Method for forming a model of a geologic formation, constrained by dynamic and static data

• US 6,381,543 B1
• Filed: 08/08/2000
• Issued: 04/30/2002
• Est. Priority Date: 09/02/1999
• Status: Expired due to Term

• Current Assignee: Institut Francais Du Petrole
• Original Assignee: Institut Francais Du Petrole
• Inventors: Guerillot, Dominique, Pianelo, Laurent
• Primary Examiner(s): McElheny, Jr., Donald E.
• Application Number:
  US09/634,583
• Time in Patent Office: 630 Days
• Field of Search: 702/11, 702/12, 702/13, 702/14, 702/16, 367/73, 703/10
• US Class Current: 702/13
• CPC Class Codes: G01V 1/282 Application of seismic models, synthetic seismograms
• CPC Class Codes: G01V 1/306 for determining physical properties of the subsurface, e.g. impedance, porosity or attenuation profiles

Abstract

Method for forming, by means of an inversion technique, a model of an underground zone, constrained by static data: data obtained by seismic exploration or measured in situ (logs), and by dynamic data: production measurement, well testing, etc. From an a priori selected meshed geologic model and from relations between parameters or physical quantities characteristic of the medium: acoustic impedance and cabsolute permeability for example, a simultaneous inversion of the two parameters is performed by minimizing a global cost function, which has the effect of considerably decreasing the number of possible solutions and of improving characterization of the underground zone. The method can be used notably for modelling hydrocarbon reservoirs, of zones likely to be used as gas, waste storage places, etc.

Claims

1. A method for automatically forming, by means of an inversion technique, a representation of variations, in an underground zone producing fluids, of physical quantities, constrained both by fluid production data and exploration data, these data being obtained by measurements or observations, characterized in that it comprises:

selection of an a priori geologic structure of the underground zone with distribution of at least a first physical quantity (k) estimated from exploration data and of at least a second physical quantity (σ

) estimated from production data,

implementation of two initial direct models respectively depending on the two physical quantities, and

optimization of at least one model implemented by iterative application of an inversion of at least one physical quantity, said quantities being related to one another by a combination relation, so as to obtain a representation of the structure of the zone in connection with the physical quantities considered, best meeting the exploration and production data combination.

2. A method as claimed in claim 1, characterized in that it comprises iterative optimization of the model implemented by performing simultaneous inversion of first physical quantity (k) and of second physical quantity (σ

) with optimization of a global cost function J(k,σ

) and depending on these two quantities.

3. A method as claimed in claim 2, characterized in that a cost function J(k,σ

) is optimized, which is the sum of terms measuring the differences between the predictions of each direct model and the data obtained by measurement or observation of the corresponding quantity and of difference terms formed by taking account of said combination relation.

4. A method as claimed in claim 1 characterized in that each model implemented is optimized by minimizing the objective function through determination of the exact value of the gradient thereof, using the adjoint state method.

5. A method as claimed in claim 1 characterized in that the production data are measurements obtained in wells through the underground zone and the exploration data are seismic signals picked up by seismic receivers in response to waves transmitted in the formation, the first physical quantity is the permeability of the medium and the second physical quantity is the impedance of the medium in relation to these waves.