Application of autoregressive model with exogenous inputs to identify and analyse patterns of solar global radiation and ambient temperature

摘要

This work presents the results and analysis of the application of autoregressive model with exogenous inputs (ARX) models on radiation and temperature patterns obtained at a high-tech monitoring station in Bogota, Colombia. The station was set up using virtual instruments to measure and analyse data in order to evaluate the solar potential available to design photovoltaic (pv) systems. The monitoring station was developed using an SCXI system as hardware, the data acquisition card 6024 E from National Instruments and the LabVIEW package as software. The data allow us to identify parameters for the models using Mean squared error (MSE) technique with an algorithm implemented in the MATLAB software. After applying the model for a seven-year period, the obtained results showed that the correlation between the irradiance measurement obtained by the monitoring station and the pattern evaluated by the algorithm was 84.21%, whereas the ambient temperature was 67.76% during the same period of time.