ABSTRACT
Solar energy is a clean source of energy that falls under the category of renewable and sustainable energy and is widely accessible around the globe. In solar power systems, solar tracking systems are critical for optimizing energy output from the sun. Single and dual-axis solar trackers have traditionally been used to move solar panels in different directions based on the sun’s beams in order to enhance energy. To optimize the energy gain, the deployed tracker must actively follow the sun’s rays and adjust its location appropriately. The key components needed for constructing tracking systems are sensors, microcontroller-controlled control circuits, and servomotors with supports and mountings. Two servo motors are used to adjust the position of the solar panel so that the sun’s beam remains aligned with the solar panel. The suggested solar tracking system, which is based on machine learning, allows the solar panel to spin in any direction. The proposed machine 140learning-based classification technique examines and categorizes sensor outputs as defective or not. If the sensor results are awarded the class label “Non-faulty,” the servo motor adjusts the position and direction of the solar panel depending on the input sensor data. If the sensor readings are classified as “Faulty,” the servo motor will modify the position and direction of the solar panel using a regression-based machine learning technique.
