An Optimized Maximum Second-Order Cyclostationary Blind Deconvolution and Bidirectional Long Short-Term Memory...

Area of Science:

• Mechanical Engineering
• Signal Processing
• Artificial Intelligence

Background:

• Extracting fault features from rolling bearings in noisy environments is challenging.
• Accurate identification of bearing fault conditions is crucial for predictive maintenance.

Purpose of the Study:

• To develop an effective technique for diagnosing rolling bearing faults under strong noisy environments.
• To enhance fault feature extraction and identification accuracy.

Main Methods:

• Utilized parameter-optimized maximum second-order cyclostationary blind deconvolution (CYCBD) and bidirectional long short-term memory (BiLSTM) networks.
• Employed an adaptive golden jackal optimization (GJO) algorithm to refine CYCBD parameters.
• Filtered and denoised rolling bearing fault signals using optimized CYCBD before fault classification with BiLSTM.

Main Results:

• Demonstrated strong noise reduction performance and high diagnostic accuracy.
• Achieved an approximate 9.89% improvement in accuracy compared to other methods at a signal-to-noise ratio (SNR) of -9 dB.
• Successfully classified rolling bearing faults in noisy backgrounds.

Conclusions:

• The optimized CYCBD-BiLSTM technique is effective for diagnosing rolling bearing faults in noisy environments.
• The proposed method offers a robust solution for industrial applications requiring reliable fault detection.