MINOR FAULT DETECTION BY INTEGRATION OF SEISMIC ATTRIBUTES IN AN OIL RESERVOIR

摘要

Seismic section can often help detect the exact location and movement of minor faults, but sometimes the poor quality of the data makes that impossible. It is well known that the existence of minor faults and fractures in an oil reservoir play an important role in increasing its productivity. The minor faults may cut the cap rock and cause the oil to leak from the reservoir. So, before making any decisions for drilling in an oil field it is very important to know the exact locations of minor faults. In order to detect minor faults, a single seismic attribute is usually applied, but the results are not satisfactory. In this paper, for minor fault detection, we introduce a method based on a combination of seismic attributes in a Neural Network system. Firstly, different attributes like energy, similarity, dip variance, polar dip and polar dip angle with different time gates were applied on a seismic section. Then to combine these attributes together and apply them, a workflow was constructed, in which an artificial neural network system was designed, the above-mentioned attributes were introduced to the system and hand-picked faults were input to the ANN system. When the training of the system completes, the ANN estimates an output cube that indicates the faults location. The obtained results based on this study showed that using a combination of attributes in the ANN system is more reliable than applying a single attribute to locate the minor faults in an oil reservoir.