Wind Power Prediction Based on Multidimensional Power Technical Indicators and LSTNet Algorithm
A wind power prediction method combining multidimensional technical indicators and the LSTNet algorithm is proposed to address volatility in wind power time series. Using financial time series analysis, K-lines and technical indicators are integrated into the framework. A sliding time window creates a power K-line, followed by feature extraction and model building. Case study results from an Inner Mongolia wind farm show the model improves Mean Squared Error (MSE) by 24.2% over the benchmark.
Authors
- Bowen Cheng (ORCID: https://orcid.org/0000-0001-7470-8077)
- Hailong Li (ORCID: https://orcid.org/0000-0001-6449-9080)
- Yongsheng Wang (ORCID: https://orcid.org/0000-0003-3673-8376)
- Haotian Chen (ORCID: https://orcid.org/0000-0003-3896-4369)
- Chunlei Liu (ORCID: https://orcid.org/0000-0001-8816-4832)
- Xuehui Wang (ORCID: https://orcid.org/0000-0001-9577-3951)
Institutions
- Dalian University of Technology (CN)
- Inner Mongolia University of Technology (CN)
Publication Details
- Published
- 2024-11-29
- DOI
- https://doi.org/10.1109/ei264398.2024.10991360
- Primary Topic
- Smart Grid and Power Systems
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- National Natural Science Foundation of China