Title: Industrial internet of things data monitoring algorithm based on improved graph convolutional network

Authors: Bin Hu; Changyi Jin

Addresses: School of Electronic Information Engineering, Henan Polytechnic Institute, Nanyang, 473000, China ' School of Electronic Information Engineering, Henan Polytechnic Institute, Nanyang, 473000, China

Abstract: Monitoring and analysis of industrial internet of things (IIoT) data require high spatiotemporal correlation, heterogeneity, and periodicity. Traditional methods fail to effectively capture these properties. This study proposes an improved IIoT data monitoring model based on a graph convolutional network (GCN), integrating temporal convolutional networks (TCNs) and dilated convolutions to enhance spatiotemporal feature extraction. A federated learning mechanism is also incorporated to ensure data security. Experimental results show the model achieves an average accuracy of 0.88072, a coefficient of determination of 0.94431, and an explanatory variance score of 0.96651 - outperforming existing methods. Additionally, the FL-TGCN model attains a lower average RMSE of 1.76 compared to the long short-term memory model in time series optimisation. These findings confirm that the proposed model exhibits strong robustness, effective component integration, and stable generalisation in multi-sensor IIoT environments, indicating significant application potential.

Keywords: graph convolutional network; GCN; industrial internet of things; IIoT; time series; dilated convolution; federated learning.

DOI: 10.1504/IJSCC.2026.152968

International Journal of Systems, Control and Communications, 2026 Vol.17 No.2, pp.197 - 215

Received: 10 Apr 2025
Accepted: 12 Jun 2025
Published online: 17 Apr 2026 *

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