E-Thesis 260 views 259 downloads
Forecasting and Prediction of Solar Energy Generation using Machine Learning Techniques / ALI LARI
Swansea University Author: ALI LARI
DOI (Published version): 10.23889/SUThesis.66094
Abstract
The growing demand for renewable energy sources, especially wind and solar power, has increased the requirement for precise forecasts in the energy production process. Using machine learning (ML)techniques offers a revolutionary way to deal with this problem, and this thesis uses machinelearning (ML...
| Published: |
Swansea University, Wales, UK
2024
|
|---|---|
| Institution: | Swansea University |
| Degree level: | Doctoral |
| Degree name: | Ph.D |
| Supervisor: | Egweb, A. |
| URI: | https://cronfa.swan.ac.uk/Record/cronfa66094 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Abstract: |
The growing demand for renewable energy sources, especially wind and solar power, has increased the requirement for precise forecasts in the energy production process. Using machine learning (ML)techniques offers a revolutionary way to deal with this problem, and this thesis uses machinelearning (ML) to estimate solar energy production with the goal of revolutionizing decision-making processes through the analysis of large datasets and the generation of accurate forecasts.Solar meteorological data is analyzed methodologically using regression, time series analysis, and deep learning algorithms. The study demonstrates how well machine learning-based forecasting works to anticipate future solar energy output. Quantitative evaluations show excellent prediction accuracy and verify the techniques used. For example, the key observations made were that the Multiple Linear Regression methods demonstrates reasonable predictive ability with moderate Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE) values yet slightly lower R-squared values compared to other methods.The study also provides a reflective analysis of result significance, methodology dependability, and result generalizability, as well as a summary of its limits and recommendations for further study. The conclusion provides implications for broader applications across energy sectors and emphasizes the critical role that ML-based forecasting plays in predicting solar energy generation. By utilizing renewable energy sources like solar power, this approach aims to lessen dependency on non-renewable resources and pave the way for a more sustainable future. |
|---|---|
| Keywords: |
Renewable energy |
| College: |
Faculty of Science and Engineering |
| Funders: |
Qatar Research Leadership Program |