To calculate the structural strength of this box-keel, you would need access to a FEA code such as ALGOR. A simple static linear version would be adequate. I used ALGOR for all the FEA work on my hull. It would be important to model the stiffened cylindrical hull and the stiffened box-keel stimulatingly subject to an external pressure say 1.5-2.0 times the max operating pressure. You could then examine the stresses in every area of the combined hull and box-keel to see if you have reached the maximum allowable stress for the A516 Grade 70. As mentioned by Dan, the slot that is cut out of the pressure hull will seriously compromise the strength of the pressure hull. For 300' ft of fresh water (130 psig), for a 12"x72" box dimension, there would be an equivalent of a 112,000 lbs force on each side of the box trying to collapse the box/hull. By cutting out a section of the "T" stiffeners from the pressure hull, you significantly increase the likelihood of overall buckling of the pressure hull. The other problem with this design is that flat plates, even with stiffeners, don't take external pressure very well at all. At the mid bay between stiffeners, the stresses would be high. My guess is that if you ran an FEA on this design, to keep the max stresses in the elastic range you would have to limit yourself to a very shallow depth. I would be very careful in adopting this concept for a 1 atm boat even with using 1/2" plate and 1/4" stiffeners for the box. If left the "T" stiffeners in tack for the cylindrical section, your maximum operating depth would improve but would still be significantly below that of the K-boat.
----- Original Message -----