Identification in nuclear and thermal energy moderator temperature coefficient estimation via noise analysis and black-box modeling of heat transfer

This PhD thesis consists of two advanced case-studies of system identification. On the one hand, noise analysis techniques are applied to determine a safety coefficient in a nuclear reactor. On the other hand, heat transfer in a borehole heat exchanger is modeled.

The first part of this doctoral research investigates whether noise analysis techniques can estimate the Moderator Temperature Coefficient (MTC) of a nuclear reactor. MTC is an important safety parameter in a Pressurized Water Reactor. These represent about two-third of the current fleet of nuclear reactors worldwide. The determination of MTC is not straightforward at the end of an operational cycle. Reducing the uncertainty of the MTC estimate associated with the current measurement techniques has economic and safety advantages. Guidelines are established that determine the measurement time required to obtain a specified precision for a given measurement setup. It is shown that special experimental conditions are required to obtain an unbiased estimate of MTC. However, these experimental conditions could not be realized during the time period of this PhD. Hence, the decision was taken to continue the research on another application of identification.

The second case-study in this thesis concerns the modeling of heat transfer in a borehole heat exchanger. Geothermal boreholes are used in Ground Coupled Heat Pump (GCHP) systems that provide an energy efficient alternative to traditional air-conditioning systems for space heating and cooling. Heat is extracted from or injected into the ground through vertical borehole heat exchangers. The energy efficiency of the GCHP depends on the temperature of the fluid coming out of the borehole. A good model for the heat transfer can, for example, be used to develop a good control strategy for a GCHP system. Part II of this thesis explains how such a model can be found and what the advantages/disadvantages of this model are compared to the current state-of-the-art models. The realization of these two case-studies required the adaptation and extension of the classical system identification theory.