مقاله انگلیسی رایگان در مورد مدل سازی برق فتوولتائیک خورشیدی با کمک سیستم پیش بینی پارامترهای آب و هوا – تیلور و فرانسیس 2022

Abstract

Introduction

Methodology

Data collecti

Performance of ANN forecaster

Performance of hybrid-ANN forecaster

Performance of hybrid-ANN forecaster-S

Conclusion

Nomenclature

Acknowledgments

Disclosure statement

References

Abstract

     The integration of solar photovoltaic (SPV) system to the grid has introduced a new source of intermittency in the grid, and the grid has to react smartly to the changes that occur in the penetration of SPV power. Accurate modeling of weather-dependent SPV power will be helpful in forecasting the penetration of SPV power into the grid. An SPV power output forecasting model has been developed based on artificial neural network (ANN) approach. Two forecasters, namely ANN forecaster and two-stage hybrid-ANN forecaster, are developed with operational and weather parameters. The historical data of SPV power (P), hours of operation of SPV system (to), daily global solar radiation (H), and ambient temperature(T) are used as modeling parameters. The combination of modeling parameters {P, H, T, to} is identified as the best combination that influences the forecasting of day-ahead power output. A relative root mean square error (RRMSE) of 5.74% was obtained with the combination of {P, H, T, to}. An RRMSE of 6.04% was observed with the combination of {P, H, T} as inputs, and the hours of operation of the SPV plant could be ignored in the model. The historical power data of the SPV plant is identified as the crucial parameter in the SPV power forecast model and has given an RRMSE of 7.25%.

Introduction

In recent times, the consumption of energy from conventional sources has been increased, which causes the rapid depletion of fossil fuel-based resources. The emission of greenhouse gases has increased because of the utilization of conventional sources over the past few decades (Fikru and Gautier 2015). In addition to that, the energy demand has been increasing enormously in both developed and developing countries due to the increase in population, urbanization, and industrialization. The portion of this energy demand can be met by renewable energy sources (RES) such as wind and sun. Solar energy can be captured by installing solar collectors, solar ponds, solar chimneys, and solar photovoltaics (SPVs). Out of all these installations, the SPV installation growth rate has increased enormously. The generation of electricity using SPV is widely accepted (Larson, Nonnenmacher, and Coimbra 2016). PV power generation is ecofriendly because it produces no toxic emission, of low maintenance, of noiseless operation, and it is abundantly available (Yan et al. 2021). Up to now, several thousands of SPV systems of 1 kW to several hundreds of megawatts have been installed and integrated into the grid worldwide (Tina, Scavo, and Gagliano 2020). The integration of SPV systems to the grid has introduced a new resource of intermittency in the grid (Sepasi et al. 2017). This is due to the sensitivity of the SPV system to local weather conditions and variation in solar radiation availability. The SPV power output is uncertain since the parameters like irradiance level, ambient temperature, surface temperature of the panel, dust deposits, wind speed, cloud cover, and relative humidity etc., will have an impact on the SPV power generation. Accurate modeling of SPV power is required by the utilities to control the high instabilities of the electric grid due to unpredictable PV power penetrations (Wan et al. 2015).

Conclusion
A day ahead SPV power forecast models are developed by using ANN approach. Daily GSR (Hm, Hp), ambient temperature, hours of operation of SPV system, and historical data of SPV power are used as input parameters in the modeling of the day ahead forecast models. The trial and error method has revealed that the ANN model with three neurons in the hidden layer with linear-log-sigmoid transfer function has shown the best performance.

     The effect of each parameter and the combination of parameters on the model accuracy is identified by developing various forecast models, and a total of 21 forecast models are developed. Out of the 21 developed forecast models, ANN forecaster-4 with the combination of inputs {P, H, T, to} has shown the best performance. The combination has shown a strong influence on the SPV power output. The MAPE of 4.18% and the RRMSE of 5.74% are derived with 4-3-1 ANN architecture. Historical data of measured power is the best suitable modeling parameter to predict the SPV power. Historical power measurements will not be available before the installation of the SPV system, and in such cases, ANN forecaster-2 with {Hm, T} could be used as a forecast model.