| CPC G01V 1/50 (2013.01) [E21B 49/00 (2013.01); G01V 1/226 (2013.01); G01V 2210/322 (2013.01); G01V 2210/34 (2013.01)] | 11 Claims |

|
1. A method comprising:
disposing a plurality of sensors within a borehole extending into a subterranean formation at an oil and gas well site, wherein the plurality of sensors comprise fiber-optic sensors;
receiving, by a processor, raw seismic data from the plurality of sensors, wherein the raw seismic data comprises distributed acoustic sensing (DAS) data collected by the plurality of sensors;
processing, by the processor, the raw seismic data to generate a plurality of processed seismic data waveforms and metadata associated with the plurality of processed seismic data waveforms, the processing the raw seismic data comprising:
stacking neighboring traces of the raw seismic data to cancel noise in a first processed seismic waveform; and
applying deep learning technology to the raw seismic data to identify a potential arrival of a seismic signal in a second processed seismic waveform;
combining, by the processor, the plurality of processed seismic data waveforms and the metadata associated with the plurality of processed seismic data waveforms to generate an output file; and
streaming, by the processor, the output file to an off-site data processing center that is located remotely from the oil and gas well site.
|