A coupled WRF-PV mesoscale model simulating the near-surface climate of utility-scale photovoltaic plants

Abstract

To simulate the local or regional climate of utility-scale photovoltaic (PV) plants, a new PV-associated energy balance model was developed for the Weather Research and Forecasting Model (WRF). The resultant tool can be used to improve our understanding of the interaction between utility-scale PV plants and land-surface processes. The effects of PV panels were represented by transferring the solar radiation into electricity and heat flux and imposing an extra drag force on the mean flow in the model domain. The proposed PV model is more advanced than previous models in the following ways: (1) The model is universal based on its analytical expression of the sensible heat of PV panels and the influence of the physical shielding of the panels on the upward longwave radiation. (2) The photoelectric conversion efficiency (ε) in the model is temperature dependent and restrained between seasons. (3) The model includes the soil and latent heat flux parameterizations under observation constraints in a novel way. (4) The model bases the dynamic roughness length (Z0) on the eddy covariance flux observation. This new PV model was then incorporated with the WRF mesoscale meteorological model, and the model evaluation in two utility-scale PV plants demonstrated acceptable performance of the new WRF-PV mesoscale model in simulating the max/mean daily surface temperature, downward shortwave radiation, and wind speed with correlation coefficients of 0.61–0.99, and the hourly sensible/latent heat fluxes, with correlation coefficients of 0.59–0.93 during different seasons. The consistency of the performance of the model across these near-surface meteorological variables confirms the versatility and reliability of the new WRF-PV model as a suitable tool for simulating the climate of utility-scale PV plants.