A New Control Strategy for Three Phase Grid Connected PV System Under Severe Partial Shading Conditions

Abstract

Photovoltaic energy is one of the most usable renewable energy on earth, however it is highly affected by several parameters such as irradiance and temperature which directly affect its efficiency, especially in the presence of PSC (partial shading condition) which generates multiple power peaks consisting of one high global maximum called GMPP (global maximum power point) and other lower local maximum called LMPP (local maximum power point). Under this condition, conventional techniques such as P&O (Perturb &Observe), IC (Incremental conductance) and nonlinear controllers fail to track the GMPP and are trapped in one of the LMPP which can be the lower peak, thus, provoking a significant drop in power. This paper presents a new control strategy for three phase grid connected PV system based on ANN (artificial neural network) for GMPPT (GMPP tracking) under PSC. The proposed approach was compared to the latest GMPPT algorithms. Results showed the good performance of the proposed controller in tracking speed with 0.0045s, high precision of 99.73% with negligible oscillations around GMPP of 0.01W. In addition, a classical controller was used to control the three phase inverter in order to inject a synchronized sinusoidal current to the grid with a low THD (total harmonic distortion) ratio of 1.23%. The overall system worked perfectly under severe partial shading conditions