This is a general comment, more to start the discussion than an answer to your question, as I see you have got no response yet.
It is important to bear in mind following:
Normally you specify a valve mounted into a pipeline under the condition that the weight of the pipeline itself should be carried by separate supports, and not by the valve. The pipeline centers is requiered to be aligned, to prevent/minimize any transverse forces to the valve and valve flanges.
Ideally then, the forces between the valve flanges and pipeline flanges should be sufficient to give enough friction to hold the valve in place (aligned), and to keep the flange sealings correctly thightened and compressed without (extra / disaligned ) deformation. If the valve has a high weight (releative to 'normal' thightening forces), a separate support under the valve will be requiered, normally also with a sliding possibillity to neuter any axial movement by the pipeline by temperature differences over time.
In practise you will see more flanges thightened to an unnecssary high torque, or uneven over the bolt circle, or skew mounted or supporting pipeline by missing pipeline support, more often than you see flanges to loose thightened. Result by false thightening and mounting: leaking and damaged sealing material and surfaces. Even unproperly working valves will occur, for instance oftern by knife-gate valves.
Thigtening near yield strength could easily give yield forces by unforeseen extra transfer of forces to the bolts. Thightening force half the yield strength should in my mind be seen more as an upper limit than the necessary torque. 'Old proper workmanship rules' is often cited as 'normal fastening by hand plus a quarter turn, then test for any leakages'.
Bonnet thightening torque should be adviced by the factory, or seen similar to above, given that surfaces and sealing is 'as new'. If damaged, repair is required/recommended, not extra thightening as is often done both by leaking older bonnet-flanges and pipeline flanges.
