CEEMDAN-LWT De-Noising Method for Pipe-Jacking Inertial Guidance System Based on Fiber Optic Gyroscope

Area of Science:

• Geotechnical Engineering
• Signal Processing
• Navigation Systems

Background:

• Long-distance curved pipe jacking relies on inertial guidance systems, often using fiber optic gyroscopes (FOGs).
• Environmental factors like vibration, electromagnetism, and temperature introduce significant random noise into FOG signals, degrading guidance accuracy.
• Effective noise reduction is crucial for reliable FOG-based pipe jacking guidance.

Purpose of the Study:

• To propose and validate a hybrid de-noising method for fiber optic gyroscope (FOG) signals used in pipe jacking.
• To enhance the accuracy and reliability of inertial guidance systems by mitigating environmental noise.
• To improve the measurement precision of pipe-jacking attitude angles (pitch, roll, heading).

Main Methods:

• A hybrid de-noising approach termed CEEMDAN-LWT (Complete Ensemble Empirical Mode Decomposition with Adaptive Noise - Lifting Wavelet Transform) is introduced.
• FOG signals are decomposed using CEEMDAN to obtain intrinsic mode functions (IMFs), with relevant components selected based on correlation coefficients.
• High-frequency IMFs are further processed using LWT for enhanced resolution, followed by soft threshold de-noising and signal reconstruction.

Main Results:

• The CEEMDAN-LWT method demonstrated superior de-noising performance compared to standalone CEEMDAN or LWT, evidenced by improved SNR, RMSE, and Deviation metrics.
• Experimental validation on pipe-jacking dynamic and environmental interference tests confirmed the method's effectiveness and adaptivity.
• Navigation accuracy significantly improved, with absolute errors for pitch, roll, and heading angles reduced by 39.86%, 59.45%, and 14.29%, respectively.

Conclusions:

• The CEEMDAN-LWT hybrid de-noising method effectively suppresses random errors in FOG signals caused by environmental disturbances.
• This technique substantially enhances the measurement accuracy of pipe-jacking attitude, crucial for precise guidance in complex underground environments.
• The proposed method offers a robust solution for improving the performance of inertial guidance systems in demanding civil engineering applications.