Predictive diagnostics of RBMK reactor flowmeters

Abstract

At the moment, there are 8 rectors of the RBMK type in operation. The safety of their operation is ensured by monitoring devices for key indicators. In this article, the main attention is paid to the control of the coolant flow through the fuel channel, carried out with the help of ball flow meters of the STORM-32M and STORM-8A types. During operation, there is a gradual wear of the raceway and the ball, which leads to the appearance of a negative error in the device, in other words, the readings become underestimated. In this regard, as well as due to the widespread use of ball flowmeters at nuclear power plants, the task of diagnosing the state of flowmeters is urgent. One of the approaches to determining the flow rate of the coolant is an approach based on the measurement and use of information about the activity of the coolant by the tightness control system of the fuel element shells (SKGO). The use of this radiation approach to control the flow of coolant has its advantages and difficulties. The advantage is that real experimental information about the presence of a coolant in the fuel channel is involved, and the difficulties are associated with receiving and processing the signals of the SCGO in real time and, consequently, with low efficiency of solving the problem. In this paper, we propose a method for predictive diagnostics of the flow meter condition using the design features of the multiple forced circulation circuit (CMPC) of the reactor. The adaptive parameter of the  model was selected as a diagnostic parameter. Studies have shown that the increase in time variance in experimental flow values has a clear effect on the selected diagnostic feature , as a result of which it can be said that the observation of this kind of variance change in experimental data can serve as a sign of a malfunction of the flow meter.