Cross-Section Adjustment Techniques for BWR Adaptive Simulation

Abstract

Computational capability has been developed to adjust multi-group neutron cross-sections to improve the fidelity of boiling water reactor (BWR) modeling and simulation. The method involves propagating multi-group neutron cross-section uncertainties through BWR computational models to evaluate uncertainties in key core attributes such as core k-effective, nodal power distributions, thermal margins, and in-core detector readings. Uncertainty-based inverse theory methods are then employed to adjust multi-group cross-sections to minimize the disagreement between BWR modeling predictions and measured plant data. For this work, measured plant data were virtually simulated in the form of perturbed 3-D nodal power distributions with discrepancies with predictions of the same order of magnitude as expected from plant data. Using the simulated plant data, multi-group cross-section adjustment reduces the error in core k-effective to less than 0.2% and the RMS error in nodal power to 4% (i.e. — the noise level of the in-core instrumentation). To ensure that the adapted BWR model predictions are robust, Tikhonov regularization is utilized to control the magnitude of the cross-section adjustment. In contrast to few-group cross-section adjustment, which was the focus of previous research on BWR adaptive simulation, multi-group cross-section adjustment allows for future fuel cycle design optimization to include the determination of optimal fresh fuel assembly designs using the adjusted multi-group cross-sections.

The major focus of this work is to efficiently propagate multi-group neutron cross-section uncertainty through BWR lattice physics calculations. Basic neutron cross-section uncertainties are provided in the form of multi-group cross-section covariance matrices. For energy groups in the resolved resonance energy range, the cross-section uncertainties are computed using an infinitely-dilute approximation of the neutron flux. In order to accurately account for spatial and energy resonance self-shielding effects, the multi-group cross-section covariance matrix has been reformulated to include the uncertainty in resonance correction factors, or self-shielding factors, which are used to calculate the self-shielded multi-group cross-sections used in the lattice physics neutron transport model. Using the reformulated multi-group cross-section covariance matrix, Efficient Subspace Methods (ESM) are then used to propagate multi-group cross-section uncertainty through the lattice physics calculation.