Abstract: Inter-basin water transfer projects suffer from the issues of unclear water supply risk transfer laws and unsatisfactory dynamic risk management effect. To explore the risk transfer law, this paper identifies risk factors, quantifies the risks using the expert scoring and mutation coefficient method, and develops a water supply risk transfer model based on the infectious disease dynamics. We have applied this model in a case study of the initial stage of the Hanjiang-to-Weihe River Water Diversion Project, and examined six risk schemes to reveal its risk transfer law. The results show the number of months under water supply risks is positively correlated with the number of risk sources. When the runoff forecast error risk is not considered, an earlier implementation of the optimal operation measures leads to a shorter water supply risk period and a higher recovery level; When the error risk is considered, the control effect of optimal operation measures on water supply risks is insignificant. No positive correlation is observed between the risk transfer capability and the number of risks; risk damage is positively correlated with the number of risk sources. The results are of great significance for prevention and control of water supply risks and risk reduction and control programs of a water diversion project.

Key words: infectious disease dynamic model, water supply risk, transmission law, back-propagation neural network, inter-basin water transfer project