Non-Contact Geomagnetic Detection Using Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise and Teager Energy Operator

Experimental and numerical analysis of non-contact magnetic detecting signal of girth welds on steel pipelines

The quality of the girth welds on pipelines is a critical point regarding the safe operation. Non-contact pipeline magnetic detection (NPMD) is a non-destructive detection technology based on the metal magnetic memory (MMM) method. However, present studies mostly focus on the qualitative analysis of girth welds instead of accurate quantitative analysis of the stress status. Here, many hydraulic tests in sealed pipelines are performed to investigate the magnetic signal under different internal pressures and detection heights. A numerical model of magnetic signal is established and verified by the experimental results. The results show the characteristics of the signal that the y component has sinusoidal fluctuations when the x and z component reach the extreme values. A new parameter K_vs is proposed to comprehensively reflect the stress status of the girth welds. It is founded that the residual strength ratio (RSR) reduces from 0.97 to 0.83 when the K_vs max increases from 7500 to 13500 nT/m The magnetic signals decay exponentially in the second order when the detection height varies within 0.1-1.0 m. This study provides a theoretical and experimental basis for identifying the stress status of the girth welds on pipelines.

A Novel Noise Reduction Method of UAV Magnetic Survey Data Based on CEEMDAN, Permutation Entropy, Correlation Coefficient and Wavelet Threshold Denoising

Despite the increased attention that has been given to the unmanned aerial vehicle (UAV)-based magnetic survey systems in the past decade, the processing of UAV magnetic data is still a tough task. In this paper, we propose a novel noise reduction method of UAV magnetic data based on complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN), permutation entropy (PE), correlation coefficient and wavelet threshold denoising. The original signal is first decomposed into several intrinsic mode functions (IMFs) by CEEMDAN, and the PE of each IMF is calculated. Second, IMFs are divided into four categories according to the quartiles of PE, namely, noise IMFs, noise-dominant IMFs, signal-dominant IMFs, and signal IMFs. Then the noise IMFs are removed, and correlation coefficients are used to identify the real signal-dominant IMFs. Finally, the wavelet threshold denoising is applied to the real signal-dominant IMFs, the denoised signal can be obtained by combining the signal IMFs and the denoised IMFs. Both synthetic and field experiments are conducted to verify the effectiveness of the proposed method. The results show that the proposed method can eliminate the interference to a great extent, which lays a foundation for the further interpretation of UAV magnetic data.