Comparative life cycle assessment of two different SOFC-based cogeneration systems with thermal energy storage integrated into a single-family house nanogrid

This paper presents the life cycle assessment (LCA) of two different solid oxide fuel cell (SOFC)-based combined heat and power (CHP) systems with thermal energy storage, one based on a reversible solid oxide fuel cell (r-SOFC) and hydrogen storage, and the other on a natural gas-fed SOFC. Both systems have been included alternately into the same single-family house nanogrid, that also includes a photovoltaic (PV) system for renewable energy production. The hourly electrical and thermal demands over an entire year of the user are calculated by means of the commercial software TRNSYS 17, the simulation of the two CHP systems operation is performed using the commercial software HOMER PRO, and the life cycle assessment is performed by means of the software SimaPro 9 and Ecoinvent 3.3.The main finding of this study is that, in the considered Italian scenario, the r-SOFC-based CHP system has, in general, a more adverse environmental impact than the natural gas-fed one. This is essentially because the higher electricity production of the natural gas-fed SOFC, mainly due to its continuous operation, determines a higher self-consumption referred to the total alternating current (AC) load on a yearly basis. On the other hand, the r-SOFC with H₂ storage shows a more adverse environmental impact, despite the higher power self-consumption from the PV system. Indeed, relatively to the r-SOFC system, in the cold seasons large part of the electrical energy necessary to power the H₂ production and AC loads is taken from the Italian national grid, which has a relatively high share of electricity produced by fossil fuel thermal power plant. Moreover, the effects of an increased renewable energy (solar) share in the country energy mix are also reported and discussed, and the results of a Monte Carlo analysis aimed at evaluating the effects of the uncertainty relative to the input parameters for the LCA analysis are presented as well.