Логин:
Пароль:
Регистрация
Забыли свой пароль?

Development of methods for automated stratigraphic correlation by well log data

UDK: 550.8.072
DOI: 10.24887/0028-2448-2018-7-74-76
Key words: linking, correlation, stratigraphy, automation, results, rock
Authors: O.V. Nadezhdin (BashNIPIneft LLC, RF, Ufa), N.A. Evgrafov (BashNIPIneft LLC, RF, Ufa), A.V. Markov (BashNIPIneft LLC, RF, Ufa), D.V. Efimov (BashNIPIneft LLC, RF, Ufa)

Stratigraphic correlation, determining layers in separate strata that are exactly the same age are an important part of geological modelling. Stratigraphic correlation is carried out in order to establish the boundaries of geological layers and to identify and classify them in different wells. The results of stratigraphic correlation have a significant impact on evaluating the spatial distribution of rock properties and estimating reserves in layered formations, creating petro-physical models of such formations and solving some field development problems. Stratigraphic correlation is usually done using manual labor of skilled geology specialists; however, it may become quite labor-intensive when hundreds of wells need to be processed. For some large oilfields consisting of thousands of wells a topic of automated multi-well stratigraphic correlation becomes relevant.

The article further develops methods for automated multi-well stratigraphic correlation, including the analysis of lithofacies generated from well logs.

The aim of this project is to increase the efficiency of geology specialists in solving stratigraphic correlation problems. The task was to carry out stratigraphic correlation of hundreds of wells from multiple oilfields using the automated approach and to compare results with the expert evaluations.

The automated stratigraphic correlation approach includes preparing and analyzing a selection of wells with expert evaluations of stratigraphic intervals of interest (usually it is about 10-15% of wells used for correlation). Next step is choosing a base well and a reference interval for the stratigraphic unit of interest for that well. The reference interval’s position will then have to be found for the rest of the wells. Then we use automated multi-well correlation of the reference interval which contains stratigraphic units of interest. A correlation scheme is created based on cluster analysis methods. The approach allows displaying a background of certain well logs from all the wells used in stratigraphic correlation at once and carrying out interactive correlation.

The approach described is an express method for stratigraphic correlation. The last step may be further elaborated by displaying not only well logs but also core data as well as any other data that can help identify different stratigraphic units.

Automated multi-well correlation results stand well against expert evaluations for the majority of wells with 80% of them having an error of less than 5% when compared with expert marks.

References

1. Kovalevskiy E.V., Gogonenkov G.N., Perepechkin M.V., Geological model update using the automatic correlation of wells (In Russ.), Nedropol'zovanie XXI vek, 2007, no. 4, pp. 28–31.

2. Sharafutdinov T.R., Shaybakov R.A., Testing an algorithm of the well log data autocorrelation on the example of Achimov sequence of Pravdinskoye field (In Russ.), Nauchno-tekhnicheskiy vestnik OAO “NK “Rosneft'”, 2012, no. 1, pp. 18–22.

3. Gutman I.S. et al., Detal'naya korrelyatsiya dlya postroeniya trekhmernykh geologicheskikh modeley zalezhey UV (Detailed correlation for construction of three-dimensional geological models of hydrocarbon deposits), Moscow: Neft' i gaz Publ., 2001, 79 p.

4. Forgy E.W., Cluster analysis of multivariate data: efficiency versus interpretability of classifications, Biometrics, 1965, no. 21, pp. 768–769.В  В В 



Stratigraphic correlation, determining layers in separate strata that are exactly the same age are an important part of geological modelling. Stratigraphic correlation is carried out in order to establish the boundaries of geological layers and to identify and classify them in different wells. The results of stratigraphic correlation have a significant impact on evaluating the spatial distribution of rock properties and estimating reserves in layered formations, creating petro-physical models of such formations and solving some field development problems. Stratigraphic correlation is usually done using manual labor of skilled geology specialists; however, it may become quite labor-intensive when hundreds of wells need to be processed. For some large oilfields consisting of thousands of wells a topic of automated multi-well stratigraphic correlation becomes relevant.

The article further develops methods for automated multi-well stratigraphic correlation, including the analysis of lithofacies generated from well logs.

The aim of this project is to increase the efficiency of geology specialists in solving stratigraphic correlation problems. The task was to carry out stratigraphic correlation of hundreds of wells from multiple oilfields using the automated approach and to compare results with the expert evaluations.

The automated stratigraphic correlation approach includes preparing and analyzing a selection of wells with expert evaluations of stratigraphic intervals of interest (usually it is about 10-15% of wells used for correlation). Next step is choosing a base well and a reference interval for the stratigraphic unit of interest for that well. The reference interval’s position will then have to be found for the rest of the wells. Then we use automated multi-well correlation of the reference interval which contains stratigraphic units of interest. A correlation scheme is created based on cluster analysis methods. The approach allows displaying a background of certain well logs from all the wells used in stratigraphic correlation at once and carrying out interactive correlation.

The approach described is an express method for stratigraphic correlation. The last step may be further elaborated by displaying not only well logs but also core data as well as any other data that can help identify different stratigraphic units.

Automated multi-well correlation results stand well against expert evaluations for the majority of wells with 80% of them having an error of less than 5% when compared with expert marks.

References

1. Kovalevskiy E.V., Gogonenkov G.N., Perepechkin M.V., Geological model update using the automatic correlation of wells (In Russ.), Nedropol'zovanie XXI vek, 2007, no. 4, pp. 28–31.

2. Sharafutdinov T.R., Shaybakov R.A., Testing an algorithm of the well log data autocorrelation on the example of Achimov sequence of Pravdinskoye field (In Russ.), Nauchno-tekhnicheskiy vestnik OAO “NK “Rosneft'”, 2012, no. 1, pp. 18–22.

3. Gutman I.S. et al., Detal'naya korrelyatsiya dlya postroeniya trekhmernykh geologicheskikh modeley zalezhey UV (Detailed correlation for construction of three-dimensional geological models of hydrocarbon deposits), Moscow: Neft' i gaz Publ., 2001, 79 p.

4. Forgy E.W., Cluster analysis of multivariate data: efficiency versus interpretability of classifications, Biometrics, 1965, no. 21, pp. 768–769.В  В В 




Attention!
To buy the complete text of article (a format - PDF) or to read the material which is in open access only the authorized visitors of the website can. .

Mobile applications

Read our magazine on mobile devices

Press Releases

16.07.2019
08.07.2019
04.07.2019
SPE 2019
ТАТАРСТАНСКИЙ НЕФТЕГАЗОХИМИЧЕСКИЙ ФОРУМ