Abstract: With the continuous advancement of the energy revolution, accurate investment in power grid has become an effective means to meet the growth of power demand and stimulate the internal potential of power grid enterprises. Accurate investment is reflected in the two aspects of "stability" and "precision". Scientific investment system requires power grid enterprises to use more effective investment demand forecasting methods. At present, the analysis method of influencing factors of power grid enterprise investment is relatively simple, and the prediction accuracy of investment demand is not high. Based on the Delphi method, this paper analyzes the internal and external factors that affect the investment demand of power grid, uses the gray correlation analysis method to screen out the factors that have a greater impact on the demand forecast, and constructs the PSO-GM（1, N） investment forecast model of power grid. The simulation training of the investment demand forecast of Shandong Power Grid in recent years verifies the feasibility of the model, which effectively improves the investment demand of power grid. The effectiveness and accuracy of the prediction are helpful for management decision-makers to grasp the future investment trend and lay the foundation for the optimization of power grid investment.