Adaptive parameter-optimized NLM algorithm to denoise vibration signals of hydropower units

Xiang  Wu; Zhibo Jiang; Renbo Tang; Yun Luo; Kefei  Zhang

doi:10.59400/sv.v59i1.1688

Adaptive parameter-optimized NLM algorithm to denoise vibration signals of hydropower units

Xiang Wu Power China Zhongnan Engineering Corporation Limited, Changsha 410014, China
Zhibo Jiang Power China Zhongnan Engineering Corporation Limited, Changsha 410014, China
Renbo Tang Power China Zhongnan Engineering Corporation Limited, Changsha 410014, China
Yun Luo Power China Zhongnan Engineering Corporation Limited, Changsha 410014, China
Kefei Zhang Hubei University of Technology, Wuhan 430068, China

Article ID: 1688

DOI: https://doi.org/10.59400/sv.v59i1.1688

Keywords: Bayesian parameter optimization; non-local means algorithm; denoising vibration signals; monitor and diagnose operating status

Abstract

Monitoring and diagnosing the operating state of hydropower units is crucial, which becomes a hot research topic in the industry. the vibration signals provide a reliable indication to detect the abnormal working conditions of hydropower units. however, the vibration signals is affected by the environment noise inevitably, making it difficult to truly reflect the operating state of hydropower units. the non-local means (NLM) algorithm is proved to be effective in denoising the vibration signals, however, whose parameters depend on the human experience, which hinders its application and development. in the present work, based on the Bayesian parameter optimization (BPO), the parameters of NLM are set adaptively, the BPO-NLM denoising algorithm is proposed. by conducting the simulation, the denoising effectiveness of BPO-NLM is improved remarkably than that of the traditional NLM. at different snr, RMSE of the signal denoised by BPO-NLM is much smaller than that of the traditional NLM, while snr of the signal denoised by BPO-NLM is much larger, namely, the effective component of the signal is enhanced, while the noise component is suppressed.

References

[1]Huang H, Qin A, Mao H, et al. The Prediction Method on the Early Failure of Hydropower Units Based on Gaussian Process Regression Driven by Monitoring Data. Applied Sciences. 2020; 11(1): 153. doi: 10.3390/app11010153

[2]Liu Y, Xu Y, Liu J, et al. Real-time comprehensive health status assessment of hydropower units based on multi-source heterogeneous uncertainty information. Measurement. 2023; 216: 112979. doi: 10.1016/j.measurement.2023.112979

[3]Guo W, Wu F. Hydraulic-mechanical coupling vibration performance of pumped storage power station with two turbine units sharing one tunnel. Journal of Energy Storage. 2022; 53: 105082. doi: 10.1016/j.est.2022.105082

[4]Li X, Xu Z, Guo P. Swing Trend Prediction of Main Guide Bearing in Hydropower Units Based on MFS-DCGNN. Sensors. 2024; 24(11): 3551. doi: 10.3390/s24113551

[5]Shen A, Chen Y, Zhou J, et al. Hydraulic Vibration and Possible Exciting Sources Analysis in a Hydropower System. Applied Sciences. 2021; 11(12): 5529. doi: 10.3390/app11125529

[6]Kong F, Song C, Zhuo Y. Vibration Fault Analysis of Hydropower Units Based on Extreme Learning Machine Optimized by Improved Sparrow Search Algorithm. Journal of Vibration Engineering & Technologies. 2022; 11(4): 1609-1622. doi: 10.1007/s42417-022-00660-3

[7]Gemechu E, Kumar A. A review of how life cycle assessment has been used to assess the environmental impacts of hydropower energy. Renewable and Sustainable Energy Reviews. 2022; 167: 112684. doi: 10.1016/j.rser.2022.112684

[8]Xu R, Tan X, Wang H, et al. Stability of hydropower units under full operating conditions considering nonlinear coupling of turbine characteristics. Renewable Energy. 2024; 223: 120009. doi: 10.1016/j.renene.2024.120009

[9]Oppliger J, Denner MM, Küspert J, et al. Weak signal extraction enabled by deep neural network denoising of diffraction data. Nature Machine Intelligence. 2024; 6(2): 180-186. doi: 10.1038/s42256-024-00790-1

