I'm modeling a above ground piping and buried pipeline, the buried length is very longer.I have no idea how long the buried pipeline shall be inputed. Somebody says use virtual anchor.
Can somebody tell me how to define the length for the buried virtual anchor to be modelled in the program? and how to define the buried model to similar with real one?

You can find many reference material about this. I would like to recommend to read an article ”Stress analysis methods for underground pipe lines, L.C.Peng, May 1978, Pipe Line Industry”.

Virtual anchor location may be calculated roughly as below;

L = Ft / Ff
where,
L = length of pipe from virtual anchor to the point of entry or exit from the ground(ft)
Ft : force due to thermal compressive stress (lbs)
Ff : frictional resistance of pipe to thermal growth (lbs/ft)

However, it will depend on buried conditions. If you want to get more correct number, you have to use computer software. There are some software for buried pipe and soil analysis.
I don’t see a problem to model your piping even a few kilometers into Caesar II. Caesar II will give the virtual anchor location and thermal movements.

Thanks to Sun Wee for thinking of L.C. Pengs's papers. You can find these papers at the following address:

http://www.pipestress.com/Pages/PEpapers.html

Also, you might want to look at the non-mandatory appendix in B31.1, Appendix VII, "Procedures for the Design of restrained Underground Piping".

Regards, John.

Be careful with your interpretation of the phrase "virtual anchor" or "virtual anchor length". This is not a distance to an anchor. It is instead a buried length of pipe, long enough so that the axial friction along this length prevents any load or displacement at one end from translating to the other end. A straight run with no branches longer than this length need not be modeled beyond this length.

But don't add an anchor in CAESAR II (CAESAR II). As Sun Wee suggests, just run it way out there and let CAESAR II "bury" it.

CAESAR II prints out the calculated Virtual Anchor Length (VAL) in the buried modeler. But this length assumes the soil holds the pipe fixed until the friction load is exceeded. Our soil model uses a bilinear stiffness to model these friction effects so some "slippage" must occur to build up the friction force along the length. (This slippage is the x in f=kx.) This requires longer runs to set up the "anchor". I run lengths much longer than the calculated value.

There are some closed form solutions available to estimate the virtual anchor length for a buried pipeline that is unrestrained at one end, and is subjected to internal pressure and temperature differential. However, these approximations assume the axial soil reistance is fully mobilized along the virtual anchor length (which is not exactly correct). The virtual anchor length calculation is a function of many variables such as pipeline diameter, thermal expansion corficient, soil type, internal pressure, and the change in temperature that the buried pipeline experiences between installation (i.e. temperature at which the line was buried-restrained) and operation. P.J. Schnackenberg (Pipe Line Industry, Nov. 1976, pp. 53-56)presents a simplified solution for virtual anchor length.

However, if you use CAESAR II, just keep increasing the buried pipline segment length until the nodal displacemnts go to zero at some point along the buried pipeline segment. By definition, this is the point at which the virtual anchor occurs.