Title: Network protocol-based heterogeneous data mapping for distribution equipment

Authors: Hui Fan; Fan Peng; Yongchao Wu; Lijuan Gao

Addresses: State Grid Hebei Electric Power Co., Ltd., Shijiazhuang 050000, China ' Hengshui Power Supply Branch, State Grid Hebei Electric Power Co., Ltd., Hengshui 053000, China ' Hengshui Power Supply Branch, State Grid Hebei Electric Power Co., Ltd., Hengshui 053000, China ' Hengshui Power Supply Branch, State Grid Hebei Electric Power Co., Ltd., Hengshui 053000, China

Abstract: Current multi-source heterogeneous data mapping for distribution equipment faces challenges in data accessibility, real-time monitoring, and large-scale transmission. This paper proposes an improved method combining convolutional neural networks (CNN) with network communication protocols. It dynamically adapts CNN feature requirements at the protocol layer, adjusts data fragmentation according to convolution granularity, and automatically converts heterogeneous formats into tensor structures, realising collaborative optimisation of protocols and models. Experimental results show that under stable grid conditions, the proposed method maintains stable detection time: 3.56 ms at 1000 data points and 5.83 ms at 5000, significantly outperforming self-organising map and other mainstream algorithms. It effectively enhances mapping accuracy and operational efficiency, supporting real-time integration and utilisation of multi-source heterogeneous data in smart distribution networks while reducing equipment operation and maintenance costs.

Keywords: network communication protocol; convolutional neural network; multi-source heterogeneity; distribution equipment; data mapping.

DOI: 10.1504/IJWMC.2026.155719

International Journal of Wireless and Mobile Computing, 2026 Vol.31 No.2, pp.103 - 119

Received: 29 Apr 2025
Accepted: 13 Oct 2025
Published online: 11 Aug 2026 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article