Optimal and fast field reconstruction with reduced basis and limited observations: Application to reactor core online monitoring

Abstract The fast reconstruction of neutronic field in a nuclear core using reduced modeling and limited observations has attracted considerable attention. In particular, four design parameters are considered for developing efficient and robust field reconstruction in this framework, including the choice of reduced basis, the order of reduced dimension, the number of sensors and their placement. In this work, a systematic study has been brought to show the effect of these design parameters on the performance of field reconstruction by investigating i) five basis selection methods and ii) five different sensor placement methods. From a series of practical engineering applications based on HPR1000 reactor core, it is observed that the POD models and the sensor placement determined by the discrete empirical interpolation method and the oversampled discrete empirical interpolation method provide an optimal reconstruction, where the computational complexity, the accuracy, and the robustness to noise are well-balanced. The result is helpful for the practical implementation of the neutronic field reconstruction methods with reduced basis and limited observations in nuclear reactor cores.