Computer implemented method for characterizing a target fluid of a hydrocarbon reservoir under uncertainty

Patent metadata

Field	Value
Publication number	US-11519897-B2
Application number	US-201816497632-A
Country	US
Kind code	B2
Filing date	Mar 27, 2018
Priority date	Mar 29, 2017
Publication date	Dec 6, 2022
Grant date	Dec 6, 2022

Abstract

The present invention is related to computer implemented method for characterizing a target fluid of a hydrocarbon reservoir under uncertainty providing a set of fluid properties from a limited amount of information.

First claim

The invention claimed is:

1. A computer implemented method for assessing the viability of fluid production from a hydrocarbon reservoir using characteristics of a target fluid present within the hydrocarbon reservoir, wherein characterization of the target fluid, at least in part, is determined based on properties of the hydrocarbon reservoir in which the target fluid is located, or on properties of the hydrocarbon reservoir rock that stores the target fluid, and the viability of fluid production is simulated through facilities of the reservoir having a predetermined field architecture including at least one pipe connecting a first point of fluid extraction located within the reservoir and a final end point of fluid extraction located at surface facilities of the reservoir, the method comprising the following steps:
a) establishing m predetermined fluid properties (p 1 , p 2 , . . . , p m ) including at least a first fluid property (p 1 ) directed to one of the phases of the target fluid present in the hydrocarbon reservoir, either liquid or gas, and a second fluid property (p 2 ) directed to a gas/liquid ratio of the target fluid;
b) assigning a probability density function (ƒ 1 ) for the first fluid property (p 1 ) and a probability density function (ƒ 2 ) for the second fluid property (p 2 );
c) providing m−2 correlations c 3 , . . . , c m dependent at least on p 1 and/or p 2 properties such that p i =c i (p 1 , p 2 ); i=3, . . . , m;
d) for each property p i , i=3, . . . , m
generating, automatically by a computer system, r sample values c i ({tilde over (p)} 1 , {tilde over (p)} 2 ) wherein {tilde over (p)} 1 is a random sample according to its probability density function ƒ 1 and {tilde over (p)} 2 is a random sample according to its probability density function ƒ 2 ;
generating, automatically by a computer system, a probability density function ƒ i for property p i from the r sample values c i ({tilde over (p)} 1 , {tilde over (p)} 2 );

e) based on the probability density functions (ƒ 1 , ƒ 2 , . . . , ƒ m ), determining the characterizing properties p i , i=1, . . . , m of the target fluid present in the hydrocarbon reservoir;
f) using the characterizing properties of the target fluid determined in step e), assessing the fluid production viability of the reservoir by simulating the evolution of the target fluid along the piping of the predetermined field architecture.

2. The method according to claim 1 wherein the first fluid property (p 1 ) directed to one of the phases of the target fluid is the American Petroleum Institute (API) gravity and the second fluid property (p 2 ) directed to the gas/liquid ratio of the target fluid is the gas-oil ratio (GOR) ratio.

3. The method according to claim 1 , wherein correlation c i (p 1 , p 2 ) also depends on at least one additional fluid property p j , being j<i.

4. The method according to claim 3 , wherein the fluid properties p i , i=3, . . . , m are sorted such that dependency of correlations is c 3 =c 3 (p 1 , p 2 ), c 4 =c 4 (p 1 , p 2 , p 3 ), . . . , c i =c i (p 1 , p 2 , . . . , p i-1 ) wherein not necessarily all properties p 3 , . . . , p i-1 appear as dependent parameters.

5. The method according to claim 1 , wherein it further comprises
providing an analog data base, said analog data base executable in a computer system, comprising records of fluids with fluid properties from known reservoirs, wherein at least one record of the analog data base further comprises the first fluid property (p 1 ), the second fluid property (p 2 ) or both;
providing a similarity module for the comparison of fluid properties of the target fluid and fluid properties of records of the analog data base, the module providing a similarity value as a result of said comparison,
wherein before determining a property p i by generating r sample values in step c), i being an integer value in the interval [3,m], the following steps are carried out:
i. determining a threshold parameter (sim);
ii. selecting the set of records of the analog data base, by means of the similarity module, having a similarity value higher than the threshold parameter (sim) when compared to a target fluid,
iii. determining a modified correlation c′ i as c′ i =c i *C+Δc i , being C and Δc i constant values, said constant values being determined by imposing that the sum of the errors between the value provided by the new correlation c′ i and the property value provided by each of the selected records of the analog data base is minimum;
iv. in step c) using the modified correlation c′ i when the r sample values c i ({tilde over (p)} 1 , {tilde over (p)} 2 ) are computed.

6. The method according to claim 5 , wherein the properties compared between the target fluid and the records of the analog data base comprises geochemical properties, thermodynamic properties or both, said properties being quantitative, multiple binary or binary variables or any combination thereof.

7. The method according to claim 5 , wherein the target fluid and at least one record of the analogous data base comprise at least one binary property wherein
each type of the binary properties is set to 0 or 1 and comprises a weighing value and
the similarity module comprises a binary property contribution value, said contribution value being weighed with the weighing value, said weighing value determined according to the following rules for each binary property:
if the target fluid does not have the binary property and/or the record does not have the binary property, then the contribution value to the similarity value is set to 0 and the weighing value of the binary property is not taken into account;
if the target fluid and the record have the binary property but the target fluid value is different to the record value the contribution value to the similarity value is set to 0 so it does not account the weighing value of the property;
if the target fluid and the record have the binary property and the target fluid value is equal to the record value the weighing value is set to 1 so the full weighing value of the property is the contribution value to the similarity value.

8. The method according to claim 5 , wherein the target fluid and at least one record of the analogous data base comprise at least one multiple binary property wherein
each type of the multiple binary properties is set to 0 or 1 and comprises a weighing value and
the similarity module comprises a multiple binary property contribution value, said contribution value being weighed with the weighing value, said weighing value determined according to the following rules for each multiple binary property:
if the target fluid does not have the multiple binary property and/or the record does not have the multiple binary property, then the contribution value to the similarity value is set to 0 and the weighing value of the multiple binary property is not taken into account;
if the target fluid and the record have not in common any of the multiple integer combination of values, the contribution value to the similarity value is set to 0 and the weighing value of the property is not taken into account;
if the target fluid and the record have in common at least one of the multiple integer combination of values, the weighing value is set to 1 so the full weighing value of the property is the contribution value to the similarity value.

9. The method according to claim 5 , wherein the target fluid and at least one record of the analogous data base comprise at least one quantitative property wherein
each type of the quantitative properties has a continuous value and comprises a weighing value and,
the similarity module comprises a quantitative property contribution value, said cont

Assignees

• Repsol Sa

Inventors

• Peña Díez José Luís
• Merino García Daniel
• Benito Iglesias Verónica
• Sanz Sanz Paula

Classifications

• E21B47/06

  Measuring temperature or pressure · CPC title

• E21B49/0875

  determining specific fluid parameters · CPC title

• G06F17/15

  Correlation function computation {including computation of convolution operations (arithmetic circuits for sum of products per se, e.g. multiply-accumulators G06F7/5443; digital filters, e.g. FIR, IIR, adaptive filters H03H17/00)} · CPC title

• E21B49/081

  with down-hole means for trapping a fluid sample (E21B49/10 takes precedence) · CPC title

• E21B47/07

  Temperature · CPC title