Solar photovoltaic (PV) systems have become an integral part of today's advanced energy infrastructure due to its low kinetic energy, its abundance availability, and its freedom from human interference. Solar PV systems have the potential to greatly reduce our reliance on fossil fuels, but their intermittent nature means they cannot provide a constant source of electricity. The system's security should be well thought out, and it should be able to withstand a lot of abuse. The current energy system faces a significant difficulty in ensuring continuous supply. In this study, a three-phase, two-stage photovoltaic system that is managed by artificial neural networks (ANN). A DC-DC boost converter with maximum power point tracking (MPPT) based on the incremental conductance (INC) method is incorporated in the first stage. In the next step, an ANN-based controller optimizes the performance of a three-phase switching PWM inverter that is connected to the grid by controlling currents along the d-q axis. Comprehensive simulations were carried out using MATLAB or Simulink to evaluate the system's performance under various illumination and temperature conditions. Results show that the suggested approach outperforms the baseline in a number of areas. Better dynamic reactions, accurate tracking of reference currents within permissible bounds, and quick settling periods after startup are all displayed by it. These findings show that our method has the potential to greatly improve the efficiency and dependability of solar PV systems. The results of this study have implications for renewable energy in general and present a viable path toward enhancing the resilience and sustainability of energy infrastructure.
Read full abstract