Abstract: To improve the accuracy of investment risk assessment of hydropower BOT projects, and in view of the characteristics of influential factors of BOT investment risk such as nonlinear, vagueness and uncertainty, we build a hydropower BOT project investment risk identification matrix. Based on the risk identification result, we apply the radial basis function neural network to create a hydropower BOT project investment risk assessment RBF model, develop a risk assessment simulation program by Matlab and predict the risk level of investment project to provide a quantitative method for risk management. A case analysis shows that the investment risk value of the project is 0.5816, which is consistent with the actual situation, indicating that the model is feasible in investment risk assessment of hydropower BOT projects.