Prediction of Global Horizontal Irradiance Using an Explainable Data Driven Machine Learning Algorithms

Abstract

Estimating global horizontal irradiance (GHI) with a high level of accuracy and precision is very challenging due to the volatile climate parameters and location constraints. To overcome this challenge, several machine learning (ML)-based techniques such as Decision Trees (DT), Random Forest (RF), Extreme Gradient Boosting (XGB), and Extra Trees (ET) are implemented to forecast the GHI. The first stage of model development is to select the optimal subset of features by using the variance inflation factor feature selection method. In the second stage, the selected features are fed into the ML models and trained. The predictive performance of the ML models is improved the result of removal of insignificant input features. The predictive accuracy of the ML models is compared and evaluated by performance metrics such as mean absolute error (MAE), root mean square error (RMSE), and coefficient of determination (R²). Conclusively, after feature selection it is seen that the ET algorithm outperforms the others because of its lowest MAE and RMSE value of 3.01 and 1.748, respectively, as compared to the other models, indicating its relevancy, legitimacy, and viability for the estimation of GHI. The higher R² value of 0.99 obtained by the ET model indicates that it is best fitted with the dataset. Additionally, optimal shapely additive explanation values have been used as feature attributions for determining the magnitude and direction of the impact of each feature on the outcome.

Keywords:

• global horizontal irradiance
• extra trees regressor
• shapely additive explanation
• variance inflation factor
• estimation

DISCLOSURE STATEMENT

The authors assert that they do not have known competing financial interests or personal relationships that may have appeared to impact the findings presented in this article.

DATA AVAILABILITY

Upon reasonable request, the data will be made available.

Additional information

Notes on contributors

Rahul Gupta

Rahul Gupta has 12 years of research and academic experience and is pursuing his Ph.D. at Netaji Subhas University of Technology, New Delhi. He is GATE-qualified and serves as a Teaching-cum-Research Fellow (TRF) in the Department of Electrical Engineering at Netaji Subhas University of Technology. He completed his B.Tech. in Electrical and Electronics Engineering from Abdul Kalam Technical University, Lucknow and M.Tech. in Electrical Engineering from Uttarakhand Technical University, Dehradun in 2016. His research interests are in renewable energy, machine learning, and deep learning.

Anil Kumar Yadav

Anil Kumar Yadav received the Ph.D. degree in Instrumentation and Control Engineering from the University of Delhi, Delhi, India in 2017. He is currently working as an Assistant Professor in the Department of Instrumentation and Control Engineering, Dr. B. R. Ambedkar NIT Jalandhar, Jalandhar (Punjab), India. He has 14 years of teaching and research experience and published more than 80 research papers in Journals and Conferences of repute. His research interests include renewable energy, hybrid systems, electric vehicle, and nonlinear and intelligent control.

Shyama Kant Jha

Shyama Kant Jha received B. Sc. Engineering degree in Electrical Engineering from Bhagalpur College of Engineering, Bhagalpur with Distinction marks and M.E. degree in Electrical with specialization in Control & Instrumentation from Delhi College of Engineering (DCE) [Presently Delhi Technological University (DTU)], Delhi University. He was awarded Ph. D degree from University of Delhi, New Delhi. Currently he is Professor in the Department of Instrumentation & Control Engineering at Netaji Subhas University of Technology (NSUT) [formerly Netaji Subhas Institute of Technology (NSIT)/Delhi Institute of Technology (DIT)], New Delhi. His teaching and research interests include optimal control, robust control, sustainable energy, bio-inspired control, and electric drives etc. He has published/presented good number of papers in international and national journal, conferences.