Development of data-driven models for wind speed forecasting in Australia
Edited book (chapter)
Chapter Title | Development of data-driven models for wind speed forecasting in Australia |
---|---|
Book Chapter Category | Edited book (chapter) |
ERA Publisher ID | 1821 |
Book Title | Predictive modelling for energy management and power systems engineering |
Authors | Neupane, Ananta (Author), Raj, Nawin (Author), Deo, Ravinesh (Author) and Ali, Mumtaz (Author) |
Editors | Deo, R., Samui, Pijush and Roy, Sanjiban Sekhar |
Page Range | 143-190 |
Chapter Number | 6 |
Number of Pages | 48 |
Year | 2021 |
Publisher | Elsevier |
Place of Publication | Netherlands |
ISBN | 9780128177723 |
9780128177730 | |
Digital Object Identifier (DOI) | https://doi.org/10.1016/B978-0-12-817772-3.00006-9 |
Web Address (URL) | https://www.sciencedirect.com/science/article/abs/pii/B9780128177723000069 |
Abstract | To provide a modeling framework for wind energy harnessing, there are several data-driven models used for wind speed prediction at a global scale but the accuracy of one particular model in a specific location does not guarantee its applicability in the others, in the climatologically diverse wind farm locations. Therefore, to resolve this challenge, it is important to develop and implement advance machine learning models for wind speed forecasting in a specific location. This chapter is based on the development and comparison of data-driven modeling for short (6 hourly) and medium-term (daily) wind speed forecasting horizons in wind potential regions in Australia. In this chapter, artificial neural networks, multiple linear regressions, random forests, M5 tree, and autoregressive integrated moving-average models are developed. The research literature shows that these models have not been used comparatively for wind speed forecasting in specific wind belt locations in Australia and, most importantly, for multiple forecasting horizons, including short-term and daily scales that this chapter aims to undertake. Thus this chapter makes an important contribution to the wind speed forecasting area, and in particular, the future opportunity for renewable energy researchers to explore these data-driven models to generate wind speed predictions with artificial intelligence techniques. It emphasizes the important implications for renewable energy feasibility studies, and several other applications in the climate dynamics and modeling areas where a prior knowledge of the future wind regimes is required. |
Keywords | wind speed forecasting; Australia |
ANZSRC Field of Research 2020 | 419999. Other environmental sciences not elsewhere classified |
469999. Other information and computing sciences not elsewhere classified | |
Public Notes | Files associated with this item cannot be displayed due to copyright restrictions. |
Byline Affiliations | School of Sciences |
Deakin University | |
Institution of Origin | University of Southern Queensland |
https://research.usq.edu.au/item/q5yq1/development-of-data-driven-models-for-wind-speed-forecasting-in-australia
287
total views12
total downloads1
views this month0
downloads this month