Search

The performance assessment of photovoltaic (PV) systems is a complex process. Several meteorological data sources are available to evaluate the PV system generation. Different computing models can be applied to determinate the solar irradiance on the plane of the array (POA). The cooling effect of the PV module due to the wind speed should not be neglected. The present study may support several users to perform more accurate PV energy predictions, providing important suggestions to develop future PV system projects with more high reliability. Perez and Hay-Davies models for the computing of the irradiance on tilted surfaces are combined with three meteorological datasets, characterized by different monitoring period and meteo station location, to estimate the POA irradiance, the module temperature and PV energy output for a PV system located in the Mediterranean climate area. Prediction results are performed by the PVsyst tool and compared with the actual data. Simulations are carried out taking into account the wind effects on the PV module performance. Results demonstrate that the geographic features of the location, in which the weather station is located, have higher impact on the estimations of the PV system performance than the distance between the PV system and the meteo station. Perez and Hay-Davies models provide predictions of the PV energy and the module temperature with a difference up to 3% and 1% respectively. Yearly average wind speed in the range 2–4 m/s fosters a cooling effect up to 3% higher than one due to the wind magnitude less of 2 m/s, increasing the PV energy up to 1%.

This paper discusses the performance of the temperature perturbation-type ADMS-Temperature and Humidity Model (ADMS-TH) and the Computational Fluid Dynamics (CFD)-based model ENVI-met for the prediction of urban air temperature using measurements collected in the city of Lecce (IT) in summer 2012. The goal is to identify the most important factors influencing numerical predictions. Direct comparisons with measured data and statistical indices show that modelled results are within the range of acceptance. Daily trends are well captured although an underestimation of maximum temperature is observed. In ADMS-TH this is due to an underestimation of sensible heat fluxes during daytime, while in ENVI-met it can be attributed to an underestimation of turbulent momentum and thermal diffusivity. Overall, ADMS-TH did predict the temperature cycle with higher accuracy than ENVI-met and its performance was particularly good during the night. ENVI-met required an ad-hoc tuning of surface boundary conditions to predict nocturnal cooling, satisfactorily. ©

The urban heat island (UHI) phenomenon may produce several cascade effects on citizens’ health, energy consumption and air quality. Numerical modelling is recognised to be a powerful tool for the analysis of the UHI, although the question of which model to use (as implied in the ‘fit-for-purpose’ approach) much depends on the application and on the result of satisfactory validation against field measurements. In this paper, two different modelling approaches are applied, namely the integral-semi-Gaussian model ADMS-TH and the CFD-based model ENVI-met, to assess the UHI phenomenon in a city of south Italy (Lecce). Modelling results are validated against field measurements collected during summer 2012. The results suggest that the integral model has the ability of capturing the UHI cycle at city scale, while CFD modelling did not provide any substantial improvements in terms of local geometric effects on temperature distribution.