<![CDATA[This research proposes and implements a Maximum Power Point Tracking (MPPT) control system on solar panels using a quadratic boost converter controlled by a backpropagation-based Artificial Neural Network (ANN) algorithm. The system is designed to maximize the output power of a 2x50 WP solar panel by reading current and voltage data, then calculating the power change (ΔP) which is used as JST training input to produce the optimal duty cycle value. This value is then used to adjust the PWM signal that controls the operation of the converter. The testing was conducted using real hardware connected to the Arduino Mega 2560, and programming was done through MATLAB Simulink. The JST training results show a very low Mean Squared Error (MSE) and high prediction accuracy with a regression coefficient (R) value approaching 1. The system has proven capable of reaching a maximum power point (MPP) of 52 watts in just 21 seconds with minimal power fluctuations. Thus, this JST-based MPPT control system demonstrates efficient, accurate, and responsive performance in optimizing the output power of solar panels]]>
Copyrights © 2025
