Electricity consumption probability density forecasting method based on LASSO-Quantile Regression Neural Network
- Submitting institution
-
Birmingham City University
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 11Z_OP_D0084
- Type
- D - Journal article
- DOI
-
10.1016/j.apenergy.2018.10.061
- Title of journal
- Applied Energy
- Article number
- -
- First page
- 565
- Volume
- 234
- Issue
- -
- ISSN
- 0306-2619
- Open access status
- Compliant
- Month of publication
- -
- Year of publication
- 2019
- URL
-
https://www.sciencedirect.com/science/article/abs/pii/S0306261918316301
- 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
-
-
- Research group(s)
-
-
- Citation count
- 32
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- The paper proposed a new probability density forecasting method to predict electricity consumption, based on Least Absolute Shrinkage and Selection Operator-Quantile Regression Neural Network (LASSO-QRNN). The method not only improves predictive accuracy of electricity consumption, but also outputs a probability distribution instead of a single-valued prediction, which provides insights into the consumption pattern. Accurate consumption forecasting is essential to energy investment planning. The superiority of the method is proven through the experiment on two large real-world data sets comprehensively, Guangdong province in China and California in U.S., in comparison with four other state-of-the-art methods including RBF, BP, QR, and NLQR.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -