Home // International Journal On Advances in Intelligent Systems, volume 16, numbers 1 and 2, 2023 // View article
Authors:
Bernard Lee
Anthony Constantinides
Keywords: Graph Neural Network; AI Process Control; Green Energy; Solar-Thermal Generation; Securitization; Tokenization
Abstract:
HedgeSPA is launching a pilot project for green electricity and hydrogen production in collaboration with a consortium including the Fraunhofer Institute for Reliability and Microintegration (IZM), part of Germany's leading applied R&D institute Fraunhofer Institute. Using IZM's high-temperature and high-pressure sensors designed for the German chemical industry, HedgeSPA will deploy its graph-theoretic AI engine on some of the most advanced hardware supplied (with validation) by Amazon Web Services (AWS) and Nvidia to control the green electricity and hydrogen production processes from the cloud using solar-thermal generation, which can continue generating electricity for up to 12 hours after sunset. While solar-thermal technologies are proven, past deployments required large sites to achieve economies of scale from centralized human support and data centers. Large sites are more difficult to find and as such their regulatory approval processes can be arduous. For the economics of smaller sites to work, they need to minimize or even eliminate human support and data centers from full automation with an AI engine that can adapt to different topographies and respond to weather and environmental challenges using reinforcement learning. In the future, we may further upgrade electricity generation from turbine engines to thermophotovoltaic (TPV) cells, while using the residual heat to superheat water molecules for high-temperature electrolysis, or to Stirling engines (used in conventional-powered submarines built by countries such as Japan and Singapore) with fewer moving parts and no dependency on a water source to produce steam. Today's market has plenty of ready buyers of any green electricity, green hydrogen, or synthetic natural gas produced. The key challenge appears to be financing for the supply side, which can be addressed by securitizing buyers' cash flows to pay for any upfront investments, similar to how homeowners can pay for mortgages instead of rents when an efficient financing mechanism is available to finance the upfront investments for construction companies to build residential homes. We propose to facilitate the financing with a classic waterfall structure under a tokenization mechanism to issue back-to-back tokens against the asset-back securities with high-integrity carbon tokens attached. All of the above fits with a widely-adopted international legal framework for renewable power plants, with only a few new elements that appear manageable based on qualified legal opinions.
Pages: 1 to 18
Copyright: Copyright (c) to authors, 2023. Used with permission.
Publication date: June 30, 2023
Published in: journal
ISSN: 1942-2679