Traditionally, the brittle strength evaluation of reactor pressure vessel was the central issue in lifetime assessment of Ukrainian nuclear power plants (NPPs). The problem of swelling of the reactor core baffle only recently got due attention from the side of operator.
Here the most efforts were given on numerical modeling of austenitic steel 08Kh18N10T swelling and its effect on induced stresses in core baffle and distortion of its geometry. The calculation shows that essential changing of core baffle dimensions is expected after 35–40 years of operation. Eventually this can lead to the contact with the core barrel. Yet, these predictions contain the big number of uncertainties related to the input data used in analysis: fluence distribution; temperature variation due to heat release induced by neutron and gamma radiation; thermal-hydraulic boundary condition between the baffle and coolant; and, especially, the adopted law of swelling in dependence with above factors as well as mechanical stresses.
So, the second task was to measure the real geometry of baffle after 27 years of operation, to determine its change and compare these results with the numerically calculated data with accounting for the design tolerances. Thus, the spatial measurement system (SMS) equipped with ultrasonic gages was designed. It contains the central vertical beam which can move in vertical direction and rotate. To the lower end of the beam four horizontal levels are attached, which are equipped with device resistant to the hot water and radiation. The gages are used to measure the shortest distances to the edges of baffle. Two types of results were obtained. The first one are the measurements in the different horizontal planes obtained by rotation the SMS around the vertical axis with angular steps equal to 1 degree. These results were difficult to handle with and required a special mathematical treatment due to the possible shift of the centre of measurement. The second set of measurements was performed by moving the SMS in vertical direction. These data demonstrate the change of distance with the height.
The results clearly show that problem of swelling do exists, and, in general, the measured patterns of the distortions along the vertical and angular coordinates correspond to numerically obtained results. Further work on baffle integrity is however needed.
