Integrating Machine Learning and IoT Technologies for Advancements in Solar Energy Systems

This research focuses on the synergistic application of machine learning and Internet of Things technologies for the improvement of the performance of renewable energy systems based on the example of solar energy conversion. The research covers the aspects of a solar energy conversion system implementation, in which a set of solar panels, a multilevel inverter, and various sensors that monitor solar irradiance, temperature, and voltage parameters have been used. The sensors have an IoT feature of wireless data transfer to a central control unit. Machine learning algorithms, in particular, Artificial Neural Networks, Support Vector Machines, Decision Trees, and Random Forest, are used to analyze historical and real-time data to provide predictive maintenance, to forecast energy demand of the users, and to detect faults. The control framework of the system uses a Proportional-Integral-Derivative control loop to adjust the output voltage of the multilevel inverter through comparison with the required grid or load demand. The performance of the control shows significant improvements of the conversion efficiency of energy, the precision of voltage regulation, and the capacity for predictive maintenance. Particularly, it is defined that the accuracy of the prediction of the response to maintenance of the ANN algorithm is 96.7% that is the highest among the studied models. As a result, it is defined that renewable systems are dynamic, and adaptive control is essential for their effective functioning taking into consideration different environmental conditions. The results can be useful for the future design and operation of such systems, and it has the potential to create a sustainable and reliable energy supply by improving the efficiency of renewable energy systems.