[10]Chen C, Chen Z, Zhou Y, et al. A reliable evaluation approach for multichannel signal denoising algorithms based on a novel arterial pulse acquisition system. Heliyon. 2024; 10(5): e26140. doi: 10.1016/j.heliyon.2024.e26140

[11]Sahoo GR, Freed JH, Srivastava M. Optimal Wavelet Selection for Signal Denoising. IEEE Access. 2024; 12: 45369-45380. doi: 10.1109/access.2024.3377664

[12]Wei M, Sun X, Zong J. Time–Frequency Domain Seismic Signal Denoising Based on Generative Adversarial Networks. Applied Sciences. 2024; 14(11): 4496. doi: 10.3390/app14114496

[13]Bai Y, Chen Z, Xie J, et al. Daily reservoir inflow forecasting using multiscale deep feature learning with hybrid models. Journal of Hydrology. 2016; 532: 193-206. doi: 10.1016/j.jhydrol.2015.11.011

[14]Yu X, Zhang X, Qin H. A data-driven model based on Fourier transform and support vector regression for monthly reservoir inflow forecasting. Journal of Hydro-environment Research. 2018; 18: 12-24. doi: 10.1016/j.jher.2017.10.005

[15]Ashraf FB, Haghighi AT, Riml J, et al. A Method for Assessment of Sub‐Daily Flow Alterations Using Wavelet Analysis for Regulated Rivers. Water Resources Research. 2021; 58(1). doi: 10.1029/2021wr030421

[16]Guo W, He N, Ban X, et al. Multi-scale variability of hydrothermal regime based on wavelet analysis—The middle reaches of the Yangtze River, China. Science of The Total Environment. 2022; 841: 156598. doi: 10.1016/j.scitotenv.2022.156598

[17]Darbon J, Cunha A, Chan T. From nano to macro. In: Proceedings of the 5th IEEE international symposium on biomedical imaging; 14–17 May 2008; Paris, France.

[18]Tracey BH, Miller EL. Nonlocal Means Denoising of ECG Signals. IEEE Transactions on Biomedical Engineering. 2012; 59(9): 2383-2386. doi: 10.1109/tbme.2012.2208964

[19]Lv Y, Zhu Q, Yuan R. Fault Diagnosis of Rolling Bearing Based on Fast Nonlocal Means and Envelop Spectrum. Sensors. 2015; 15(1): 1182-1198. doi: 10.3390/s150101182

[20]Lv Y, Yuan R, Song G. Multivariate empirical mode decomposition and its application to fault diagnosis of rolling bearing. Mechanical Systems and Signal Processing. 2016; 81: 219-234. doi: 10.1016/j.ymssp.2016.03.010

[21]Van M, Kang H, Shin K. Rolling element bearing fault diagnosis based on non‐local means de‐noising and empirical mode decomposition. IET Science, Measurement & Technology. 2014; 8(6): 571-578. doi: 10.1049/iet-smt.2014.0023

[22]Van M, Kang H. Bearing‐fault diagnosis using non‐local means algorithm and empirical mode decomposition‐based feature extraction and two‐stage feature selection. IET Science, Measurement & Technology. 2015; 9(6): 671-680. doi: 10.1049/iet-smt.2014.0228

[23]Laha SK. Enhancement of fault diagnosis of rolling element bearing using maximum kurtosis fast nonlocal means denoising. Measurement. 2017; 100: 157-163. doi: 10.1016/j.measurement.2016.12.058

[24]Bing P, Liu W, Wang Z, et al. Noise Reduction in ECG Signal Using an Effective Hybrid Scheme. IEEE Access. 2020; 8: 160790-160801. doi: 10.1109/access.2020.3021068

[25]Guo Z, Wan Y, Ye H. An Unsupervised Fault-Detection Method for Railway Turnouts. IEEE Transactions on Instrumentation and Measurement. 2020; 69(11): 8881-8901. doi: 10.1109/tim.2020.2998863

[26]Zhang X. Two-step non-local means method for image denoising. Multidimensional Systems and Signal Processing. 2021; 33(2): 341-366. doi: 10.1007/s11045-021-00802-y

