Electromagnetic (EM) downhole tool telemetry has the potential to yield the greatest cost savings over other methods such as optical, mud pulse, power-line communication, and acoustic telemetry. However, its reliability is challenged in wellsites with large formation conductivity extremes, rig noise, and tool depth. As a result, conventional methods incur additional hardware costs to improve their reliability, such as by adding redundancy with a different telemetry method or by making measurements closer to the tool from an adjacent well. The purpose of our study was to evaluate the feasibility of applying deep-learning methods to attenuate rig noise within the EM telemetry signal band (1-25Hz) and separate the downhole tool source signal without incurring additional hardware costs. Successful architectures for seismic noise attenuation and speech source separation were evaluated on synthetic and field data sets of EM source and rig noise data. Architectural adjustments were explored to shed light on which features are most useful with very low frequencies and very low signal-to-noise ratios present in EM telemetry. The signal-preserving and noise attenuation performance of proposed architectural improvements was evaluated in comparison to the original architectures and conventional methods.

This content is only available via PDF.
You can access this article if you purchase or spend a download.