ABSTRACT

The objective of this work was to implement and test a method for the determination of the control rod positions in a Pressurized Water Reactor (PWR). The method was developed earlier in a co-operation between the Dept. of  Reactor Physics, Chalmers University of Technology, and the Ringhals Power Plant.

The method is based on the information in the axial flux shape on the control rod position. The relationship between rod position and flux profile is rather implicit and cannot be explicitly inverted. Artificial Neural Network (ANN) techniques are used to unfold the rod position from the measured flux profile. More specifically, it is possible to determine the position of a control rod in the D-Bank from the flux measurement performed in a controlled assembly or the neighboring assemblies.

Programming codes and methods have been developed to be used for the determination of the D-bank positions in the range between coast-up and coast down in a cycle. The codes were developed by using real measurements from Ringhals 4 power station and by simulated data obtained with the Studsvik Scandpower reactor core simulator SIMULATE3.
 

Gothenburg, Sweden 2000

Albert Nagy