Sector coupling between green Hydrogen and steel: machine learning-based estimation of levelized cost of steel

As per the Paris Agreement of the 15th Conference of the Parties (COP 15), nations pledged to set their national decarbonization and Net Zero goals or Nationally Determined Contributions (NDCs) as part of a larger shared goal of preventing the rise of global average temperatures to more than 1.5°C compared to the pre-industrial average as per the recommendations of The Intergovernmental Panel on Climate Change (IPCC), to avoid the possibilities of large scale irreversible climate catastrophes emerging from the effects of global warming. Accordingly, India has also set its ambitions to go Net Zero by 2070. Also, as per India’s NDC, targets have been set for various carbon-intensive industries like Oil & Gas, Power, Cement, Transport, Steel, etc. Amidst these pledges and targets that span across sectors, green hydrogen is emerging as a key technology with a promise to decarbonize several of these end-uses, as a chemical feedstock in some, while as an energy source in others. However, not all end-uses are equal in their technological readiness or maturity to adopt green hydrogen economically and at scale. In this paper, the possibility of sector coupling between green hydrogen and steel is studied and quantified at the most granular detail (site-level), as steel is one of the most carbon-intensive and critical sectors and must be abated if a nation is to fulfill its NDCs. Steel facility level optimization results available in the public domain are referenced, and an in-house Machine Learning (ML) Modeling framework is developed to estimate the Levelized Cost of Steel (LCOS) based on the Renewable Energy (RE) characteristics of specific geolocations. The correlation between the modeled LCOS and the Levelized Cost of Hydrogen (LCOH) at those geolocations is computed, whereby the presence of a moderately positive correlation between the two commodities would establish grounds for sector coupling.