Etude et Simulation d’un système photovoltaïque appliqué machine asynchrone

Abstract

Optimum operation and continuous industrial mechanisms can be contemplated without the presence of a system that prevents the state early anomalies that may arise in the various organs of equipment, and quickly diagnose faults. For this purpose, an application of artificial neural networks to detect defects in a photovoltaic system was developed. The work presented in this paper concerns the modeling, the diagnosis of a photovoltaic panel associated with a DC-DC converter controlled by MPPT, intended to lead to an induction motor through a DC-AC. A photovoltaic generator can operate over a wide range of voltage and current output, but it can deliver maximum power for the particular values of current and voltage. Indeed, the characteristic I (V) of the generator depends on the solar irradiance and temperature. These climatic variations cause fluctuations in the maximum power point. Because of this fluctuation is often interposed between the generator and the receiver one or more controlled static converters to continue the maximum power point. These commands known as MPPT (Maximum Power Point Tracking), are associated with DC-DC converter, which ensures the coupling between the PV generator and the asynchronous machine by forcing the first to deliver its maximum power and detect defects in the photovoltaic panel with ANN (Artificial Neural Networks).