Prediction of Solar Photovoltaic Conversion Efficiency: A Model Combined with Machine Learning
The photovoltaic conversion efficiency of photovoltaic materials is essential in solar power generation. This paper briefly introduced organic photovoltaic materials, proposed to predict the photovoltaic conversion efficiency of organic photovoltaic materials with a Convolutional Neural Network (CNN) algorithm, a machine learning algorithm, to reduce the trial-and-error cost in the development of organic photovoltaic materials, and compared the prediction performance of three algorithms, support vector machine (SVM), back-propagation neural network (BPNN), and CNN algorithms, using MATLAB software. The results showed that the prediction model of photovoltaic conversion efficiency constructed by the CNN algorithm had the largest goodness-of-fit, had a smaller error in predicting the photovoltaic conversion efficiency of organic photovoltaic materials, and took the least time.
Keywords: photovoltaic power generation, organic photovoltaic materials, photovoltaic conversion efficiency, convolutional neural network