Hydrogen Blending in Iberian Gas-Electricity Networks using Mixed Integer Linear Programming

Abstract

This paper develops an hourly optimization model coupling hydrogen (H2) and natural-gas (NG) transport with electricity dispatch through combined-cycle gas turbines (CCGT). Mixed gas nonlinearities are approximated using McCormick relaxations and a tangent-based pressure-flow approximation. An iterative concentration reconstruction and bound-refinement procedure reduces the mixing inconsistencies, while fixed flow directions and relaxed thermal commitments make the weekly management tractable. A 2030 Iberian case study compares hydrogen mass-fraction limits from zero to 2.8% during a representative week. H2 utilization ranges from nearly 0% to 85.7%, showing that H2 injection potential strongly depends on project location, downstream gas demand, available NG dilution and network flow and pressure constraints. Increasing the maximum H2 mass fraction from 0.7% to 2.8% raises avoided direct CO2 emissions from 4.08 to 17.88 kt. The impact on CCGT generation is comparatively small, while gas composition, flows and pressures are more significantly affected.

This paper develops an hourly optimization model coupling hydrogen (H2) and natural-gas (NG) transport with electricity dispatch through combined-cycle gas turbines (CCGT). Mixed gas nonlinearities are approximated using McCormick relaxations and a tangent-based pressure-flow approximation. An iterative concentration reconstruction and bound-refinement procedure reduces the mixing inconsistencies, while fixed flow directions and relaxed thermal commitments make the weekly management tractable. A 2030 Iberian case study compares hydrogen mass-fraction limits from zero to 2.8% during a representative week. H2 utilization ranges from nearly 0% to 85.7%, showing that H2 injection potential strongly depends on project location, downstream gas demand, available NG dilution and network flow and pressure constraints. Increasing the maximum H2 mass fraction from 0.7% to 2.8% raises avoided direct CO2 emissions from 4.08 to 17.88 kt. The impact on CCGT generation is comparatively small, while gas composition, flows and pressures are more significantly affected.