It seems as if there is some confusion.
I assume you are not referring to a cogeneration facility where there may be a required export steam pressure requirement, and are instead referring to operating the HRSG HP evaporator under either constant pressure CP or under variable pressure VP , when the plant is being dispatched from the ISO .

The great majority of installed HRSG's are drum type boilers and are not once thru steam generators, so I will assume a drum type boiler.

The method of operation of the HP evaporator ( CP or VP) is defined early in the selection of the major capital equipment. If the steam turbine is selected as a full arc admission machine it usually operates best as a VP machine, with the HP throttle valves at least 95% open in the load range 65-100% MCR. Minimizing the throttling of these valves reduces steam cycle efficiency losses, as throttling always lowers cycle efficiency. However, it requires the HP evaporator to change its pressure and saturation temperature during VP load changes , which involves changing the energy content of a large mass of water and steam. This implies that there will be a dynamic hysteresis delay due to the thermal inertia of this water and metal mass changing temperature. To account of this effect, the spray attemporation system may need to be oversized and the superheater metal selection may need to be upgraded due to the temporary overheat that occurs during rapid increases in load under VP conditions.

If the steam turbine is a partial arc machine ( 4 throttle valves in parallel) then it can be operated as a CP machine, and the HP evaporator can operate under constant throttle pressure over the load range 65-100% MCR. There is no significant dynamic hysteresis caused by the thermal inertia of the HP evaporator water and metal as its metal temperature will change only slightly during the load increase. The main dynamic issue is the steam turbine's temperature will change downstream of the throttle valves as load changes , which leads to fatigue damage to the steam turbine and also lowers the cold reheat steam temperature as load drops.

Once thru steam generators have much lower inventory of water and metal so their thermal inertia is much smaller than a drum boiler, and the VP issues are not significant in that case.

"In this bright future, you can't forget your past..." Bob Marley

If it is like at our power station, then:

Pressure mode controls turbine first stage pressure (which is closely proportional to turbine power) and operates the governor valves to keep first stage pressure constant. Grid frequency changes are ignored.

Power mode, I believe, allows a power level to be set but allows variation of the 'set' power as the grid frequency varies, according to a "Droop" curve prescribed by the grid controllers.

Both are standard control modes so switching from one to the other is not an issue, but could be an issue for the grid or other 'external' things.

"Primary frequency response" could be like our droop curve. It defines how the suppliers to the grid react to frequency changes, which are usually due to load changes.

I can't answer the question about combined production, but my guess would be that the control system compensates automatically.

And I can't help with the books question.

see following related thread: http://www.eng-tips.com/viewthread.cfm?qid=415569

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]