From conventional to surrogate-based flexibility assessment: application to a bio-methanol process under uncertainty

Over the past decade, a primary concern in the process industry has been the transformation of traditional chemical processes into more sustainable alternatives, particularly in response to the emission targets established by the EU for 2050. To address this challenge, various technologies leveraging cogeneration, biomasses, and energy integration have been recently developed. However, these approaches are usually associated to a reduction in the degrees of freedom and to an increased sensitivity of the process to potential uncertainties. In such scenarios, external disturbances could harm the profitability and sustainability of the system, and designing based solely on nominal operating conditions may not be sufficient to evaluate the actual plant performance. This work has the purpose of investigating the performances of a Biogas-to-Methanol process under perturbed conditions, with a focus on the total energy consumption and related equivalent carbon dioxide emissions. The methodology employed allows obtaining two significant outcomes. Firstly, it enables the prediction of specific emission variations for a general process tailored for a specific inlet biogas composition, when encountering different feedstocks. Secondly, it provides valuable hints on how to effectively deal with these variations or to guide decision making during the design phase. By applying the methodology to the specific case study discussed in this work, it has been verified that the majority of the expected perturbations can be withstood by the system with an equivalent carbon dioxide emissions increase lower than 10%. Furthermore, by means of a surrogate modelling procedure, it has been possible to derive an analytical function that allows quantifying the equivalent carbon dioxide emissions within a wide range of process conditions and with the best trade-off between computational efforts and accuracy. The proposed model can provide additional support in further applications such as process optimization or optimal scheduling.