Analysing Photovoltaic Grid-Connected H5 Inverter with Artificial Intelligence for Induction Motor Load Testing and Evaluation

Abstract

Recent advancements in transformerless photovoltaic (PV) grid-connected inverters have positioned them as a prominent technology for distributed PV power generation systems. This is attributed to their demonstrably superior efficiency, reduced cost, and compact size. Conversely, they are susceptible to common-mode currents, leading to significant electromagnetic interference and security risks, thereby compromising system reliability. We introduced an enhanced H5 variant topology, termed H5-D topology, which integrates a clamping diode to mitigate common-mode voltage fluctuations. The paper presents the simulation results comparing both topology’s configurations, with a focus on the H5-D topology’s efficiency in suppressing common-mode currents. Additionally, experimental prototypes were constructed and evaluated for both the original and the H5-D configurations topology. The results unequivocally demonstrate the superior performance of the H5-D topology. The experimental findings corroborate the effectiveness of the recommended/suggested topology in addressing current issues related to common-mode issues, thereby enhancing the reliability and performance of photovoltaic inverter systems.