Application of Machine Learning for the Automation of the Quality Control of Noise Filtering Processes in Seismic Data Imaging

Mohamed Mejri,Maiza Bekara

doi:10.3390/geosciences10120475

Mohamed Mejri, Maiza Bekara

Open Access

https://doi.org/10.3390/geosciences10120475

Copy DOI

Abstract

Seismic imaging is the main technology used for subsurface hydrocarbon prospection. It provides an image of the subsurface using the same principles as ultrasound medical imaging. As for any data acquired through hydrophones (pressure sensors) and/or geophones (velocity/acceleration sensors), the raw seismic data are heavily contaminated with noise and unwanted reflections that need to be removed before further processing. Therefore, the noise attenuation is done at an early stage and often while acquiring the data. Quality control (QC) is mandatory to give confidence in the denoising process and to ensure that a costly data re-acquisition is not needed. QC is done manually by humans and comprises a major portion of the cost of a typical seismic processing project. It is therefore advantageous to automate this process to improve cost and efficiency. Here, we propose a supervised learning approach to build an automatic QC system. The QC system is an attribute-based classifier that is trained to classify three types of filtering (mild = under filtering, noise remaining in the data; optimal = good filtering; harsh = over filtering, the signal is distorted). The attributes are computed from the data and represent geophysical and statistical measures of the quality of the filtering. The system is tested on a full-scale survey (9000 km2) to QC the results of the swell noise attenuation process in marine seismic data.

Highlights

For the oil and gas industry, localizing and characterizing economically worthwhile geological reservoirs is essential
The seismic dataset is always contaminated by different types of noise and unwanted reflections; the noise attenuation is an important step in a typical seismic data processing sequence
Machine learning techniques such as deep neural networks or support vector machines [2] have been widely used in various domains to automatically predict several patterns from relevant features and was used to automate numerous human observation-based assessment processes

Summary

Introduction

For the oil and gas industry, localizing and characterizing economically worthwhile geological reservoirs is essential. As with every important step, performing quality control (QC). After each noise attenuation process is essential to ensure that noise has been sufficiently removed while no useful signal has been distorted. Improving the efficiency and the turnaround of seismic processing projects can be achieved by automating the QC process. Machine learning techniques such as deep neural networks or support vector machines [2] have been widely used in various domains to automatically predict several patterns from relevant features and was used to automate numerous human observation-based assessment processes. We investigate the use of those techniques to automate the quality control process of seismic data

Methods

Results

Conclusion