AN APPROACH FOR FAILURE DETECTION AND DIAGNOSIS IN PHOTOVOLTAIC POWER SYSTEMS BASED ON WEIBULL DISTRIBUTION FUNCTION

Abstract

This paper proposes a method to predict photovoltaic (PV) energy production based on the hypothesis that during the hours of activity of the plant, its production behaves as a stochastic process with a Weibull distribution function of two parameters which take into account the characteristics of the photovoltaic plant and the weather parameters measured in real time by a sensor network. The versatility and accuracy of the model has been validated through the deployment of a real-time SCADA system. The system developed includes power generation prediction and real-time comparison between the power expected and the power measured, for failure detection. With the implementation of the proposed system was achieved a production improvement of more than 7% with respect to the periods with failures. Keywords: Information Systems, photovoltaic; energy; prediction; failure, Weibull distribution