Health assessment and health trend prediction of wind turbine bearing based on BO-BiLSTM model DOI Creative Commons

Zhenen Li,

Yujie Xue

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Март 17, 2025

During the long-term operation of wind turbines, due to environmental factors and equipment aging, health reliability each component will gradually decline, leading failure. To assess status timely grasp subsequent changes development trends, it is necessary extract degradation characteristics, including time domain, frequency time-frequency domain characteristics. These characteristics can reflect operating equipment, help build indicator curves, evaluate high-speed shaft bearings turbines. Selecting reasonable an important prerequisite for constructing a index curve, using evaluation indicators construct comprehensive function screen The feature fusion method based on self-organizing mapping network used fuse multiple selected features into curve that bearing process. Finally, quantitative analysis performed scientifically bearings. Bearings are one key components Based constructed in this article, appropriate prediction model predict trend A effective trends turbine great practical significance formulating scientific maintenance measures farms. work article be divided following four parts: (1) Extracting vibration signals turbines; (2) Comprehensive monotonicity, correlation, robustness constructs degenerate features; (3) Use network. integrates curve; (4) optimize BiLSTM hyperparameters through Bayesian establish BO-BiLSTM achieve more accurate trend.

Язык: Английский

Federated Learning and Neural Circuit Policies: A Novel Framework for Anomaly Detection in Energy-Intensive Machinery DOI Creative Commons
Giulia Palma, Giovanni Geraci, Antonio Rizzo

и другие.

Energies, Год журнала: 2025, Номер 18(4), С. 936 - 936

Опубликована: Фев. 15, 2025

In the realm of predictive maintenance for energy-intensive machinery, effective anomaly detection is crucial minimizing downtime and optimizing operational efficiency. This paper introduces a novel approach that integrates federated learning (FL) with Neural Circuit Policies (NCPs) to enhance in compressors utilized leather tanning operations. Unlike traditional Long Short-Term Memory (LSTM) networks, which rely heavily on historical data patterns often struggle generalization, NCPs incorporate physical constraints system dynamics, resulting superior performance. Our comparative analysis reveals significantly outperform LSTMs accuracy interpretability within framework. innovative combination not only addresses pressing privacy concerns but also facilitates collaborative across decentralized sources. By showcasing effectiveness FL NCPs, this research paves way advanced strategies prioritize both performance integrity industries.

Язык: Английский

Процитировано

1

Health assessment and health trend prediction of wind turbine bearing based on BO-BiLSTM model DOI Creative Commons

Zhenen Li,

Yujie Xue

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Март 17, 2025

During the long-term operation of wind turbines, due to environmental factors and equipment aging, health reliability each component will gradually decline, leading failure. To assess status timely grasp subsequent changes development trends, it is necessary extract degradation characteristics, including time domain, frequency time-frequency domain characteristics. These characteristics can reflect operating equipment, help build indicator curves, evaluate high-speed shaft bearings turbines. Selecting reasonable an important prerequisite for constructing a index curve, using evaluation indicators construct comprehensive function screen The feature fusion method based on self-organizing mapping network used fuse multiple selected features into curve that bearing process. Finally, quantitative analysis performed scientifically bearings. Bearings are one key components Based constructed in this article, appropriate prediction model predict trend A effective trends turbine great practical significance formulating scientific maintenance measures farms. work article be divided following four parts: (1) Extracting vibration signals turbines; (2) Comprehensive monotonicity, correlation, robustness constructs degenerate features; (3) Use network. integrates curve; (4) optimize BiLSTM hyperparameters through Bayesian establish BO-BiLSTM achieve more accurate trend.

Язык: Английский

Процитировано

0