Optimization Application of the Whole Process Control Strategy of Power Grid Infrastructure Projects Based on Bi-LSTM

Nowadays, there is a lack of effective management and control strategies in the process of power grid (PG) infrastructure projects, and there are problems such as delayed progress and insufficient resource allocation. This paper introduces a management and control strategy optimization model based on Bi-LSTM to improve the accuracy of project management and real-time adjustment capabilities. Firstly, the key indicators of the project, such as key progress, resource consumption and risk control, are selected as input variables to build a project management and control model based on Bi-LSTM (Bidirectional Long-Short Term Memory), and data preprocessing such as denoising and normalization is performed. Secondly, by using the bidirectional structure of Bi-LSTM, a real-time feedback mechanism is established to dynamically predict the changing trends of various indicators during the project execution. The model learns historical data from different stages of the project to identify possible risk points and key nodes, so as to optimize resource allocation and provide risk warnings in advance. Finally, based on the actual needs of PG infrastructure projects, a refined decision support system is developed to ensure that the results of model prediction can be quickly converted into actionable project adjustment plans. The experimental results show that the optimization application of control strategy based on Bi-LSTM can significantly improve the control effect of PG infrastructure projects in terms of progress and resource utilization. In the later stage (after step 15), the predicted values are closer to reality, such as the predicted progress in steps 18 and 19 reaching 91.13% and 93.75<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">%</sup>, almost accurate. The optimization application of control strategy based on Bi-LSTM not only has advantages in resource optimization and risk control, but also can improve the flexibility and efficiency of project management through dynamic adjustment, providing practical and feasible optimization solutions for the full process control of PG infrastructure projects.