saam met die veranderinge aan hitte-oordrag deur ondeursigtige geboukomponente oorweeg word, of spesifieke oppervlaktes oorweeg word. Die derde boumodel bied insig in hoe vogbuffermateriale die omgewingstoestande van die gebou- omgewing beïnvloed vir ‘n tipiese somer- en winterweek in Suid Afrika. Resultate dui aan dat higroskopiese materiale die gebou-omgewing direk beïnvloed. Verder blyk dat die sorspsie isoterm en die aanvanklike voginhoud van die materiaal simulering-resultate noemenswaardig beïnvloed. Resultate dui aan dat indien voldoende vogbuffermateriale aan die omgewingstoestande van Suid Afrika blootgestel word (in terme van oppervlakte en volume), verwagte gesimuleerde resultate anders sal wees, afhangende van die insluiting van vog in hitte-oordrag deur boumateriale. Die finale boumodel dien as bevestiging van die gevolgtrekkings rakende die belangrikheid van higrotermiese analise. Die model beklemtoon ook die behoefte aan verbeterde modelleringsriglyne vir Suid-Afrikaanse geboue en bykomende vereistes ten opsigte van geboue se materiaaleienskappe. Die akkuraatheid van die higrotermiese analise, in vergelyking met hitte-alleen-analise, bevestig dat higrotermiese analise nodig is om die omgewingstoestande in geboue in Suid Afrika met vogbuffer- komponente akkuraat te simuleer. Alhoewel resultate die invloed van vogbuffermateriale op gesimuleerde temperatuur en relatiewe humiditeit toon, is verdere navorsing nodig in Suid-Afrika om onafhanklike higrotermiese analise moontlik te maak, dit wil sê higrotermiese analise met behulp van data spesifiek tot Suid-Afrika, insluitend data vir boumateriale, gebou ontwerp, weer, grond, besetting, skedules wat verband hou met boustelsels, asook hitte-wins data. Toekomstige navorsing vir verbeterde simulering van Suid- Afrikaanse geboue moet fokus op die uitbreiding van insetdata vir simulering, asook wanneer higrotermiese analise nodig is.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/124504
This item appears in the following collections: