Axial neutron flux density distribution in RBMK-1000: reconstruction using the Prizma-M program

Abstract

A computational-experimental procedure for reconstructing the axial neutron flux density and powerrelease distributions using the Prizma-M program is described. An approach to reconstructing the axial neutron flux density and power-release distributions in the RBMK-1000 assembly channels by using extended detectors with a hafnium dioxide emitter to scan the core is presented. This approach made it possible to compare the reconstruction accuracy of the semi-empirical and computational-experimental procedures used in Prizma-M. It is shown that the reconstruction error for the axial neutron-flux density distribution does not exceed the attested error of the Prizma-M program.A method of reconstructing the axial neutron flux density and energy release distributions was adopted in 1985 in RBMK-1500 at the Ignalina NPP. Based on a semiempirical method, it uses a harmonic expansion of the flux density distribution without a neutron-physical calculation and signals from eight neutron detectors belonging to the control-and-protection system (CPS) and uniformly distributed over the height of each of 20 assemblies [1]. Ionization chambers placed in (n, γ) converter channels served as fast-response neutron flux density detectors. In 2002, this method was incorporated into RBMK-1000 in the newly introduced program Prizma-M of the SKALA-Mikro information-and-measurement system. A special feature of the Prizma-M program here was that 72 four-section in-reactor height-monitoring detectors were used. Each section with a hafnium dioxide emitter had a sensitive part with length equal to 1/4 the core height. The length of the communication line section located in the core was equal to 3/4 the core height. The top and bottom parts of the communication line were connected to the emitter, which made it possible to decrease the uncertainties related with the background signal of the communication line.For the first nine years, a semiempirical method was used in the Prizma-M program. Subsequently, on the basis of positive experience in reconstructing the radial-azimuthal distribution by means of a neutron-physical calculation, the computational-and-experimental procedure was adopted for reconstructing the axial distribution of the neutron flux density and energy release, retaining the semiempirical method in the auxiliary operations.In the computational-and-experimental method, just as in the semiempirical method, an expansion of the form (1) is used for the axial flux density distribution of thermal neutrons in the jth assembly channel. Here k is an index of a point along the core height, k = 1, 2, ..., 28; h k is the distance to the kth point from the core top, m; B j is a vector of the coefficients in the harmonic expansion of the axial neutron flux density distribution in the jth assembly channel; ϕ T (h k ) = [ϕ 1 (h k ), ϕ 2 (h k ),