[27]Wei Z, Wang Y, He S, et al. A novel intelligent method for bearing fault diagnosis based on affinity propagation clustering and adaptive feature selection. Knowledge-Based Systems. 2017; 116: 1-12. doi: 10.1016/j.knosys.2016.10.022

[28]Lu Z, Xiao JW, Huang Z. Metric Learning Based Rolling Bearing Faults Diagnosis with Curvelet Transform. Journal of Physics: Conference Series. 2019; 1229(1): 012037. doi: 10.1088/1742-6596/1229/1/012037

[29]Snoek J, Larochelle H, Adams R. Practical Bayesian optimization of machine learning algorithms. Advances in Neural Information Processing Systems; 2012. pp. 2951-2959.

[30]Dwyer R. Detection of non-Gaussian signals by frequency domain Kurtosis estimation. ICASSP ’83 IEEE International Conference on Acoustics, Speech, and Signal Processing. 1983; 8: 607-610. doi: 10.1109/icassp.1983.1172264

[31]Antoni J, Randall RB. The spectral kurtosis: application to the vibratory surveillance and diagnostics of rotating machines. Mechanical Systems and Signal Processing. 2006; 20(2): 308-331. doi: 10.1016/j.ymssp.2004.09.002

[32]An X, Yang J. A method of eliminating the vibration signal noise of hydropower unit based on NA-MEMD and approximate entropy. Proceedings of the Institution of Mechanical Engineers, Part E: Journal of Process Mechanical Engineering. 2016; 231(2): 317-328. doi: 10.1177/0954408915595763

Published

2024-10-28

How to Cite

Wu, X., Jiang, Z., Tang, R., Luo, Y., & Zhang, K. (2024). Adaptive parameter-optimized NLM algorithm to denoise vibration signals of hydropower units. Sound & Vibration, 59(1), 1688. https://doi.org/10.59400/sv.v59i1.1688

Download Citation

Issue

Vol. 59 No. 1 (2025)

Section

Article

This work is licensed under a Creative Commons Attribution 4.0 International License.

Editor-in-Chief

Prof. Jun Yang

Institute of Acoustics, Chinese Academy of Sciences, China

ISSN

1541-0161 (Print)

2693-1443 (Online)

Publication Frequency

Bi-monthly

About the Publisher

Academic Publishing insists on taking academic exchange and publication as the main line, carrying out comprehensive management based on science and technology, and fully exploring excellent international publishing resources. Within 5 years, it will form a strategic framework and scale with science (S), technology (T), medicine (M), education (E), and humanities and arts (H) as the main publishing fields. Academic Publishing is headquartered in Singapore and based in Malaysia, with the United States and China providing the main scientific and academic resources. At the same time, it has established long-term good cooperative relations with other publishing companies, scientific research communities, and academic organizations in more than a dozen countries and regions. Academic Publishing uses English and Chinese as its main publishing languages, mainly publishing books, journals, and conference papers in print and online. The vast majority of publications follow the international open access policy, providing stable and long-term quality and professional publications. With the joint efforts of the expert team and our professional editorial team, our publications will gradually be indexed by international databases in stages to provide convenient and professional retrieval for various scholars. At the same time, manuscripts we accept will be subject to the peer review principle, and cutting-edge and innovative research articles will be preferentially accepted for peer reference and discussion. All kinds of our publications are welcome for peer to contribute, access, and download.

more

Volume Arrangement

Vol 59 Iss 2 (2025)

Vol 59 Iss 1 (2025)

