If you do Time History analysis or Spectrum analysis, remember that you have to extract higher modes to a sufficient degree may be in the range 200-300 Hz as the stiff axial modes get excited in in Water Hammer Loads.

Also if you are using Joukowski eqn. to compute the dp i.e. pressure rise, you may get very high values, which actually may not be the case.Try to consult Hydraulics group of your company for more realistic dp values.It basically requires numerical solution of the Transient Bernoulli's eqn.

Controlling valve closure time in conjunction to the above method ( to be done in co-ordination with Hydraulics group) is the best possible solution to this problem.

Best Regards