Research and Development Program in Reactor Diagnostics and Monitoring with Neutron Noise Methods. Stage 9. Final Report

This report constitutes Stage 9 of a long-term research and development program concerning the development of diagnostics and monitoring methods for nuclear reactors. The program executed in Stage 9 consists of four parts and the work performed in each part is summarized below. Development of reactor kinetics and dynamics for systems with non-stationary boundaries Studies of the space-time dependent neutron flux in systems with a non-constant volume, i.e. having time-dependent boundaries. They were motivated by the need of calculating the neutron noise induced by vibrating control rods and core barrel vibrations. A general formalism was developed in the frame of linear noise theory in the one-group approximation and homogenous bare systems. It was shown that within this formalism, closed form solutions can be obtained, and the original problem of a moving boundary can be replaced by an absorber of variable strength, placed at the static boundary of the stationary system. The various reactor kinetic approximations such as the point kinetic, adiabatic, quasistatic etc., are important in a number of diagnostic problems and regarding parameter estimation, including measurement of reactivity. The validity of the various approximations has been investigated in critical or near-critical cores. However the formalism of these approximations, on which the investigation of their validity is based, is not directly applicable to source-driven subcritical cores. Measurement of reactivity in a core during start-up represents the case of a source-driven subcritical core, in which the validity of e.g. the point kinetic approximation, on which most reactivity measurement methods are based, is not well understood. Hence we have proposed a systematic definition of the kinetic approximations in subcritical source-driven cores, a study of the related formalisms and an investigation of validity of the approximations. A PhD project is being conducted concerning the theory of the so called Cf-252 method for the measurement of the reactivity in subcritical cores.The existing Cf-252 detectors that were used by others so far have all been of the ionisation chamber type, where the Cf-252 was electrolytically fixed on one of the plates of the ion chamber. We have suggested another, simpler construction, which has its origin in a small scintillation detector construction which we previously used as a neutron detector. By replacing the neutron converter with Cf-252, a small size Cf-252 detector could be constructed. We have made an estimate of the possibilities of constructing such a detector. It was found that it is fully possible to manufacture a functioning Cf-252 detector with the scintillation principle, and the construction can be made in Chalmers. The possibility of extracting quantitative information by image processing of flow structures has been suggested by us a long time ago. Extracting time-dependent intensity data from various pixels, corresponding to various geometrical points in the flow, could make it possible to determine flow velocity profile as well as local radial and axial correlation lengths. By determining the distribution of the local correlations lengths and velocity profiles, useful information could be obtained for developing an objective flow regime indicator, among others. Performing such an analysis was hindered so far by the difficulty of extracting time-resolved signals from individual pixels, since earlier this required a so-called frame grabber with manual control for each frame, which resulted in a very slow and time-demanding procedure. However, digital video recording offers quite different, improved data acquisition possibilities. At the same time, in the frame of an on-going project at the Department, digital video recordings were made of the transport of injected ink in pure water, i.e. one-phase flow. These recordings gave the possibility of testing the principles of the method of correlation analysis of video images. In this Stage a first pilot investigation was made. Time-resolved signals from individual pixels of the recording were extracted, and by using two axially displaced pixels along the same streamline, the transit time of the ink in the water could be determined by correlation analysis. By selecting axially displaced pixels at various radial positions, a velocity profile in the pipe could be determined. The results are quite encouraging and a thorough investigation of the two-phase flow recordings is planned in the continuation