No abstract

Aiming at the problems in the actual construction process of power grid infrastructure projects, such as the lack of time series data, the difficulty in analyzing the correlation between engineering quantity and investment amount, and the deviation of investment budget caused by engineering quantity estimation deviation, this paper studies the power grid infrastructure based on Bootstrap data expansion and gray correlation, Project engineering investment forecasting model. Firstly, the time series distribution of various expenses and engineering costs is analyzed, and the Bootstrap data expansion method is used to construct various cost and expense databases and cost data distribution characteristic models of power grid infrastructure projects. , established a calculation and evaluation model of the correlation between engineering quantity and value of power grid infrastructure projects based on the grey relational analysis method; through the coordinated preparation and adjustment model of the designed plan and budget, a long-term short-term memory neural network-based grid infrastructure project investment prediction model was established. Finally, through the analysis of simulation experiments, it is shown that the model studied can not only improve the accuracy of investment plan and budget management, solve the problem of inconsistency between investment plan and budget and actual project progress, reduce waste and reduce cost; it can also realize the infrastructure construction of power grid enterprises. The project is more standardized and efficient implementation.