SAFPWR home page
Simulation of unsym mslb by means of sg1 temp temperature decline
ref: rtvunsym_2c.dat (download)
We already noticed that, at "state point" condition, when safety variables reach their most adverse value, the core is almost critical, the unfaulted sg do not exchange much power with primary water, and the faulted sg1 conditions (p3n1, qcn1) are almost stationary.
Thus, it should be conceivable to try "capturing" the state point condition by simulating the sg1 effect through a slowly declining tcn1 ramp. This is made possible by enabling the option loop_1 / Lstl1 / input_t1cn= t
S1= .., steam_gen_1, steam_gen_2,..
must still be entered for the sole purpose of specifying heat_exchange coefficients on the secondary side of the sg's. Thereafter steam_gen keywords are no longer invoked.
read /
S00gvg /: convergence parameters
epsqgv and
omqgv are solely required for allowing adjustment of initial t1cn2 field in such a manner that qcn1 = total pumps power.(cf ini_tinj )
loop_1
Lstl1
input_t1cn= t: sg1 is represented as a uniform, but time-variable, secondary temperature field (t1nj(j)= t1cn.
t1cn= 560.436: guess value.
We select here the initial saturation temp tln1 calculated by the reference transient run.
Itp_sec/
sec= 0, 10, 2000/: t1cn ramp started at sec=10
t1cn= 2*560.436, 480: initial value up to moment where rauc is set to 2%.
Normally the application must be run in two parts: the first limited to initialization, search for consistent t1cn, which must retained in the second part as initial value for the t1cn(sec) ramp.
This complication could be avoided by specifying dt1cn(sec), the deviation from initial value, in place of t1cn(sec)
The declining t1cn ramp must be slow enough for keeping the core nearly critical during the phase of interest, when min margins are observed.
As the ramp is slow, a large value dsec= 2 can be tolerated.
480 K is an arbitrary value lower than the state point value observed in the reference case.
loop_2 / Lstl2
jqli= 11*0: to force sg2 to remain isolated.
The water in the unfaulted loops has the only effect of mitigating core water cooling and creating radial temperature gradient in it.
Steam_gen data are limited, as already noted, to j9n and rfnj specification.
safety_inj / Lstis
sec= 1000: boration by SI is not relevant because the core will remain critical anyway.
SI is enabled here for controlling pressu response so that the pressure at state point be more or less replicated.
Results
After a transient, of no interest, for returning core to power, one observes (chart 01, chart 02), as expected, a monotonic succession of critical cores; ql1gv - q2c is just the total pump power.
The conditions at sec= 2200 of the present case replicates fairly well those of the ref case at sec= 250.
This makes it feasible to test the SAFPWR transient model by means of static neutronic design tools, or even to carry out real tests bringing the core safely to conditions close to the state point and take system and in-core measurements to compare with the predictions.