The distribution of wind loads and vulnerability of metal clad roofing structures in contemporary Australian houses

Jayasinghe, Nandana Chana (2012) The distribution of wind loads and vulnerability of metal clad roofing structures in contemporary Australian houses. PhD thesis, James Cook University.

[img]
Preview
PDF (Thesis)
Download (5MB) | Preview
View at Publisher Website: https://doi.org/10.25903/qdx8-mm38
 
850


Abstract

Windstorms cause most of the damage to houses, worldwide. The roof is subjected to the largest wind loads and is usually the most vulnerable part of the house. However, data on the transfer of wind loads within the roof structure is scarce. Such data is required for the application of structural reliability analysis and for development of building codes. The fluctuating nature and variable distribution of wind loads, combined with the change from linear to non-linear structural behaviour as the loads increase, can pose challenges for calculating structural response. This is also required for developing the performance based design of structures and for understanding wind load transfer within the roof structure and the effect of progressive failure on the sharing and redistribution of loads. Most vulnerability models do not accurately incorporate the structural behaviour of the houses during windstorms, which may produce unreliable estimates of damage. These vulnerability models have mainly focused on the wind loads acting on the connection tributary area and the strength of the connections.

This research study analysed the transmission of wind loads within a commonly used roof structural system of contemporary houses obtained from a survey in the cyclonic region of Australia. The distribution of wind pressure on the roof of such a typical house was determined using a wind tunnel model. The wind loads on selected roof fixings were further analysed to obtain the wind load data in terms of probabilistic parameters. The strength capacities of the roofing connection were also determined in terms of probabilistic parameters using available test data. The structural response of a roof was studied by testing sub-assemblies of the roof applying point and line loads and measuring the reactions at batten-to-truss connections and the deflections at selected locations in a range of conditions and damage states to simulated loads. The results were also compared with analytical solutions. The variation of reactions with increasing load was discussed in terms of a reaction coefficient. The coefficients were assessed for loads in the linear and non-linear states of roofing components/connections. The study integrated the wind loading and structural information on transfer of wind load effects through the structure to determine the loads and vulnerability of batten-to-truss connections. These outputs were compared with the results obtained from conventional methods for calculating load on connections and vulnerability assessments.

The study found that loads on the batten-to-truss connection of these contemporary houses are influenced by the flexibility of the battens and cladding, and the directional stiffness characteristics of the cladding, as is the redistribution of batten-to-truss connection loads following failure of the cladding fastener and batten-to-truss connections. As a result estimates based on application of pressures to connection tributary area, which is the normal design practice, can lead to underestimation of the connection loads. The study shows that a larger tributary area should be considered to obtain the batten-to-truss connection loads on these structural systems. Furthermore, the study shows that estimates of the vulnerability of a batten-to-truss connection based on the incorporation of load distribution effects in the reliability analysis are greater than those obtained from the methods used in current practice. Hence, the study suggests that load sharing effects must be incorporated when determining the vulnerability of connections. The vulnerability estimates on cladding fixings and truss-to-wall connections were also determined and the results show that the cladding fixings are the most vulnerable and then batten-to-truss connections and truss-to-wall connections respectively.

A main outcome of the thesis is the establishment of an improved procedure for analysing the variation of the connection loads with time taking account of the spatial and temporal variation in wind pressures and the structural response characteristics of the roof system, which is a necessary first step in the assessment of their vulnerability. These outcomes make a significant contribution to understanding the wind loading distribution and developing vulnerability functions for houses to windstorms. The results can also be used to assess the system reliability for a well defined limit state and hence can contribute significantly to performance based evaluation of masonry block houses in cyclonic regions. The results could also be used as a basis to study adaptation measures and for the development of software models for assessing building vulnerability to windstorms.

Item ID: 39226
Item Type: Thesis (PhD)
Keywords: cyclones; houses; metal roofing; roofing; roofs; strength; structural frames; vulnerability; wind tunnel models; wind; wind loads; wind-pressure
Date Deposited: 18 Jun 2015 03:02
FoR Codes: 09 ENGINEERING > 0905 Civil Engineering > 090506 Structural Engineering @ 50%
09 ENGINEERING > 0905 Civil Engineering > 090503 Construction Materials @ 50%
SEO Codes: 87 CONSTRUCTION > 8702 Construction Design > 870201 Civil Construction Design @ 50%
87 CONSTRUCTION > 8701 Construction Planning > 870104 Residential Construction Planning @ 50%
Downloads: Total: 850
Last 12 Months: 23
More Statistics

Actions (Repository Staff Only)

Item Control Page Item Control Page