On line surveillance of core parameters for a pressurized water reactor

A surveillance algorithm for monitoring the Doppler coefficient and the overall fuel-to-coolant heat transfer coefficient for a pressurized water reactor has been developed for on line applications. The surveillance is achieved by comparison of the values of the parameters given by an adaptive identification process to the reference values corresponding to normal behaviour. The on-line adaptive identification method is based on a quasilinearization technique, sensitivity function calculation and a recursive least squares algorithm. The method was tested using a theoretical model for the Oconee I PWR and experimental results obtained during operation of Oconee I. The physical model of the plant consists of a stable variable model of 29th order where only some components of the state vector are measurable. The tests were achieved by adding a small time-varying perturbation signal to the flux controller set point while the reactor was operating at 95% of full power