Study on Geometric Matrix of Axial Power Reconstruction in PWR

Based on the polynomial fitting, the geometric matrix model in the algorithm for axial power reconstruction from the current signal of the ex-core nuclear instrumentation detectors is deduced, and the verification is achieved by numerical calculation and engineering measurement data for CPR1000 reactor type. Geometric matrix models for different numbers of section power signal and model suitable for self-powered detector are deduced, the error analysis of the 4-section power signal reconstruction is carried out, and the influence of the input signal number on the reconstruction accuracy is discussed. The results show that the essence of the geometric matrix is a mapping relationship that expands the power input of a finite number of sections into a full axial distribution, the matrix includes geometric information such as section division, axial extrapolation distance, and power reconstruction point position; the 4-section power reconstruction accuracy cannot meet the requirement of engineering applications, and the signal of self-powered neutron detector should be used as the input for axial power reconstruction.