Search

The problem of predicting the oscillations of the main thermodynamic parameters of the core in the first loop of the sodium cooling system of IBR-2M reactor is examined. Attention is focused mainly on the prediction of the temperature and sodium flow at the entry into the core as well as the thermal power. It is shown that the prediction makes it possible to reduce by a factor of 3 the influence of slow oscillations of reactivity on the power and thereby reduce the operational requirements for the automatic power stabilization system. Neural-net prediction using nonlinear autoregression nets with feedback is proposed. The results agree with experiment to within ~5%.

The pattern recognition methodologies and artificial neural networks were used widely for the IBR-2M pulsed reactor noise diagnostics. The cluster analysis allows a detailed study of the structure and fast reactivity effects of IBR-2M and nonlinear autoregressive neural network (NAR) with local feedback connection allows predicting slow reactivity effects. In this work we present results of a study on pulse energy noise dynamics and prediction of liquid sodium flow rate through the core of the IBR-2M reactor using cluster analysis and an artificial neural network.

In this work we present the results of a study on pulse energy noise dynamics of the IBR-2M; the study employs statistical methods of time series (pulse energy) processing and hierarchical cluster analysis. It is shown that the power spectrum changes of the pulse energy fluctuations per cycle (∼11 days) has a transition region duration of ∼3 days that takes place after the reactor has reached the nominal power of 2 MW. The power noise is subsequently divided into four stable clusters, with three of which describe the noise transition region. The fourth cluster constitutes a stable structure that does not depend on noise level (amplitude of the power spectrum) or on reactor operation time. The noise transition region is formed by the vibration of the moving reflectors once the reactor has been reached.

This paper presents an artificial neural network method for long-term prediction of liquid sodium flow rate through the core of the IBR-2M reactor. The nonlinear autoregressive neural network (NAR) with local feedback connection has been considered as the most appropriate tool for such a prediction. The predicted results were compared with experimental values. NAR model predicts slow changes of liquid sodium flow rate up to two days with an error less than 5%.