Analysis of short-term operational forecast deviations and controllability of utility-scale photovoltaic plants

Abstract

The growing deployment of utility-scale photovoltaic (PV) in electricity industries raises many concerns for power system operation. Due to the highly variable and somewhat unpredictable characteristics of short-term PV output, the system operator faces challenges in balancing electricity supply and demand, which increases the need for short-term frequency management. Although many studies have been carried out to enhance the understanding of short-term PV output characteristics, the uncertainty around forecast targets for electricity market dispatch has not been studied. This paper analyses 4-s SCADA output data from PV plants and all other generators operating in the Australian National Electricity Market (NEM) to assess the deviation from their forecast dispatch targets. During the study period, PV penetration was less than 5% of the total generation capacity. The deviations of PV plants from forecast targets increases almost proportionally with their operating level. The correlation between deviations from the targets and system frequency deviations shows that PV plants contribute towards increasing frequency deviations, yet the impact seems minor, with only 0.06–0.09 correlation coefficients. Despite their variability during normal operation, PV plants demonstrate accurate output control when following dispatch instructions during network constrained periods with fluctuations of only ±1% of the plant rated capacity.