Melbourne Olympic Park Stadium
Though Melbourne already has a world class stadium to population ratio (WCSPR) of about one to one hundred, the sleeping giant soccer (aka football) recently woke up and noticed through bleary eyes that this sports-mad city did not yet have a world class soccer facility in its proverbial trophy cabinet.
Now under construction, this new stadium will allow soccer and rugby fans to get close to the action like never before, rather than conforming to the geometric restrictions of Melbourne's much-beloved oval pitch stadiums, where 'Aussie Rules' football and cricket reign supreme.
COX Architects, working alongside Arup as structural engineers designed a 'bioframe' roof, comprising a single structural layer of steel tubes that use a combination of arch, cantilever and shell actions for stability. With the structural mass of the roof weighing in at only 45 kg/m2, it makes many a lesser stadium look positively flabby in comparison.
How GSA Was Used
GSA, with its user-friendly interface and wide range of analysis functions, was an integral part of realising this lightweight roof. In the initial stages of the project the wireframe of the complex geometry was imported directly into GSA using .DXF file format, ensuring consistency between architectural and analysis models.
Once the geometry was established, GSA's grid loading capabilities allowed roof member loads to be input (and amended) more quickly than if line loads were individually applied to each element.
To inform calibration of wind tunnel tests, modal analysis was used to estimate the natural frequency of the roof structure, the results of which were fed back into the analysis model for use in checking strength and serviceability criteria.
To assess whether each of the CHS members (over 4000 in total) had adequate strength, and in order to account for the buckling behaviour of the bioframe, the 'Dallard Method' was adopted. The 'Dallard Method' is ideally suited to analysis using GSA, as it makes use of GSA's P-Delta and buckling solvers. As part of this process it was necessary to have an analysis package that is capable of re-calculating the stiffness of the structure in its displaced conditions, and GSA was again able to deliver.
Conveniently, the buckled mode shapes and associated displacements are already normalised, ready for direct application to Dallard's method of buckling assessment. Also beneficial was GSA's graphic output, such as deflected shape and displacement contours, as it allowed clear identification of the effective length of each mode shape. Furthermore, the ease with which load cases and design actions are able to be enveloped within GSA was a factor in reducing the time required for analysis.
When checking the final load combinations, the ability to plot combined stress contours meant structural adequacy could be quickly ascertained.
Throughout all stages of analysis, GSA's ability to easily import and output data in spreadsheet format was crucial, and its user-friendly and excellent graphic representations allowed the model to be quickly manipulated and understood.