The sensitivity of power system expansion models

Abstract

Power system models are a widely used tool in academia. They support the societal and political decision-making process in the context of the transformation of power systems. In particular, they can propose approaches to ease the integration of renewable resources. However, model results are subject to uncertainties originating from uncertain model parameters or model formulations. Here, we show that the error caused by uncertain assumptions for model parameters or model designs can be measured by a mathematically well-defined concept: a distance measure (or metric). When applied in a systematic manner, this metric can be used to describe the sensitivity of many kinds of optimization models. Therefore, it is found that power system models are especially sensitive to the temporal resolution. We emphasize that our approach could be used as a standard to disclose sensitivity information in power system studies and, hence, contribute to openness and transparency.