Acoustic-Based Condition Recognition for Pumped Storage Units Using a Hierarchical Cascaded CNN and MHA-LSTM Model

Linghua Kong, Nan Hu, Hongyong Zheng, Xulei Zhou, Jian Wang, Weijiao Li, Yang Lu, Ziwei Zhang () and Jianyi Lin ()
Additional contact information
Linghua Kong: Fujian Xianyou Pumped Storage Power Co., Ltd., Putian 351267, China
Nan Hu: Fujian Xianyou Pumped Storage Power Co., Ltd., Putian 351267, China
Hongyong Zheng: Fujian Xianyou Pumped Storage Power Co., Ltd., Putian 351267, China
Xulei Zhou: Fujian Xianyou Pumped Storage Power Co., Ltd., Putian 351267, China
Jian Wang: Fujian Xianyou Pumped Storage Power Co., Ltd., Putian 351267, China
Weijiao Li: State Key Laboratory of Regional and Urban Ecology, Xiamen Key Lab of Urban Metabolism, Research Center of Urban Carbon Neutrality, Institute of Urban Environment, Chinese Academy of Sciences, Xiamen 361021, China
Yang Lu: Unisound AI Technology Co., Ltd., Xiamen 361022, China
Ziwei Zhang: State Key Laboratory of Regional and Urban Ecology, Xiamen Key Lab of Urban Metabolism, Research Center of Urban Carbon Neutrality, Institute of Urban Environment, Chinese Academy of Sciences, Xiamen 361021, China
Jianyi Lin: State Key Laboratory of Regional and Urban Ecology, Xiamen Key Lab of Urban Metabolism, Research Center of Urban Carbon Neutrality, Institute of Urban Environment, Chinese Academy of Sciences, Xiamen 361021, China

Energies, 2025, vol. 18, issue 16, 1-23

Abstract: As an important regulating resource in power systems, pumped storage units frequently switch operating conditions due to peak shaving and frequency regulation, making the condition transitions complex. Traditional methods struggle to achieve high-precision classification. This paper proposes a hierarchical cascade deep learning model based on noise signals, which integrates a convolutional neural network (CNN) with a multi-head attention long short-term memory network (MHA-LSTM) to address the differentiated recognition of steady-state and transitional conditions. The CNN efficiently extracts multi-scale spatial features from sound spectrograms, enabling fast classification under steady-state conditions. The MHA-LSTM combines attention mechanisms with time-series modeling. This enhances its ability to capture long-range dependencies in the signals. And it significantly improves classification accuracy in ambiguous boundaries and transitional scenarios. Testing on 3413 noise samples shows that the proposed method achieves an overall accuracy of 92.22%, with steady-state condition recognition exceeding 98%, and recall and F1 score above 90% for major categories. Compared with other approaches, this model provides a high-precision classification tool for unit health monitoring, supporting the intelligent operation and maintenance of power plants.

Keywords: pumped storage units; acoustic condition recognition; convolutional neural network; long short-term memory network; multi-head attention; hierarchical cascaded model (search for similar items in EconPapers)
JEL-codes: Q Q0 Q4 Q40 Q41 Q42 Q43 Q47 Q48 Q49 (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/1996-1073/18/16/4269/pdf (application/pdf)
https://www.mdpi.com/1996-1073/18/16/4269/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jeners:v:18:y:2025:i:16:p:4269-:d:1722109

Access Statistics for this article

Energies is currently edited by Ms. Agatha Cao

More articles in Energies from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().