Ifp Energies Nouvelles

Vincent Clochard

Michel Leger

Youri Hamon

Maxime Moreaud

Mathieu Feraille

Sylvain Desroziers

Antoine Bouziat

Thierry Tonellot

Marie-Lise Bernard

The present invention relates to a method of detecting a geological constituent (cge) of a rock sample, by use of a machine learning method (ALG) trained with a limited number of training images (IAP), through segmentation (SEG) and partitioning (DEC) of training images (IAP) into multiple patches.

Method of detecting at least one geological constituent of a rock sample

The invention relates to a method of monitoring the interior of a pipeline () positioned in contact with a soil (S) below a water mass (E), comprising implementation of the following steps by data processing means ():

Method of monitoring the interior of a subsea pipeline

The invention is a method of monitoring an underground formation into which a gas is injected or from which a gas is produced, by a stratigraphic inversion of seismic data with anamorphosis of impedances. Pre-injection and post-injection seismic data are acquired. A pre-injection seismic impedance cube IPis determined by a stratigraphic inversion of the pre-injection seismic data. An anamorphosis function &#x3b1; is defined by a function comprising a positive lower limit B, an upper limit B, an identity interval defined between limits Band B, with B&#x3c;B&#x3c;B&#x3c;B. A value for each limit B, B, Band Bis associated with each cell of the cube. Post-injection seismic impedances IPare determined by a stratigraphic inversion of the post-injection seismic data wherein the anamorphosis function &#x3b1; is applied to seismic impedances IP. The gas is located within the zone by identifying the cells where IP&#x3c;IP

Method of monitoring a geological gas storage site by stratigraphic inversion of seismic data

The invention is a method of constructing an image representative of a heterogeneous medium by a procedure of joint inversion of seismic data represented on different time scales having an application for underground reservoir exploration. Sequential inversion of the seismic data is performed so as to determine a first model from the seismic data expressed in a first time scale, and a second model from the seismic data expressed in the second time scale. A scale factor t(t) allowing synthetic data described in a first time scale to be expressed in a second time scale is defined by a differential equation relating the traveltime variations of a seismic wave to the models resulting from the sequential inversion. Finally, joint inversion wherein a cost function using the scale factor is minimized is carried out so as to evaluate a difference between the synthetic data and the seismic data expressed in another time scale.

Growing community of inventors

Sartrouville, France

Vincent Clochard