Irradiance Acquisition in Real Time With Long Term Data Logger and Postprocessing Using Data Mining Methods

Abstract

Nowadays, there has been an increase in the usage of renewable energies, due to the environmental situation in our planet. Among the existing renewable energies, the photovoltaic energy is an attractive alternative. However, it has an intermittent production due to the environmental factors involved in its process. Particularly partial shading can make photovoltaic arrays consume energy instead of producing it. Also, it is useful to register the environmental factors, in order to implement control measurements. For this reason, in order to measure and register real-time temperature and irradiance precisely, a low-cost Data Acquisition System (DAS) is proposed. The implemented DAS has a wireless communication system, which allows the user to have a database through an interface. Subsequently the available power is calculated through a simplified math model, and used along with the irradiance and temperature as attributes. Finally, the data is classified through Data Mining techniques, such as k-means and ID3, and later used to predict whether the bypass diode remains connected or not.