How can I include spatial variation of wind on a long-span bridge during time-domain simulation? Could anyone please suggest a simple method to take into account wind spatial variation and/or provide reference?
The problem is that in the last twenty years have given much emphasis on the numerical simulations, leaving aside the physical measurements of the properties of the wind.
.
The development of numerical simulation can read the most recent review entitled: “50 years of Computational Wind Engineering: Past, present and future”. Blockena, Bert http://dx.doi.org/10.1016/j.jweia.2014.03.008. To wind characteristics suggest look in the references of the article of CFD simulation of flow and dispersion around an isolated building: “Effect of inhomogeneous ABL and near-wall treatment” Z.T. Ai & C.M. Mak http://dx.doi.org/10.1016/j.atmosenv.2013.05.034.
.
But in the two references you will see that the results for wind data are poor.
Maybe if you do a little search in earlier bibliographies to 1990 consigas find something, an example is: “Perspectives on the full-scale measurement of Wind effects” Davenport, A. G. 1975 Journal of Industrial Aerodynamics. If you look at articles in this reference is that thou mayest find something.
.
The problem is that most of the articles nowadays are scholarly articles on CFD and most authors would certainly not have the courage to use their results to simulate real cases and put a personal responsibility against possible errors of boundary conditions which employ!
If you wanted to do a dynamic simulation of your bridge being loaded with varying wind loading due to varying speeds, I would suggest you look at the open access code TurbSim, available from the NREL website. This was written for wind turbine study but could similarly be used for any structure where a time-marching wind field is required.
Otherwise you could check out the latest Eurocode for wind loading, EN 1991-1-4:2005. This takes account of wind variation by introducing some dynamic load factors but treating the wind load essentially as a static lateral load.