The behaviour and design of prestressed steel structures, with an emphasis on trussed arches, are examined in this thesis. For long-span structural systems, where self-weight becomes an increasingly dominant component of the design loading, significant material savings can be achieved through the use of high tensile strength steel cables in conjunction with conventional steelwork. Further benefits can be achieved by prestressing the cables.

In the system currently being investigated, the prestressed
cables, which are housed within the bottom chord of tubular arched trusses,
apply a compressive force to the chord members, which is opposite in
nature to the resultant forces arising from the externally applied
gravity loads. The stability of the trussed elements under prestress and the load--deformation response of the prestressed elements to the subsequent application of tensile loading are examined analytically, numerically and experimentally, with good correlation achieved between the three approaches. The benefits of prestressing, in terms of increased member strength and stiffness, are demonstrated, and optimal prestress levels are investigated.

In instances of load reversal (e.g. due to wind uplift) in trusses without horizontal end anchorage that would allow catenary forces to develop, the presence of prestress can become detrimental. To examine this, a total of eight pin-ended cable-in-tube systems, featuring both non-grouted and grouted members, were tested in compression. Increasing initial prestress levels was found to reduce the capacity of the system in compression, but initial prestress was shown to be less detrimental than externally applied compressive loading of the same magnitude, due to the absence of second order bending moments. Finite element models were developed and, following accurate replication of test results, were used to generate parametric results for a range of member slendernesses and prestress levels. The test and FE results were compared against capacity predictions based on a proposed modified Perry-Robertson design method. Consistent, accurate and generally safe-side predictions were achieved.

Following the examination of behaviour of individual prestressed elements within the truss, a series of analytical and numerical models of the full arched truss system were developed to investigate its global structural behaviour. Parametric studies revealed that the horizontal end boundary conditions, prestress level, truss depth and diagonal member arrangements were the key parameters influencing the stiffness, load bearing capacity and failure mode of the structure.