These plants need the ability to shut down on their own, to continue the process for days without human intervention until any residual heat or water loss is no longer a problem and the sucker is stone cold dead ...
There's some other issues that need addressing as well ...
Like I said, common mode failure. The industry will have to look at what it will take for plants sited in various locales to assure that the backup power is maintained.
Nuclear plant shutdown is largely automated. I worked on a system in the early 1980s, ATOG (anticipated transient operating guidelines) that gave a visual representation of the plant status on a graphical display. A moving dot represented the plant state as a function of various process parameters, vessel pressure, downcomer temperature, time, main steam pressure, whatever the operator wanted to call up. The integrated control system would keep that little old dot right in the middle, the "sweet spot", of all the various conditions, to achieve cold shutdown as quickly as possible with minimal challenges to the safety systems.
From reading several places the newer plants can in fact do that....with the flow of water not active....
This old design of GE depends on those pumps running...and (I think) Hydrogen has to be injected into the stream to stop some of the corrosive actives in the fuel and control rods....
will try to get back to the blog where I was reading that....
The hydrogen Explosions may have been as a result of that process rather than Hydrogen generated from the Reactor....the discussion in the comments to the article didn't really resolve that however.
There is a boatlad of misinformation being spread around by all of the sources as far as I am concerned.
I realize I am not leaving sources here....will work on that...got challenges here doing stuff.