You're taking an interesting approach to the scenario, and I'm interested in how you get on too.
I would think you wll have to arrive at a whole bunch of inputs to the time history, one for each node in the area with the applied blast force distributed loads being converted into equivalent nodal forces. Don't forget to include the negative rebound phase of the blast wave as well into the time history. Are you going to establish a nominal blast center and have finite differences between the blast front hitting elements at different distances from the center.
Just remember the blast front moves very fast in relation to the typical dimension of piping so the active loading times are very short [i.e. for a blast front travelling at 300m/s it will take 0.7msec to pass a 8" line]
Best of luck - it's a massive undertaking