American Nuclear Society

Home / Store / Journals / Electronic Articles / Nuclear Technology / Volume 136 / Number 2

An Integral Response Scaling Method for Reduced-Height Test Facilities and Its Validation Using RELAP5 for Loss of RHR During Midloop Operation

Hyoung Tae Kim, Hee Cheon No

Nuclear Technology

Volume 136 / Number 2 / November 2001 / Pages 169-185


Member Price:$27.00
Member Savings:$3.00

An integral response scaling method for reduced-height test facilities is described and validated for the local and the integral phenomena using the RELAP5 thermal-hydraulic code. In the present schematic scaling methodology, the scaling laws are generated from four scaling analyses: three-dimensional differential equation-based scaling analysis, one-dimensional differential equation-based scaling analysis, component-based scaling analysis, and overall integral scaling analysis. Through use of integral response function, scaling parameters related to four types of requirements are identified: time scaling, initial-condition scaling, transient scaling, and bifurcation-phenomena scaling requirements. In the present scaling method, flow velocities in the vertical channel and at break locations of the scaled-down model can be preserved setting up the scaling as R = lR. The significance of the velocity preservation in the reduced-height facility is demonstrated by the RELAP5 code.

To validate the present scaling method, similarity criteria from the integral response scaling method are applied to the scaled-down model of the Korea Standard Nuclear Power Plant simulating core boil-off process during midloop operation. The scaled-down model is basically developed with the length and area scales of 1/5 and 1/100, respectively. The integral simulation and local calculations for pot-boiling, blowdown, heat transfer in the steam generator, and off-take are conducted for a loss of RHR during the midloop operation to show the preservation of the similarity criteria using RELAP5.

It turns out that the scaled-down model based on the present scaling method maintains well the similarity of the nondimensional parameters for the local phenomena. Furthermore, the integral simulation indicates that there are no noteworthy differences in the general trends between the prototype and scaled models.

Questions or comments about the site? Contact the ANS Webmaster.