Natural circulation loops represent important elements of many technologically relevant systems. For this reason, their instability represents a major concern, as it consists in dangerous oscillations leading to flow reversal. The analysis of these processes was addressed in various theoretical works, mainly based on mathematical approaches to the problem. The models obtained in this way suffer of a poor correspondence between simulated and experimental data. To solve this problem, in this paper, the identification of the system was adopted and a neural network model has been obtained by means of input-output measurements detected on an experimental natural circulation loop. Moreover, the neural model has been used in a predictive scheme, in order to allow long-term prediction of the birth of unstable behaviors.