Thesis (MScEng)--Stellenbosch University, 2003. / Some digitised pages may appear illegible due to the condition of the original hard copy. / ENGLISH ABSTRACT: A critical evaluation of provisions for imposed loads in the South African Loading
Code for design of structures, SABS 0160-1989 (SABS), by comparison with other
codes was performed earlier. The evaluation revealed the SABS loading code to be
generally non-conservative in its provisions for imposed loads for a range of general
and specialist occupancy classes. The SABS provision for imposed loads for
inaccessible roofs was found to be substantially non-conservative in comparison with
the other codes. An investigation into the imposed load for inaccessible roofs is
subsequently performed in order to establish a scientific rationale through which the
codified design values may be measured effectively. Due to the lack of information
and the large uncertainties involved in the imposed roof load, stochastic treatment of
the loads is implemented. This is in line with the stochastic modelling of loads as
implemented in general.
The approach applied is to select a type of building that can be regarded as a generic
example of buildings to which these loads apply, and to discretisize the load into the
various sub-mechanisms that translate into the imposed roof load. The probabilistic
models for the load mechanisms are then quantified, either through physical load
surveys, or through conducting an expert survey for those variables which are not
observable. The use of expert opinion as a resource for information is not readily
accessible in terms of yielding scientifically defendable results. Therefore, the expert
survey is performed as a calibrated experiment whereby weights were calculated for
the individual experts' opinions and their opinions combined accordingly.
The probabilistic models for the load mechanisms are then translated into load effects
by taking into account the physical process resulting in the load effects. By applying
these mechanisms in such a way as to maximise the said load effects, equivalent
uniformly distributed loads (EUDL's) were calculated for each mechanism. The
probabilistic models obtained in terms of the EUDL's pose an easily accessible
format through which existing load models and codified provisions can be evaluated.
These load models are then utilised to evaluate the SABS provisions in terms of the
level of reliability catered for by SABS ultimate limit-state design criteria. It is
concluded that the SABS conservatively provides for maintenance loads on the roof,
while the reliability for construction loads is non-conservative for large tributary areas
and highly non-conservative for small areas. The load models so obtained can
further be applied for structural reliability assessment. / AFRIKAANSE OPSOMMING: "n Kritiese evaluasie van die voorskrifte vir opgelegde belastings in die Suid-
Afrikaanse Selastingskode vir die ontwerp van strukture, SASS 0160-1989 (SASS)
deur 'n vergelyking met ander kodes is vroeër uitgevoer. Die evaluasie het getoon
dat die SASS in die algemeen onkonserwatief is in sy voorsiening vir opgelegde
belastings oor 'n bereik van algemene en spesialis okkupasie tipes. Die SASS
voorskrif vir opgelegde belastings vir ontoeganklike dakke is hoogs onkonserwatief in
vergelyking met die ander kodes. 'n Ondersoek na die opgelegde belasting vir
ontoeganklike dakke word gevolglik uitgevoer met die doelom 'n wetenskaplike
rasionaal daar te stel waardeur die gekodifiseerde voorskrifte effektief gemeet kan
word. As gevolg van die gebrek aan inligting en groot onsekerhede betrokke by die
opgelegde dakbelasting word stogastiese modellering geimplimenteer.
Die aanslag wat gevolg is, is om 'n tipe gebou te selekteer wat beskou kan word as
verteenwoordigend van die geboue waarvoor hierdie belastings van toepassing is, en
om die belasting te diskretiseer in die verskeie lasmeganismes wat die opgelegde
dakbelasting voortbring. Die waarskynlikheidsmodelle vir die lasmeganismes word
dan gekwantifiseer, óf deur fisiese opnames, óf deur die uitvoering van 'n ekspertopname
vir daardie veranderlikes wat nie waarneembaar is nie. Die gebruik van
ekspert opinie as "n bron van inligting is nie maklik toeganklik in terme daarvan om
wetenskaplik verdedigbare resultate te lewer nie. Daarom is die ekspert-opname
uitgevoer soos 'n gekalibreerde eksperiment waardeur relatiewe gewigte bereken
word vir die individuele eksperts en hulopinies daarvolgens gekombineer word.
Die waarskynlikheidsmodelle vir die lasmeganismes word dan omgeskakel in laseffekte
deur in agneming van die fisiese proses wat die las-effek voortbring. Deur die
lasmeganismes op só 'n manier toe te pas dat die betrokke las-effekte gemaksimeer
word, word ekwivalent uniforme belastings (EUS's) bepaal. Die waarskynlikheidsmodelle
in terme van EUS's bied "n maklik toeganklike formaat waardeur bestaande
lasmodelle en gekodifiseerde voorskrifte evalueer kan word. Die lasmodelle word
gevolglik gebruik om die SASS voorskrifte te evalueer in terme van die vlak van
betroubaarheid wat gehandhaaf word deur SASS limiet-staat ontwerp kriteria. Dit is
bepaal dat die SASS konserwatief voorsiening maak vir onderhoudslaste op die dak,
maar onkonserwatief tot hoogs-onkonserwatief is vir konstruksie laste. Die bepaalde
lasmodelle kan verder toegepas word in strukturele betroubaarheids analise.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:sun/oai:scholar.sun.ac.za:10019.1/53460 |
| Date | 04 1900 |
| Creators | De Villiers, Pieter Jacobus |
| Contributors | Retief, J. V., Stellenbosch University. Faculty of Engineering. Dept. of Civil Engineering. |
| Publisher | Stellenbosch : Stellenbosch University |
| Source Sets | South African National ETD Portal |
| Language | en_ZA |
| Detected Language | Unknown |
| Type | Thesis |
| Format | 257 p. : ill. |
| Rights | Stellenbosch University |
Page generated in 0.0025 seconds