ABSTRACT
Accurate forecasting of the photovoltaic (PV) power output is crucial for the stability of power systems. In this paper, the Johansen vector error correction model (VECM) cointegration approach is used to predict the PV power output of a PV system in the tropical island of Mauritius. The predicted PV power output is compared to measured PV power output under real conditions. Only solar irradiation and PV cell temperature are considered as predictors and data is recorded from a 20 kW PV system installed in the Phoenix region in Mauritius. The performance of Johansen model is then compared to a conventional ANN forecasting model. Results showed that the Johansen VECM cointegration approach outperformed the Artificial Neural Network (ANN) model with RMSE and MAE values of 298.9W and 221.6W, respectively. The Johansen model proved to be a powerful tool, showing good potential for PV power forecasting.
Acknowledgements
This work has been conducted under the HEC (Higher Education Commission Mauritius) MPhil/PhD scholarship.
Disclosure statement
No potential conflict of interest was reported by the author(s).