Empirical Model for Capacity Credit Evaluation of Utility-Scale PV Plant

Abstract

Owing to increasing photovoltaic (PV) systems installation, especially large-scale ground-mounted PV plants, the role of PV power generation is gradually changing from a supplementary energy to an alternative energy resource, with respect to conventional power generation. It is important and necessary to evaluate the capacity value of PV systems. This paper introduces an artificial neural network-based empirical model to evaluate a PV plant's capacity credit, in which the effective load carrying capability is utilized to calculate the capacity value of utility-scale PV plants. A novel metric is proposed to depict the temporal correlation between PV output and load profile, considering the intermittent nature of PV power. The impacts of PV penetration, simulation temporal granularities, and correlation similarity of varying PV and load time series on the capacity credit evaluation of PV systems are also explored. Finally, the presented empirical model is employed for capacity credit evaluation for any given conditions. The law of decreasing marginal capacity value is verified, and the optimal time scale for capacity credit estimation is obtained using the established model.