Implementation and comparison of least-square method and thin-plate spline method for PWR online power-distribution monitoring based on NECP-ONION system

Abstract

Among many methods for online power-distribution monitoring, the least-square method (LSM) and the thin-plate spline method (TPSM) are two practical and promising methods for PWR application. These two methods have been implemented in the NECP-ONION system, which has the capability to perform the 3D PWR online power-distribution monitoring including plant data processing, core calculation, online power-distribution monitoring, and sensitivity and uncertainty analysis. It is found that the weighting factor and the regularization coefficient are the most important parameters, defined as key parameters, for the LSM and the TPSM respectively, and thus should be determined in advance. The online power-distribution monitoring and uncertainty analysis with various key parameters are performed for BEAVRS benchmark problem. Numerical results show that, key parameters have important impact on the monitoring accuracy and uncertainty. With the weighting factor increase or the regularization coefficient decrease, the monitoring accuracies improve but the uncertainties of monitored results increase simultaneously. Therefore, a strategy is proposed that the key parameter is determined based on the uncertainty requirement of the online monitoring system. Then, the comparison of the monitoring accuracies for these two methods is carried out with the same maximum-uncertainty values; the monitoring accuracy of the LSM is better than TPSM on the condition of this paper.