Propagating Uncertainty in Solar Panel Performance for Life Cycle Modeling in Early Stage Design

One of the challenges in accurately applying metrics for lifecycle assessment lies in accounting for both irreducible andinherent uncertainties in how a design will perform underreal world conditions. This paper presents a preliminarystudy that compares two strategies, one simulation-basedand one set-based, for propagating uncertainty in a system.These strategies for uncertainty propagation are thenaggregated. This work is conducted in the context of anamorphous photovoltaic (PV) panel, using data gatheredfrom the National Solar Radiation Database, as well asrealistic data collected from an experimental hardware setupspecifically for this study. Results show that the influence ofvarious sources of uncertainty can vary widely, and inparticular that solar radiation intensity is a more significantsource of uncertainty than the efficiency of a PV panel. Thiswork also shows both set-based and simulation-basedapproaches have limitations and must be appliedthoughtfully to prevent unrealistic results. Finally, it wasfound that aggregation of the two uncertainty propagationmethods provided faster results than either method alone.