ICONE23-1354 PREDICTION MODEL OF EXPLOSION STRENGTH OF LIQUID LITHIUM WATER INTERACTION BASED ON A BP NEURAL NETWORK

Abstract

Liquid lithium is a promising plasma facing material in magnetic fusion devices. However, liquid lithium water interaction under the conditions of loss of coolant accidents is a principal concern to the safety of fusion devices. The prediction of explosion strength of liquid lithium water interaction is significant to the analysis and assessment of related accidents in fusion reactors. As a preliminary investigation, an experiment of lithium water interaction on the water has been conducted. The pressure and temperature in the test section were recorded during the violent interactions. The results indicate that the mass of lithium, initial lithium temperature and init ial water temperature are key factors to the explosion strength. It is a kind of complicated nonlinear relation between explosion strength and its influencing factors. Therefore, a BP neural network model for predict ing the explosion strength has been developed and the prediction results are consistent with the experimental data. The BP neural network prediction model is applicable and provides a novel method for the evaluation of liquid lithium water interaction.