A Regime-Switching Recurrent Neural Network Model Applied to Wind Time Series
- Submitting institution
-
Goldsmiths' College
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 3124
- Type
- D - Journal article
- DOI
-
10.1016/j.asoc.2019.04.009
- Title of journal
- Applied Soft Computing
- Article number
- -
- First page
- 723
- Volume
- 80
- Issue
- -
- ISSN
- 1568-4946
- Open access status
- Compliant
- Month of publication
- July
- Year of publication
- 2019
- URL
-
http://research.gold.ac.uk/id/eprint/26305/
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
3
- Research group(s)
-
-
- Citation count
- 1
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- The proposed model is more accurate (it demonstrates better statistical characteristics) on distribution forecasting than previous models on a real-world wind speed time series. The results provide empirical evidence for the importance of nonlinearities when modeling environmental time series, and explicit treatment of the temporal dimension of the model. This is significant because of the increasing importance of green energy production. A highly accurate and efficient model can help to control accurately the production and distribution of energy by green farms (opened and to be opened soon in many places in the UK).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -