A New Approach for Meteorological Variables Prediction at Kuala Lumpur, Malaysia, Using Artificial Neural Networks: Application for Sizing and Maintaining Photovoltaic Systems

Marwan Mahmoud's picture
Journal Title, Volume, Page: 
Journal of Solar Energy Engineering . 01/2012; 134(2):021005. DOI: 10.1115/1.4005754
Year of Publication: 
2012
Authors: 
Tamer Khatib
Department of Electrical, Electronic and System Engineering,Faculty of Engineering and Built Environment, University Kebangsaan Malaysia, Bangi 43600, Selangor, Malaysia
Azah Mohamed
Department of Electrical, Electronic and System Engineering,Faculty of Engineering and Built Environment, University Kebangsaan Malaysia, Bangi 43600, Selangor, Malaysia
M. Mahmoud
Department of Electrical Engineering, Faculty of Engineering and Information Technology, An-Najah National University, Nablus. Palestine
Current Affiliation: 
Department of Electrical Engineering, Faculty of Engineering and Information Technology, An-Najah National University, Nablus. Palestine
K. Sopian
Solar Energy Research Institute, University Kebangsaan Malaysia, Bangi 43600, Selangor, Malaysia
Preferred Abstract (Original): 

This research presents a new meteorological variables prediction approach for Malaysia using artificial neural networks. The developed model predicts four meteorological variables using sun shine ratio, day number, and location coordinates. These meteorological variables are solar energy, ambient temperature, wind speed, and relative humidity. However, three statistical values are used to evaluate the proposed model. These statistical values are mean absolute percentage error (MAPE), mean bias error (MBE), and root mean square error (RMSE). Based on results, the developed model predicts accurately the four meteorological variables. The MAPE, RMSE, and MBE in predicting solar radiation are 1.3%, 5.8 (1.8%), and 0.9 (0.3%), respectively, while the MAPE, RMSE, and MBE values for ambient temperature prediction are 1.3%, 0.4 (1.7%), and 0.1 (0.4%), respectively. In addition, the MAPE, RMSE, and MBE values in relative humidity prediction are 3.2%, 3.2, and 0.2. As for wind speed prediction, it is the worst in accuracy among the predicted variables because the MAPE, RMSE, and MBE values are 28.9%, 0.5 (31.3%), and 0.02 (1.25%). Such a developed model helps in sizing photovoltaic (PV) systems using solar energy and ambient temperature records. Moreover, wind speed and relative humidity records could be used in estimating dust concentration group which leads to dust deposition on a PV system.