3-D Far-Field Imaging Processing Algorithm for Borehole Acoustic Reflection Imaging Utilizing Both Acoustic Pressure and Particle Velocity

Borehole acoustic reflection imaging utilizes acoustic waves to identify well-side anomalies, but accurately obtaining the azimuth is an ongoing problem. In this study, a mathematical model was developed for the echo signal from a well-side anomaly recorded by a downhole cylindrical receiver array utilizing acoustic pressure sensors. The model was used to propose a three-dimensional (3D) multiple signal classification (3D-MUSIC) method for calculating the azimuth of the well-side anomaly based on the orthogonality of the echo signal subspace and noise subspace. Then, a mathematical model was derived for the signal recorded by a downhole receiver array utilizing acoustic vector sensors, which provide information on both the acoustic pressure and particle velocity. This model was used to propose a three-dimensional acoustic vector array MUSIC (3D-AVA-MUSIC) method for calculating the azimuth of a well-side anomaly with improved noise suppression and detection accuracy. Numerical simulations, physical simulations, and field tests were conducted to evaluate the performances of the proposed methods compared with existing methods. Compared with the existing downhole beamforming and 3D slowness time coherence methods, 3D-MUSIC and 3D-AVA-MUSIC demonstrated higher accuracy. In addition, 3D-AVA-MUSIC effectively utilized acoustic pressure and particle velocity information to perform better than 3D-MUSIC.