Comparison of machine learning and deep learning algorithms for hourly global/diffuse solar radiation predictions

Abstract

Summary Due to the advancement and wide adoption/application of solar‐based technologies, the prediction of solar irradiance has attracted research attention in recent years. In this study, the predictive performance of machine learning models is compared with that of deep learning models for both global solar radiation (GSR) and diffuse solar radiation (DSR) prediction. Different studies have proposed the use of different models for solar radiation prediction. While some used machine learning models, the use of deep learning algorithms were considered by others. Although these algorithms were concluded to be appropriate for solar radiation prediction, variation in their performances brings about an intriguing quest to compare and determine the most appropriate algorithm. The three most common deep learning models in the literature namely; artificial neural network, convolutional neural network, and recurrent neural network (RNN) are considered within the scope of this study. Also, two traditional machine learning models namely polynomial regression and support vector regression (SVR) is considered as well as an ensemble machine learning model called random forest. These models have been applied to four different locations in Nigeria and the typical meteorological year data for 12 years in an hourly time step was used to train/test the model developed. Results from this study show that deep learning models have a better GSR and DSR prediction accuracy in comparison to machine learning models. However, the duration for training and testing the machine learning models (except SVR) is shorter than that of deep learning models making it more desirable for low computational applications. The application of RNN for GSR prediction in Yobe (with an r value of 0.9546 and root means square error/mean absolute error of 82.22 W/m2/36.52 W/m2) had the overall best model performance of all the models developed in this study. This study contributes to the existing literature in this field as it highlights the disparities between machine learning and deep learning algorithms application for solar radiation forecast.

Development and validation of a deep learning‐based model for predicting burnup nuclide density

“…Based on a comparison between the predictive model and the actual WHR simulated cycle, it can be concluded that data science can be used as an alternative to thermodynamic modelling to avoid time‐consuming calculations. Comparing machine learning and deep learning algorithms for estimating hourly global/diffuse solar radiation, Bamisile et al 8 found that all deep learning models produced in this work performed better than machine learning models. Cheng and Yu 9 discussed machine learning technologies applied to smart energy and electric power systems, beginning with an introduction to seven types of machine learning (ML) methods and systematically reviewing their applications in Smart energy and electric power system (EEPS), and concluding with a discussion of ML development under the big data thinking and a future development outlook for AI 2.0 and ML in Smart EEPS, etc.…”

Optimal demand response programs selection using CNN‐LSTM algorithm with big data analysis of load curves

“…Deep learning is a subfield of machine learning, and ANNs form the backbone of deep learning algorithms. In fact, the distinguishing feature of a single ANN from the deep learning algorithm is the number of node layers or the depth of the neural networks, which must be more than three in the deep learning algorithm [35]. In this article, the CNN-LSTM algorithm is used.…”

Comprehensive assessment, review, and comparison of AI models for solar irradiance prediction based on different time/estimation intervals

“…The expansion of solar energy-based technologies and applications will continue 40 . Therefore, the reliable estimation of solar radiation including its hourly, daily average, monthly average, annual, 41 and seasonal variability is of paramount importance for the estimation of solar energy capacity and potential 42 . As mentioned earlier, the high cost and technological complexity attached to the measurement of solar radiation makes it a more difficult task in many meteorological stations.…”