Civil engineering is by its very nature a multi-disciplinary field and this is reflected by the varied applications of computational methods published in the Proceedings of the Institution of Civil Engineers – Engineering and Computational Mechanics (EACM). This breadth is exemplified by the three papers published in the current issue of EACM. All three papers illustrate how computational mechanics can be applied to solve long-standing problems of practical engineering interest. Fluid–structure interaction (FSI) has featured in the proceedings of the ICE since the first half of the nineteenth century; Kara et al. (2015) present a new approach to modelling this complex multi-physics problem. Tong et al. (2015) offer a more practical view on a related problem in bluff body aerodynamics by using computational simulations to assess the reduction in drag that can be achieved by chamfering the corners of square section columns. In the third paper, Sokołowski and Kamiński (2015) take us away from fluid mechanics to consider the modelling of uncertainty in engineering design, an issue which has again featured since the earliest years in the proceedings of the ICE.
Problems associated with FSI occur in a wide range of civil engineering applications, from aero-elastic instabilities of flexible bridges to the vortex-induced vibrations (VIVs) of immersed pipelines and tethers. Kara et al. (2015) focus on the latter, providing a thorough review of both numerical and experimental developments in the understanding of VIVs for circular cylinders. Their contribution to the computational modelling of this phenomenon is a new model coupling a large eddy simulation of the fluid flow to a single degree of freedom model of the dynamic response. Many FSI solution schemes adopt an arbitrary Lagrangian Eulerian approach, in which the mesh surrounding the moving body deforms and translates as the body moves. The authors have avoided the added complexity that this creates by adopting the immersed boundary method, in which both the fluid flow and body motion are computed in the same fixed frame of reference. Results are presented for both static and dynamic cylinders to show that the method predicts aerodynamic coefficients well and captures the lock-in phenomenon.
Flows around circular sections have their own unique features that create specific problems in practice. More general sharp-edged cross-sections generate a different set of flow-related problems and wind effects on these structures are well known to be extremely sensitive to small perturbations of shape. This applies at the member scale and at the building scale where the shape on a plan of a tall building strongly affects its response in strong winds. Tong et al. (2015) examine the effect of corner details, specifically chamfers, on the aerodynamic properties of square section columns. They adopt a k-ω RANS approach to solving the fluid flow around the bluff body, varying both the size of the chamfer and the angle of attack. Their choice of solution scheme allows realistic Reynolds numbers to be considered. Results confirmed the marked influence of corner detail on the drag coefficient, with even a 5% chamfer leading to a significant reduction. The results showed only minor variation with Reynolds number, but angle of attack had a more significant affect. Tabulated values will form a useful dataset for comparison with future simulations and wind tunnel measurements.
In the final paper of this issue, Sokołowski and Kamiński (2015) consider a novel form of beam construction, the corrugated I-beam, and present a reliability analysis to model the uncertainty in the geometry of the corrugated web. Structural engineering design codes have accounted for uncertainties in different ways over the years as new design philosophies have developed. Each reduced this complex problem to a simple factor of safety or set of partial factors. Recent advances in computational engineering have allowed a more explicit treatment of uncertainty to be made. Here, the authors outline the basic background to reliability theory and show how it can be applied to I-beams with corrugated webs, a structural form not covered explicitly in design codes. The authors compare first- and second-order reliability methods and clearly show the importance of considering the second-order method. Numerical calculations of structural reliability require a random data set to be generated. The use of Monte Carlo simulation (MCS) is intuitive but inefficient because much computational time is spent analysing parameter combinations that are far from the failure boundary. The authors compare MCS with two alternatives and recommend the stochastic perturbation technique as the most efficient.
This collection of papers is an interesting and useful contribution to the literature. These are topics of real interest to the practising engineer and which present real challenges to those working in research and development. There are many who could add to the ongoing discussion on these topics and I would encourage all to contribute with papers in future issues.
