Design and experimental validation of an IoT data-logger device for indirect measurement of solar irradiance based on a correlation model

The present work aimed to develop an autonomous instrument for indirectly measuring solar irradiance from silicon sensors. Through laboratory tests, using a precision digital solar power meter device (SM206-SOLAR), with an error range of ±10% per reading and ±0.38% W/m
²
per °C, and photovoltaic trainer equipment (Lucas Nulle), a statistical correlation model was developed between the short-circuit current of the NPA5S-12H solar module and the measured artificial irradiance. A significant linear correlation was determined with a mean absolute percentage error (MAPE) of 6.97%, an average root-mean-square error (RMSE) of ±26 W/m
²
per °C and a relative percentage root mean square error (rRMSE) of 10.76%. For real-time data collection purposes, a data-logger device with wireless communication functions based on a Wi-Fi link and IoT MQTT protocol was developed. Finally, a SQL server and web application were configured on a public Linux server to collect and present data in real-time.