more issues

Featured Articles

Soundscapes in Arab Cities: A Systematic Review and Research Agenda

In the context of Arab cities, this study explores the intricate interplay between cultural, historical, and environmental elements that shape their unique soundscapes. The paper aims to shed light on this underrepresented field of study by employing a three-fold research approach: systematic review, a comprehensive literature review, and the formulation of a future research agenda. The first part of the investigation focuses on research productivity in the Arab world regarding soundscape studies. An analysis of publication trends reveals that soundscape research in Arab cities is still an emerging area of interest. Critical gaps in the existing body of literature are identified, highlighting the importance of addressing these gaps within the broader context of global soundscape research. The second part of the study delves into the distinctive features that inform the soundscapes of Arab cities. These features are categorized into three overarching groups: (i) cultural and religious life, (ii) daily life, and (iii) heritage and history, by exploring these factors, the study aims to elucidate the multifaceted nature of Arab urban soundscapes. From the resonating calls to prayer and the vibrant ambiance of traditional cafes to the bustling markets and architectural characteristics, each factor contributes to the auditory tapestry that defines Arab cities. The paper concludes with a forward-looking research agenda, proposing sixteen key questions organized into descriptive and comparative categories. These questions emphasize the need for a more profound understanding of sound perception, sources, and the impact of urban morphology on the soundscape. Additionally, they highlight the need for interdisciplinary research, involving fields such as urban planning, architecture, psychology, sociology, and cultural studies to unravel the complexity of Arab urban soundscapes.

Prediction of Sound Transmission Loss of Vehicle Floor System Based on 1D-Convolutional Neural Networks

The Noise, Vibration, and Harshness (NVH) experience during driving is significantly influenced by the sound insulation performance of the car floor acoustic package. As such, accurate and efficient predictions of its sound insulation performance are crucial for optimizing related noise reduction designs. However, the complex acoustic transmission mechanisms and difficulties in characterizing the sound absorption and insulation properties of the floor acoustic package pose significant challenges to traditional Computer-Aided Engineering (CAE) methods, leading to low modeling efficiency and prediction accuracy. To address these limitations, a hierarchical multi-objective decomposition system for predicting the sound insulation performance of the floor acoustic package has been developed based on an analysis of the noise transmission path. This approach involves introducing a 1D-Convolutional Neural Network (1D-CNN) model for predicting the sound insulation performance of the floor acoustic package, thereby avoiding the limitations of conventional CAE approaches that rely solely on “data-driven” methods. The proposed method was applied and tested using specific vehicle models, and the results demonstrated the effectiveness and superiority of the proposed approach relative to those obtained using 2D-CNN and Support Vector Regression (SVR) models.

A Sound Quality Evaluation Method for Vehicle Interior Noise Based on Auditory Loudness Model

When designing and optimizing the hull of vehicles, their sound quality needs to be considered, which greatly depends on the psychoacoustic parameters. However, the traditional psychoacoustic calculation method does not consider the influence of the real human ear anatomic structure, even the loudness which is most related to the auditory periphery. In order to introduce the real physiological structure of the human ear into the evaluation of vehicle sound quality, this paper first carried out the vehicle internal noise test to obtain the experimental samples. Then, the physiological loudness was predicted based on an established human ear physiological model, and the noise evaluation vector was constructed by combining the remaining four psychoacoustic parameters. Finally, the evaluation vector was fitted into the subjective evaluation results of vehicle interior noise by a deep neural network. The results show that our proposed method can estimate the human subjective perception of vehicle interior noise well.

Prediction and Analysis of Vehicle Interior Road Noise Based on Mechanism and Data Series Modeling

Currently, the inexorable trend toward the electrification of automobiles has heightened the prominence of road noise within overall vehicle noise. Consequently, an in-depth investigation into automobile road noise holds substantial practical importance. Previous research endeavors have predominantly centered on the formulation of mechanism models and data-driven models. While mechanism models offer robust controllability, their application encounters challenges in intricate analyses of vehicle body acoustic-vibration coupling, and the effective utilization of accumulated data remains elusive. In contrast, data-driven models exhibit efficient modeling capabilities and can assimilate conceptual vehicle knowledge, but they impose stringent requirements on both data quality and quantity. In response to these considerations, this paper introduces an innovative approach for predicting vehicle road noise by integrating mechanism-driven and data-driven methodologies. Specifically, a series model is devised, amalgamating mechanism analysis with data-driven techniques to predict vehicle interior noise. The simulation results from dynamic models serve as inputs to the data-driven model, ultimately generating outputs through the utilization of the Long Short-Term Memory with Autoencoder (AE-LSTM) architecture. The study subsequently undertakes a comparative analysis between different dynamic models and data-driven models, thereby validating the efficacy of the proposed series vehicle road noise prediction model. This series model, encapsulating the rigid-flexible coupling dynamic model and AE-LSTM series model, not only demonstrates heightened computational efficiency but also attains superior prediction accuracy.