Аңдатпа
This paper presents the design and implementation of an integrated AI-based auxiliary device to enhance power quality and increase output capacity in renewable power plants. Specifically, a deep learning (DL) algorithm is developed to stabilize the DC-link voltage of a multifunction inverter operating within a solar and wind-based microgrid system. The proposed controller dynamically adapts to fluctuations in renewable power generation and varying load demands, ensuring stable system operation and improved power quality under diverse and unpredictable conditions. MATLAB/Simulink simulations were performed to validate system performance. Results show that the controller reduced the Total Harmonic Distortion (THD) of the grid current from 28.13% to 2.3%, meeting IEEE 519 standards. The controller also maintained DC-link voltage stability and managed fluctuating PV power outputs (ranging from 0 to 8000 W) and battery charging levels (0.1 to 1.1 units) under dynamic scenarios. These outcomes demonstrate the superior effectiveness of the proposed AI-based controller compared to conventional approaches, particularly in scenarios characterized by uneven renewable energy generation, sudden load changes, and power quality disturbances. The findings highlight the potential of deep learning techniques in addressing critical challenges in modern renewable energy systems, paving the way for more reliable and efficient microgrid operations.