Towards the applicability of solar nowcasting: A practice on predictive PV power ramp-rate control

Abstract

Solar forecasting has been widely adopted in modern power system operations to facilitate a reliable and continuous photovoltaic (PV) integration. Solar nowcasting, also known as intra-minute solar forecasting, is a new subdomain of solar forecasting. Nevertheless, despite the significant progress achieved in solar nowcasting over the last decade, one important aspect, that is, applicability—the value and operability of nowcasts in practical grid operations—is generally left out. To that end, this paper brings forth the applicability of solar nowcasting for the first time. Three time parameters involved in operational solar nowcasting are first identified, namely, forecast horizon, forecast resolution, and forecast model updating rate. Then paired with the state-of-the-art PV power ramp-rate control algorithm, i.e., predictive active power curtailment (PAPC), the effect of different time parameters is investigated, thus revealing the nowcasting applicability at large. Through four case studies and eight standardized deterministic and probabilistic solar nowcasting models, the applicability of solar nowcasting on PAPC is shown to be most characterized by the forecast horizon (up to a deviation of ramp smoothing rate around 12%, with smart persistence (SP) being the reference model), and least characterized by the forecast model updating rate (with a deviation of ramp smoothing rate less than 1% for SP). Moreover, the negatively-biased deterministic nowcasts and wider probabilistic nowcasts are found more applicable to PAPC. To promote solar nowcasting applicability on PAPC further, an outlook for future research is provided, from both a solar forecaster's and a system operator's viewpoints.