My preference for this would be the addition of a small dummy rigid element of the correct weight (mass) at the CG, as Michael suggests above.

The concentrated force will work also. The UDL method will work as long as you set the magnitude back to zero on the first pipe element following the exchanger. There are two problems with both the Force and UDL methods. First you have to be sure to add "F1" or "U1" to all of the static load cases that contain the "W" primitive. Second, neither "F1" or "U1" will transfer into dynamics if you intend to do further analysis. The additional dummy rigid is the safest way to go.

I personally suggest you DO NOT call the exchanger body rigid, it's not. Vessels are just large diameter pipes, and if you ever go vertical you would want to see wind deflections. Therefore you shouldn't indicate that vessels are rigid.