Dynamic Planning and Intelligent Decision-Making Model of Power Grid Project Reserve Integrating Multi-Source Data and Deep Learning

In this paper, a dynamic planning and intelligent decision-making model of power grid project reserve based on multi-source data and deep learning is proposed to meet the new challenges of power grid project reserve and decision-making under the background of current energy transformation and smart grid construction. In the process of data fusion, hierarchical fusion strategy is adopted, including data preprocessing, feature extraction and feature level fusion. The deep learning module adopts a model combining convolutional neural networks (CNN) and long short-term memory (LSTM) to capture the spatial and temporal features of power grid data. CNN is used to extract local spatial features such as voltage fluctuations and current changes, while LSTM captures the temporal variations of these features. The fused eigenvectors are used as the input of the deep learning model, and the data base is provided for the dynamic programming algorithm by accurately predicting the power grid operation state and load demand. According to the prediction results, the dynamic programming algorithm comprehensively considers the project cost, income, implementation time and resource constraints, and optimizes the reserve strategy of power grid projects. Through the combination of dynamic planning and deep learning, the optimal allocation of resources and efficient management of projects are realized, aiming at minimizing operating costs, maximizing power grid stability and optimizing return on investment. The experimental results show that, compared with the traditional methods, the model has significantly improved the prediction accuracy and decision-making optimization effect, with the prediction accuracy reaching 91.8% and the decision-making optimization effect reaching 84.1%. This research not only improves the scientific and intelligent level of power grid project reserve, but also provides strong support for the sustainable development of smart grid.