This paper deals with critical thermal buckling load optimization of symmetrically laminated four layered angle-ply plates with one or two different intermediate line supports. The design objective is the maximization of the critical thermal buckling load and a design variable is the fibre orientation in the layers. The first order shear deformation theory and nine-node isoparametric finite element model are used for the finite element solution of the laminates. The modified feasible direction (MFD) method is used for the optimization routine. For this purpose, a program based on FORTRAN is used. Finally, the numerical analysis is carried out to investigate the effects of location of the internal line supports, plate aspect ratios and boundary conditions on the optimal designs and the results are compared.

Steel structural elements with web-tapered I cross section, are usually made of welded thin plates. Due to the nonrectangular shape of the element, thin web section may be obtained at the maximum cross section height. The buckling strength is directly influenced by lateral restraining, end support and initial imperfections. If no lateral restraints, or when they are not effective enough, the global behaviour of the members is characterized by the lateral torsional mode and interaction with sectional buckling modes may occur. Actual design codes do not provide a practical design approach for this kind of elements. The paper summarizes an experimental study performed by the authors on a relevant number of elements of this type. The purpose of the work was to evaluate the actual behaviour of the web tapered beam-columns when applying different types of lateral restraints and different web thickness.

In this paper, experimental investigations on high strength steel (HSS) stud connected steel-concrete composite (SCC) girders to understand the effect of shear connector density on their flexural behaviour is presented. SCC girder specimens were designed for three different shear capacities (100%, 85%, and 70%), by varying the number of stud connectors in the shear span. Three SCC girder specimens were tested under monotonic/quasi-static loading, while three similar girder specimens were subjected to non-reversal cyclic loading under simply supported end conditions. Details of casting the specimens, experimental set-up, and method of testing, instrumentation for the measurement of deflection, interface-slip and strain are discussed. It is found that SCC girder specimen designed for full shear capacity exhibits interface slip for loads beyond 25% of the ultimate load capacity. Specimens with lesser degree of shear connection show lower values of load at initiation of slip. Very good ductility is exhibited by all the HSS stud connected SCC girder specimens. It is observed that the ultimate moment of resistance as well as ductility gets reduced for HSS stud connected SCC girder with reduction in stud shear connector density. Efficiency factor indicating the effectiveness of high strength stud connectors in resisting interface forces is estimated to be 0.8 from the analysis. Failure mode is primarily flexure with fracturing of stud connectors and characterised by flexural cracking and crushing of concrete at top in the pure bending region. Local buckling in the top flange of steel beam was also observed at the loads near to failure, which is influenced by spacing of studs and top flange thickness of rolled steel section. One of the recommendations is that the ultimate load capacity can be limited to 1.5 times the plastic moment capacity of the section such that the post peak load reduction is kept within limits. Load-deflection behaviour for monotonic tests compared well with the envelope of load-deflection curves for cyclic tests. It is concluded from the experimental investigations that use of HSS studs will reduce their numbers for given loading, which is advantageous in case of long spans. Buckling of top flange of rolled section is observed at failure stage. Provision of lips in the top flange is suggested to avoid this buckling. This is possible in case of longer spans, where normally built-up sections are used.

Concrete filled steel tubular (CFST) member has been widely used in the construction of high-rise buildings for its high axial bearing capacity. It can also be applied on long-span structures such as spatial structures or bridges not only for its high bearing capacity but also for its construction convenience. Concrete casting effect of CFST member is considered in the study of its bearing capacity in this paper. Firstly, in order to authenticate the applicability of constitutive relationship and yield criterion of steel and concrete based on FEM, two ANSYS models are built to simulate and compared with other's test. Secondly, in order to find the huge difference in bearing capacity due to different construction processes, two full-size CFST models are studied when they are horizontally cast and axially compressed. Finally, the effects of slenderness ratio (L/D) and confining parameter (D/t) of CFST members are studied to reveal the intrinsic links between bearing capacity and slenderness ratio or confining parameter.

The interaction between steel tube and concrete core is the key design considerations for concrete-filled steel tube columns. In a concrete-filled steel tube (CFST) column, the steel tube provides confinement to the concrete core which permits the composite action among the steel tube and the concrete. Due to construction faults and plastic shrinkage of concrete, the debonding separation at the steel-concrete interface weakens the confinement effect, and hence affects the behaviour and bearing capacity of the composite member. This study investigates the axial loading behavior of the concrete filled circular steel tube columns with debonding separation. A three-dimensional nonlinear finite element model of CFST composite columns with introduced debonding gap was developed. The results from the finite element analysis captured successfully the experimental behaviours. The calibrated finite element models were then utilized to assess the influence of concrete strength, steel yield stress and the steel-concrete ratio on the debonding behaviour. The findings indicate a likely significant drop in the load carrying capacity with the increase of the size of the debonding gap. A design formula is proposed to reduce the load carrying capacity with the presence of debonding separation.

This paper experimentally investigated the buckling capacity of built-up steel columns mainly, Cruciform Columns (CC) and Side-to-Side (SS) columns fabricated from two Universal Beam (UB) sections. A series of nine experimental tests comprised of three UB sections, three CC sections and three SS sections with different lengths were tested to failure to measure the ultimate axial capacity of each column section. The lengths used for each category of columns were 1.8, 2.0, and 2.2 m with slenderness ratios ranging from 39-105. The measured buckling loads of the tested specimens were compared with the predicted ultimate axial capacity using Eurocode 3, AISC LRFD, and BS 5959-1. It was observed that the failure modes of the specimens included flexural buckling, local buckling and flexural-torsional buckling. The results showed that the ultimate axial capacity of the tested cruciform and side-by-side columns were higher than the code predicted design values by up to 20%, with AISC LRFD design values being the least conservative and the Eurocode 3 design values being the most conservative. This study has concluded that cruciform column and side-to-side welded flange columns using universal beam sections are efficient built-up sections that have larger ultimate axial load capacity, larger stiffness with saving in the weight of steel used compared to its equivalent universal beam counterpart.