Optimization of Wavelet Threshold based on Sparrow Search Algorithm Integrating Tent Chaos and Cauchy Mutation with Application in Bearing Fault Diagnosis
DOI:
https://doi.org/10.54691/e4myef82Keywords:
Improved sparrow search algorithm, wavelet threshold denoising, tent chaotic mapping, Cauchy mutation, fault diagnosis.Abstract
Wavelet-threshold denoising is strongly affected by the selected threshold when the signal is non-stationary and the noise level is unknown. An overly large threshold may remove impact components, whereas an overly small threshold leaves residual noise after reconstruction. This paper treats threshold selection as an optimization problem and searches it with an improved sparrow search algorithm (ISSA). Tent chaotic mapping is used in the initialization stage to spread candidate thresholds over the search interval. Cauchy mutation is then applied near the current leading position to keep the search from staying too early in a narrow region. Minimum envelope entropy serves as the fitness value, so the search does not require a clean reference signal. The method is tested first on a composite signal containing sinusoidal components and a step term, and then on an outer-ring bearing fault signal. In the simulated signal test, the ISSA-based method increases the signal-to-noise ratio (SNR) by 2.67 dB and reduces the root mean square error (RMSE) by 26.6% compared with the traditional soft-threshold method. In the bearing fault test, the peak SNR of the envelope spectrum increases by 8.3 dB, and fault-related impact information remains visible after denoising.
Downloads
References
[1] Kumar, A., Zhou, Y., & Xiang, J. (2021). Optimization of vmd using kernel-based mutual information for the extraction of weak features to detect bearing defects. Measurement, 168, 108402. https://doi.org/10.1016/j.measurement.2020.108402
[2] Zhang, X., Miao, Q., Zhang, H., et al. (2021). An improved threshold function for signal denoising based on wavelet transform. Digital Signal Processing, 119, 103195. https://doi.org/10.1016/j.dsp.2021.103195
[3] Wang, H., Chen, J., & Dong, G. (2023). Adaptive threshold selection for wavelet denoising using deep reinforcement learning. Mechanical Systems and Signal Processing, 186, 109858. https://doi.org/10.1016/j.ymssp.2022.109858
[4] Wang, X., & Li, Y. (2024). Improved sparrow search algorithm based on multistrategy collaborative optimization. Processes, 12(12), 2775. https://doi.org/10.3390/pr12122775
[5] Chen, X., Zhang, B., & Gao, D. (2021). Bearing fault diagnosis base on multi-scale cnn and lstm model. Journal of Intelligent Manufacturing, 32(4), 971–987. https://doi.org/10.1007/s10845-020-01637-9
[6] Li, Y., Wang, X., Liu, Z., et al. (2022). Deep learning-based adaptive wavelet neural network for signal denoising. IEEE Transactions on Neural Networks and Learning Systems, 33(10), 5551–5565. https://doi.org/10.1109/TNNLS.2021.3128511
[7] Xue, J., & Shen, B. (2020). A novel swarm intelligence optimization approach: Sparrow search algorithm. Systems Science & Control Engineering, 8(1), 22–34. https://doi.org/10.1080/21642583.2019.1699885
[8] Liu, G., Shu, C., Liang, Z., et al. (2021). A modified sparrow search algorithm with application in 3d route planning for uav. Sensors, 21(4), 1224. https://doi.org/10.3390/s21041224
[9] Wang, D., Tan, D., & Liu, L. (2023). Cauchy mutation boosted harris hawk algorithm: Optimal performance design. Journal of Computational Design and Engineering, 10(2), 503–526. https://doi.org/10.1093/jcde/qwad006
[10] Zhang, H., & Li, X. (2025). Research on signal processing algorithms based on issa-vmd-awptd for wearable devices. Electronics, 14(14), 2761. https://doi.org/10.3390/electronics14142761
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Scientific Journal of Intelligent Systems Research
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
How to Cite
