Track: High School and Middle School STEM Competition
In this paper, although photovoltaic power generation has recently shown the most remarkable growth in the field of renewable energy worldwide, defects may occur if power outages or manufacturing facilities remain despite the increase in demand and demand for photovoltaic power generation. A machine learning algorithm that predicts the optimal development to solve these problems was obtained through experiments. By implementing the algorithm in the system, it will be able to contribute to reducing operating costs and popularizing it.
Neural network, SVM, and deep learning are used as prediction algorithms, and the optimal algorithm is selected by using the root mean square error (RMSE), which is the most used when identifying prediction errors. We propose a predictive model whose prediction rate is expanded by changing the algorithm structure and modifying constants. Then, a defect detection system is developed by applying the predicted results to the domestic regional data.
Artificial Intelligence, Solar Power, Generation Prediction System, Solar Power Generation, AI