Abstract
This paper aims to construct a data set that can be used to train neural networks to furnish the power density peak factor during reactor operation. The inputs considered were those available in the reactor protection systems, namely, the axial and quadrant power differences obtained from measured ex-core detector signals, and the position of control rods. The response of ex-core detector signals was measured in experiments performed in the IPEN/MB-01 zero-power reactor. Several reactor states with different power density distribution were obtained by positioning the control rods in different configurations. The power distribution and its peak factor were calculated for each of these reactor states. The obtained results show that the power peak factor correlates well with the control rod position and the quadrant power difference, and with a lesser degree with the axial power differences. The data presented an inherent organisation and could be classified into different classes of power peak factor behaviour as a function of position of control rods, axial power difference and quadrant power difference. The analysis of the data set indicates that the power peak factor can be determined through a neural network having as input the position of control rods. Regarding only signals of ex-core detectors, the data indicate that a neural network may estimate better the power peak factor if the input vector comprises both the axial and the quadrant power differences.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.