Rock Typing Mapping Methodology Based on Indexed and Probabilistic Self-Organized Map in Shushufindi Field

To make efficient use of image-based rock physics workflow, it is necessary to optimize different criteria, among which: quantity, representativeness, size and resolution. Advances in artificial intelligence give insights of databases potential. Deep learning methods not only enable to classify rock images, but could also help to estimate their petrophysical properties. In this study we prepare a set of thousands highresolution 3D images captured in a set of four reservoir rock samples as a base for learning and training. The Voxilon software computes numerical petrophysical analysis. We identify different descriptors directly from 3D images used as inputs. We use convolutional neural network modelling with supervised training using TensorFlow framework. Using approximately fifteen thousand 2D images to drive the classification network, the test on thousand unseen images shows any error of rock type misclassification. The porosity trend provides good fit between digital benchmark datasets and machine learning tests. In a few minutes, database screening classifies carbonates and sandstones images and associates the porosity values and distribution. This work aims at conveying the potential of deep learning method in reservoir characterization to petroleum research, to illustrate how a smart image-based rock physics database at industrial scale can swiftly give access to rock properties.

show abstract

Producing the Limit, by Re-Defining the Rock Typing Classification in a Mature Oilfield: Case Study in the Auca Field, Oriente Basin, Ecuador

Morales

Lafournère

Bringer

et al. 2017

Day 2 Thu, March 16, 2017

View full text Add to dashboard Cite

The Auca field is located in the northern Oriente basin (Ecuador) with hydrocarbon production coming from Cretaceous fluvio-estuarine and shallow marine sandstones. The field has produced more than 547 million barrels of oil since 1972 and by the end of 2015 the field recovery factor was approximately 14%. In December 2015, the reservoir management and the field re-development activities for the Auca field were awarded to Schlumberger Production Management (SPM) under the name of Shaya project. Since then, to sustain the field re-development activities, an integrated reservoir characterization process has been implemented. In this depositional environment reservoir evaluation can be very challenging, especially when using only conventional well logs. It is proposed in this paper that the acquisition of texture dependent measurements is the solution to improve the understanding of the reservoir rocks in highly heterogeneous environments. Based on our experience in Ecuador, incorporating nuclear magnetic resonance (NMR) in the petrophysical model appears to be the best way to collect the needed texture dependent data. The Rock type characterization in the field was based on mercury injection capillary pressure data. This method enables the determination of pore throat profiles for each rock type and the dominant interconnected pore system, which corresponds to a mercury saturation of 35% in a capillary pressure curve. An empirical relationship was used to relate conventional porosity and permeability to pore throat profiles, and this was used to classify rock types. With the purpose of validating reserves and optimizing the field development plan, a model based on rock type characterization was developed using existing core, log and production data. Additionally, this model was calibrated using data from multiple fields in the basin. The propagation of the model from core to logs was accomplished through a relationship between gamma ray, density, neutron and NMR logs with core porosity and permeability in key wells. These relationships are dependent on rock type, and they were used to extrapolate core characterization to those wells without cores. Maps of rock type distribution were used to classify areas according to their petrophysical properties. These maps were also used to delineate the reservoir limits, helping to validate and identify prospective areas for future drilling and workovers. This paper presents the characterization of the reservoir into rock types by integrating geological, petrophysical and production data through Neural Network Analysis, establishing a fundamental input into and support for the development of the exploitation plan.

show abstract

Rock Typing Mapping Methodology Based on Indexed and Probabilistic Self-Organized Map in Shushufindi Field

Cited by 5 publications

References 2 publications

An integrated clustering algorithm based on firefly algorithm and self-organized neural network

An integrated clustering algorithm based on firefly algorithm and self-organized neural network

Digital core repository coupled with machine learning as a tool to classify and assess petrophysical rock properties

Producing the Limit, by Re-Defining the Rock Typing Classification in a Mature Oilfield: Case Study in the Auca Field, Oriente Basin, Ecuador

Contact Info

Product

Resources

About