Day-ahead multi-energy optimisation considering hydrogen blending and integrated electricity-heat-gas networks

The use of renewable gases such as "green" hydrogen (H2) and synthetic natural gas (SNG) produced with renewable energies is a promising option to reduce carbon emissions and provide flexibility in emerging electricity-heat-gas multi-energy districts (MEDs). However, optimally operating MEDs while injecting renewable gases into the gas network requires overcoming current modelling limitations that can compromise their secure operation (e.g., for gas blend constraint management) and limit their flexibility. Existing approaches to model gas blend in multi-vector optimisation problems (e.g., applicable to MED modelling) sacrifice exactness as a means to reduce complexity, are based on specific contexts which limits their extendibility, and overlook critical infrastructures (i.e., heat networks) required to support energy decarbonisation. To address these gaps, this paper proposes a two-stage day ahead MED optimisation framework. The framework builds on an extendable algebraic linear approximation (ALA) algorithm to bring together a linear programming optimisation model of the MED's distributed energy resources, a full nonlinear integrated electricity-heat-gas network simulator, and a gas blend tracking model. The proposed framework is demonstrated with applications to a real MED in the UK where blended gas security indices (e.g., H2 concentration and Wobbe Index) and integrated network constraints are actively managed. The results demonstrate the potential of the proposed ALA algorithm to improve the computational efficiency of the framework. The findings show that comprehensive modelling of network constraints is crucial, even with modest integration of renewable gases, whereas higher hydrogen blending limits can enable a more cost-effective decarbonisation of the energy system.