Prediction of Nuclear Power Project Cost Based on Improved Non-Equidistant GM (1,1) Model and Exponential Regression Model

Today, faced with the problem of international energy supply difficulties, due to the nuclear power’s fewer of its resources consumption, little environmental pollution and other advantages, it has become one of the essential pillars of the electricity supply. This paper analyzed from a comprehensive localization of nuclear power plants, then use the non-equidistant GM (1,1) model and the exponential regression model to predict the construction cost of nuclear power, and do the error analysis. The conclusion that by increasing the degree of localization of domestic nuclear power construction, which can effectively reduce the cost of building nuclear power plants conclusions.