A multi-scale data integration algorithm for visualization of distribution lines based on deep learning

With the rapid development of smart grids, distribution lines generate multi-source heterogeneous data from sensor monitoring, historical load data, and image/video sources during operation. These data exhibit distinct multi-scale characteristics in terms of spatial, temporal, and semantic dimensions, posing significant challenges for efficient integration and real-time visualization. To address this issue, this paper proposes a deep learning-based multi-scale data integration algorithm for distribution line visualization. The algorithm employs a joint modeling approach using convolutional neural networks (CNNs) and recurrent neural networks (RNNs) to extract spatio-temporal features at different scales, and incorporates an attention mechanism to enhance the representation capability of key features. Additionally, graph neural networks are used to model the topological relationships of distribution lines, further optimizing data fusion performance. Based on this, a multi-modal data fusion and visualization strategy is designed to map multi-source data into a dynamic visualization of distribution lines, enabling intuitive presentation and real-time monitoring of line operational status. Experimental results show that this method outperforms traditional algorithms in terms of data fusion accuracy, visualization efficiency, and fault detection accuracy, validating its application value in smart grid monitoring and dispatching. The research findings are of great significance for improving the safety and reliability of distribution grid operation.