Blockchain and AI-Based Natural Gas Industrial IoT System: Architecture and Design Issues

• Published in: IEEE network Vol. 34; no. 5; pp. 84 - 90
• Main Authors: Miao, Yiming, Zhou, Ming, Ghoneim, Ahmed
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.09.2020; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Blockchain; Computer architecture; Cryptography; Energy industry; Industrial applications; Industrial facilities; Internet of Things; Load modeling; Natural gas; Natural gas industry; Prediction models; Predictive models; Smart contracts
• ISSN: 0890-8044; 1558-156X
• DOI: 10.1109/MNET.011.1900532

With the development of energy interconnection and increasing consumption of natural gas, the distributed energy supply and trusted-transaction- based natural gas Industrial Internet of Things (IIoT) has become one of the hottest research spots in the energy industry. This article puts forward natural gas IIoT architecture based on blockchain and AI, in order to address the defects of centralized energy supply architecture. The proposed architecture is introduced in detail from three aspects of infrastructure, side-chain of natural gas block based on data dimension, and backbone of natural gas block based on value dimension. Then the design issues on the integration existence of blockchain and AI in an actual natural gas IIoT scenario are discussed, including data association and interaction of blockchains, trusted identity authentication and management, energy source and virtual currency transformation, and so on. Next, an LSTM-based natural gas load prediction model is proposed by virtue of AI technology. The transaction model based on natural gas value and supply-demand relationship is also proposed by choosing the common natural gas application scenarios and taking advantage of blockchain technology. Experiments show that the proposed models can predict demand and output load of natural gas, while achieving the balance of interests of natural gas suppliers, users, and market in the transaction scenarios.