Open circuit fault diagnosis for a five-level neutral point clamped inverter in a grid-connected photovoltaic system with hybrid energy storage system

Résumé fourni par la source

Introduction. Recently, the number of high and medium voltage applications has increased dramatically. The connection between these different applications requires series-parallel combinations of power semiconductors. Multilevel converter topologies provide major advantages to these applications. In this paper, a grid-connected photovoltaic system with a hybrid energy storage system using a five-level neutral point clamped inverter is studied. Although the multilevel inverter has many advantages over the two-level inverter, it has a high probability of experiencing an open circuit fault. In this context, the five-level inverter has 24 controllable switches, one of which may experience an open circuit fault at any time. Therefore, it plays an important part in the reliability and robustness of the whole system. The novelty of this paper presents an approach to accurately detect the open circuit fault in all insulated gate bipolar transistors of a five-level neutral point clamped inverter in a photovoltaic power generation application with a hybrid energy storage system. Purpose. Before using fault-tolerant control to ensure service continuity, fault diagnosis techniques must first be used, which are the crucial phase of reliability. Methods. A detection method based on the maximum and minimum error values is proposed. These errors are calculated using the expected and measured line-to-line pole voltages. Results. The open circuit fault detection method is implemented using MATLAB/Simulink. Simulation results showed the accuracy of detecting the open circuit fault in all insulated gate bipolar transistors in a short time. Moreover, this method is adaptable to several applications and is also robust to transient regimes imposed by solar irradiation and load variations.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.