Validation of Physics Model of FARE Tool by Comparison of RFSP Flux Simulations with Measurements at Discrete Rebelling Steps
Main Article Content
Abstract
In CANDU 6 the on-power refuelling is done in the direction of the coolant flow through the channel, flow being in alternate direction in adjacent channels. The channel flow pushes the fuel string (a total of 20 bundles with the 8-bundle fuelling scheme) towards the downstream Fuelling Machine (FM), obviating the need for the upstream FM to insert the rams into the active core to push the fuel. In some channels of the outer core region, however, the coolant flow is low and the hydraulic drag is not enough to push the fuel string. In these channels a Flow Assist Ram Extension (FARE) tool is used during refuelling to augment the flow and hence enhance the hydraulic drag required to push the fuel string. Refer to Figure 1 for detail.
The FARE-Tool is a strong neutron absorber and its use results in a severe local flux depression. During the refuelling process, when the FARE-Tool travels in the immediate vicinity of a Reactor Regulating System (RRS) detector, the spatial control system causes a short-term zone-fill reduction in an attempt to maintain the reference zone powers. This short term zone drain may in turn result in a single channel ROP trip of the SDS1 or SDS2 reactor protection system. Therefore, the control room operator is always interested in minimizing the occurrence of this category of trip for obvious economic reasons. The physics modelling of the FARE-Tool used in the 2-neutron-energy-group 3-dimensional diffusion code RFSP is validated by comparing the simulated response of the in-core detectors with the measured response of the same detectors during refuelling. These measurements were recorded during a specially scheduled refuelling of channel B09 at the Point Lepreau Generating Station on 1993 July 6 at 3591 Equivalent Full Power Days (EFPDs). A ten-minute pause was ensured at each step of refuelling. The simulation-versus-measurement comparison was done using the detector response for the 102 FLUXMAP vanadium detectors, 24 ShutDown System No. 1 (SDS1) detectors, 24 ShutDown System No. 2 (SDS2) detectors and the 14 Reactor Regulating System (RRS) detectors. In addition to this the simulated absolute fluxes (i.e., fluxes computed by *INTREP module of RFSP) were compared with the measured fluxes using a least-squares-fit methodology. The standard deviation of the difference between the simulated detector response and the measured response for the FLUXMAP, SDS1, SDS2 and the RRS detectors is less than 1%. Similarly the agreement in the individual simulated zone-fill (with spatial control invoked) and the measured zone-fill in most of the refuelling steps is within +/- 5%. This indicates that the FARE-Tool physics model is very good, recognizing the fact that there is approximately 2% noise in zone-fill measurement and that a ten-minute pause at the fuelling steps is not sufficient for the vanadium detectors (whose time constant is about 6 minutes) to reach their asymptotic reading.
The FARE-Tool is a strong neutron absorber and its use results in a severe local flux depression. During the refuelling process, when the FARE-Tool travels in the immediate vicinity of a Reactor Regulating System (RRS) detector, the spatial control system causes a short-term zone-fill reduction in an attempt to maintain the reference zone powers. This short term zone drain may in turn result in a single channel ROP trip of the SDS1 or SDS2 reactor protection system. Therefore, the control room operator is always interested in minimizing the occurrence of this category of trip for obvious economic reasons. The physics modelling of the FARE-Tool used in the 2-neutron-energy-group 3-dimensional diffusion code RFSP is validated by comparing the simulated response of the in-core detectors with the measured response of the same detectors during refuelling. These measurements were recorded during a specially scheduled refuelling of channel B09 at the Point Lepreau Generating Station on 1993 July 6 at 3591 Equivalent Full Power Days (EFPDs). A ten-minute pause was ensured at each step of refuelling. The simulation-versus-measurement comparison was done using the detector response for the 102 FLUXMAP vanadium detectors, 24 ShutDown System No. 1 (SDS1) detectors, 24 ShutDown System No. 2 (SDS2) detectors and the 14 Reactor Regulating System (RRS) detectors. In addition to this the simulated absolute fluxes (i.e., fluxes computed by *INTREP module of RFSP) were compared with the measured fluxes using a least-squares-fit methodology. The standard deviation of the difference between the simulated detector response and the measured response for the FLUXMAP, SDS1, SDS2 and the RRS detectors is less than 1%. Similarly the agreement in the individual simulated zone-fill (with spatial control invoked) and the measured zone-fill in most of the refuelling steps is within +/- 5%. This indicates that the FARE-Tool physics model is very good, recognizing the fact that there is approximately 2% noise in zone-fill measurement and that a ten-minute pause at the fuelling steps is not sufficient for the vanadium detectors (whose time constant is about 6 minutes) to reach their asymptotic reading.
Article Details
Section
Articles