Contribution au Développement de Techniques de Recherche de la Topologie Optimale d’un Générateur Photovoltaïque.

Abstract

This thesis proposes a generalized technique to minimize power losses of PV arrays connected in Total Cross-Tied (TCT), under both current and voltage mismatch effects. The proposed method is based on the classification of the electrical data of the PV modules composing the photovoltaic array in order to identify the mismatch type, then applying an arrangement of the PV modules according to the mismatch type found. The design process of the proposed algorithm is detailed and its validity and performance are verified under different mismatch scenarios. The efficiency enhancement is verified for different mismaths cases and the computed results reveal that the proposed algorithm can achieve a great improvement in the PV array power. Furthermore, a comparative study with SuDoKu and genetic algorithms (GA) are performed. The obtained results under highlighted the superiority of the proposed method in comparison to the compared ones. The enhancement resides in the implementation simplicity as well as in the minimization of the number of infection points indicating smooth I-V and P-V characteristic curves.