Research on Fault Warning Algorithm for Coal Mills Based on CNN-LSTM-Attention
As the frequent failure issues of coal mills in thermal power plants operating under harsh conditions become increasingly prominent, it is particularly important to effectively predict faults and assess the remaining useful life of coal mills. This paper proposes a fault warning method for coal mills based on CNN-LSTM-Attention. This method employs deep learning techniques and selects data related to coal mill faults as input. By analyzing the operational data of coal m ills, it predicts the fault development trend and provides early warnings, thereby enhancing the operational safety and economic efficiency of thermal power units.
Authors
- Chengwen Wang (ORCID: https://orcid.org/0009-0004-5599-6383)
- Jiachen Cai (ORCID: https://orcid.org/0000-0003-3260-2551)
- Lin Gao (ORCID: https://orcid.org/0000-0003-1559-0886)
- Xiaolong Sheng (ORCID: https://orcid.org/0000-0003-4031-8139)
- Yuzhi Huang (ORCID: https://orcid.org/0009-0008-3903-7667)
- Haibin Liu (ORCID: https://orcid.org/0000-0002-8371-2403)
- Jun Li (ORCID: https://orcid.org/0000-0003-1156-3645)
- Honghui Wang (ORCID: https://orcid.org/0000-0002-2134-8968)
- Peng Chang
Institutions
- Thermal Power Research Institute (CN)
- Shaanxi Yanchang Petroleum (China) (CN)
Publication Details
- Published
- 2024-10-18
- DOI
- https://doi.org/10.1109/dtpi61353.2024.10778879
- Primary Topic
- Geoscience and Mining Technology
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Research and Development